Speak Softly but Pack a Mean Elbow, w/ Jen Stirrup

When Jen Stirrup speaks, she speaks softly.  The meaning of her words, however, speak loudly!  Jen is CEO of Data Relish, a UK-based consultancy that delivers real business value through solving all manner of business challenges.  You don't earn the nickname the Data Whisperer without knowing a great deal about Business Intelligence and AI.  Jen certainly knows not only those topics, she knows SO much more! References in this episode: Data Kind The Art Of War Blade Runner Tears Scene Episode Timeline: 4:30 - The human element of data, Bias in data, implications of Artificial Intelligence and Machine Learning, and COVID data 27:00 - The BI goal is Business Improvement, escalation and taking principled stands, Data-Driven vs Data Inspired 46:00 - Seeing the hidden costs of some business strategies, the value of even small successes, Diversity and Inclusion, and online bullying 1:29:30 - Jen's mugging story (!) Episode Transcript: Rob Collie (00:00:00): Hello friends. Today's guest is Jen Stirrup. Jen and I have had one of those long-running internet friendships that are so common these days, especially in the data world and in certain communities. But we've also had the opportunity to meet in person several times at those things that we used to do called "in-person physical conferences." She's an incredibly well-seasoned veteran of the data world, but if you're expecting us to be talking about things like star schema and DAX Optimization, that's not really what we talked about. You know that our tagline here is "data with the human element," and we definitely leaned into that human element in today's show. Now, we do talk about some of the important human dynamics about data projects. For example, how the business intelligence industry kind of lost its way in the past and forgot that it's all about improvement and how we're as an industry waking back up to that today. Rob Collie (00:00:54): We also talked about the value of having even one signature success in a large organization that other people can look at to become inspired. And she has some very interesting and well-founded semantic opinions about terms like "data-driven" and why maybe, "data-inspired" is better. Similarly, she prefers the term "data fluent" to "data literate", and she explains why. But we also touched repeatedly on the themes of ethics and inclusivity in the world of data. Now, I have a personal idea that I haven't really shared on this show before that I call "radical moderation." It's the idea that you can be polite, you can be reasonable, while at the same time advocating for sharp change. Now, this is personally what I would like to see emerge in our political sphere, for instance, a form of polite radicalism. We need to change, but we need to be nice. Rob Collie (00:01:52): There aren't many readily available examples that I could point to if I wanted to show you "this is what radical moderation looks like." But now if someone asked me for that, I can point them to this conversation we have with Jen. She is soft-spoken, she is polite, she is open-minded, including the open-mindedness that she might not always be correct. And yet, underneath all of that, is a very firm conviction that we need to be better. And I think that's the best introduction I can give this because I don't want to spoil anything upfront. So, let's get into it. Announcer (00:02:28): Ladies and gentleman, may I have your attention please? Announcer (00:02:32): This is the Raw Data By P3 Adaptive Podcast, with your host, Rob Collie, and your co-host, Thomas you know. Find out what the experts at P3 Adaptive can do for your business. Just go to p3adaptive.com. Raw Data By P3 Adaptive is data...with the human element. Rob Collie (00:02:56): Welcome to the show, Jen Stirrup. It is such a pleasure to see you again, virtually, talk to you. I'm really glad we were able to do this So, thrilled to have you here. Jen Stirrup (00:03:06): Thank you so much for having me. I'm glad we made it work in the end. Diaries, schedules, everything else, but I'm really glad to be here and it's great to speak to you. Rob Collie (00:03:15): I know bits and pieces of the Jen Stirrup story and I know bits and pieces of what you're up to. How do you describe yourself on your LinkedIn profile? Jen Stirrup (00:03:23): So I would describe myself as really trying to help people make their data better. I've just finished a post- COVID data strategy for a healthcare organization in the US and in the UK. The reason I'm doing that is to try and have a big impact. I believe in that, I think COVID has brought around a real stress and a lot of technical architectures, and a lot of data architectures as well, and there're all sorts of pressures. So I've just finished that, which has been a nice piece of work. I've been working with a religious organization on their data as well. A lot of people are accessing their services as part of a recovery from COVID. I think it's been a very difficult, challenging time for a lot of people in terms of mental health, and I like to think that by solving these problems you're actually helping people, in a way to contact, some of whom you may never meet, but that's okay. That's really what I like to do, I think, it's a way of connecting, I think. Rob Collie (00:04:22): We subtitled the show 'Data With The Human Element,' you think of the data field is like this cold, analytical, sanitary, and it's not, right? If you're doing it right, you're having an impact in the human plane, and it's a leveraged impact because you can really sort of touch a lot of people's lives via the central hub that is data. And you've got to keep the human beings in mind, even to be successful at the quote-on-quote "cold, calculating data stuff." If you don't keep the humans sort of first and foremost in your mind, you're not going to design, for example, a good data strategy, like what you just finished. Jen Stirrup (00:05:02): That's right. So I believe that the information ladder is quite important. So we start off with data, then we need to turn that into information, but then we need to turn it into knowledge and then wisdom. And I think COVID has taught us many things. I think it's maybe taught us a sense of purpose, it's something that can help drive all of us. Data can be part of that and I think that data in some ways has been replacing some of the bigger-purpose questions that perhaps we should ask ourselves more often as human beings. With artificial intelligence, particularly, I'm finding that people are replacing data with, perhaps, information, knowledge, or wisdom and say "what does the data see?" and that's fine, but we have to have the context to the data as well. Jen Stirrup (00:05:47): I think in some ways with artificial intelligence, what people are trying to do is build a little box of data and it's becoming this oracle that people are going to touch and say: "So, what does the data say?" It's like we are taking this box and we're trying to turn into some sort of God that we can touch, and it's going to give us all the answers, but if we're going to do that, it has to be a God that we are comfortable to live with, and it's one that we can choose, and one that fits in with people's ethics and their sense of purpose. So, I see data as part of fitting something that can make us all better in so many different ways, whether that is healing or bringing people together. Jen Stirrup (00:06:29): So I think if we could solve these problems where people are feeling that they are not interconnected, then we could start to try and look at that and perhaps think about making people feel whole and feel more together. Because I think what COVID has done is really helped us to focus a lot on data but perhaps not about how we could do things better. It seems that we have an opportunity to decide what goes back in to make the new normal or the next normal. And I'm worried I suppose that I don't see that happening as much as I would like. So yeah, data is important. Absolutely. We wouldn't be here without it and the fact people are struggling with it does pay my mortgage. I still would like us to ask ourselves the bigger questions as well as something that's important to me. Rob Collie (00:07:14): Let me check here. Oh yeah yup, it pays my mortgage as well. We're here for a reason that's for sure. I loved you talking about the AI, this box, that we're going to sort of elevate to the status of a God or that's how a lot of people are viewing it subconsciously. Of course, it's a box that we built. Jen Stirrup (00:07:33): Yeah. Rob Collie (00:07:33): We fed it with our context. It got fed with our assumptions and also our blind spots and now if it makes decisions, that thing starts making judgments and decisions that impact people's lives. It's a tricky proposition, it's one that's best approached very carefully. Jen Stirrup (00:07:55): I agree and I think that's why the bigger questions are important. So say for example, you may have seen the Netflix information series. It was called 'The Social Hack' or something like that. I've forgotten the name, but it was talking about the role of bias in data. One of the researchers found that their facial recognition algorithm didn't recognize a face. And the reason for that was that she's black and for me, I just thought, that's such a preventable issue and how much time do you spend looking at preventable issues? And perhaps not very much. I still see the magpie problem a lot in technology. Companies are happier buying a new technology that they see that's going to solve all their problems, but actually it's not doing that. It's maybe replacing as a bad answer to a different question. We can't see that right now in artificial intelligence. Jen Stirrup (00:08:48): There's some research going on, which will decrease the size of data sets that AI needs in order to create its algorithms and that sounds fine. It's a good piece of research, but what I'd like to see is more researches on collating datasets which are less biased, so that we can think about focusing and trying to make the algorithms fewer rather than focusing on making them smaller. Jen Stirrup (00:09:13): I know a few years ago, you probably remember, everyone talked about big data. Big data was the thing but we didn't ask ourselves if this was the right data. It might be big, but if it's missing out large sections of the population, then that's building an inequality before we get started. I think, even if you don't have the answers, asking these questions is a good thing. I don't have all the answers. There's people working in this field much much smarter than me and they all live and breathe this stuff and I read it, the things that they're doing and talking about, and I think this is such an important part of what we do every day. I think it's really important. I don't know what you think, but there's so much going on in the world of data at the moment that it feels hard to keep up sometimes. Thomas Larock (00:09:58): So first I want you both to remember in case you've forgotten, but you can purchase the Azure Data Box, that does exist. Rob Collie (00:10:07): We will just call it God in a box. Thomas Larock (00:10:09): Azure Data Box, it's actually for shipping storage to an Azure data center, but that's what they chose to call it and I said: "You put your data in the box or it gets the hose again." Right? So- Rob Collie (00:10:20): No no, Tom, it's one: "Put your data in the box." Thomas Larock (00:10:26): So, I mean, that does exist. The first point I wanted to make that you danced around, like Rob you were talking about how we're building this thing and it comes with all of our failings. And I know Jen, she leads discussions on diversity, inclusion, equality and I try to emphasize why that's so much more important and especially seeing the rise and I saw the Netflix special as well, and the Data Justice League. The idea is we need to have those programs in order to have better models. We have to be aware of the bias inherent in the stuff that has already been built. And I think there's a lot more awareness over the last 18 months regarding the products that are on the market that are already failing us because they were built with these biases. And that's a difficult thing to overcome now that you have police departments or governments deploying this technology, thinking, as Jen said, it's this God that is just going to give you all the answers. Thomas Larock (00:11:35): Jen, you also hinted on the thing about the question. So, you're replacing one problem with another, and that made me think of how vital it is that you understand the question you need answered and a lot of times that gets kind of shifted, it's fluid almost. It's like: "Oh, well we were doing this thing we think this next thing we'll solve for it." But the next thing you're getting is actually answering a completely different question than what you thought you were doing and it leads to a huge, huge disconnect. And I think the last thing I would say Jen, I've seen that research about the data sets. I'm encouraged by the idea that we could get people to understand that it's not the volume of data that makes a better model. It's the data that was chosen to be collected in the manner in which it's collected. Thomas Larock (00:12:30): So I know the research on building these models and they're saying: "Yeah, you don't need a billion rows. The accuracy tails off at some point after, say, a million rows." At some point more data doesn't make this model any more accurate but the inherent problem is how was it collected? What were the biases and how was it collected? What was missing? Was it missing at random? Was it missing not at random? The analysis necessary to conduct that research, I think is where we are sorely lacking in business. I know it exists in academia, but those people, they don't scale. There's only so many of those, and there's a lot more businesses trying to get the job done so I think that's fairly important. Jen Stirrup (00:13:13): There is a huge gap between academia and business. I guess there always has been, I do speak to academic institutions from time to time and it's clear that they are doing so much work. They really are, but how that is getting out? I am not sure. Maybe that's why they asked me to come and talk to them so I can talk to other people about what they're doing and I don't mind doing that. I think there needs to be more of that, because I think these scientists, these academics are working in this, have to get access to each other as well and the multidisciplinary aspect of it is really interesting. I did a Postgraduate in Cognitive Science about 20 years ago, and suddenly it's back round again, and it's about philosophy, linguistics, psychology, AI. And why did that go away? Jen Stirrup (00:14:03): It should never have really gone away. I think we got as an industry perhaps Goldstone and such technologies which these things were re-badged as, and we got derailed by the marketing efforts. But I think that there's real room for doing these things in a better way. I don't know if you see this, but I see, or maybe it's my age now, I've been around in the industry for a long time, but I see that people are doing and making mistakes that I first saw 20 years ago, data collection, which you rarely mentioned, Tom, that's been there for a long time and then it seemed to go away. Jen Stirrup (00:14:36): I think that's why academia does help because it gives us maybe more of that consistent backgrounds than perhaps we get from marketing noise, which was goes round in cycles and trends as people are under pressure to purchase these licenses or whatever it happens to be. I wish I had better answers for all of this, I think sometimes it's about just asking these questions, blogging, talking about them, putting them on social media so that when people are thinking, "what do I do about data strategy?" That these things are part of this. I saw a study recently saying that companies are decreasingly likely to include ethics and these questions and bigger societal questions as part of the data strategies as you're trying to get the link. But it disheartens me because I thought I could see that the voices are getting squeezed out. Rob Collie (00:15:25): Decreasingly likely, like we're trending- Jen Stirrup (00:15:28): Trending down. Rob Collie (00:15:28): You know, it'd be one thing to be flat, right? I mean that would also be disheartening, but to be decreasing, decreasingly likely to be factoring in ethics into a data strategy. Now we've been talking a lot and I think it's a good thing to continue to talk about the implications of AI and machine learning in this space, the business intelligence industry isn't particularly fraught with this kind of problem, right. Transactions happened, or they didn't, you know, and it was the number of six or a seven. I mean like, you can get it wrong, you can have bugs, right. But there isn't any like objective debate about what, there shouldn't be any way about what actually has happened. But the decider systems, are a completely different game, like where should we route this patient? This is going to have a huge impact on their life. Rob Collie (00:16:21): That's a very, very, very different game and we've been talking about sort of like, the completeness of the data that is used to train these systems, but I think it's really instructive just to stop for a moment and go, you know what, even if we were able to feed these systems a 100% comprehensive picture of today's world, we still have to accept the fact that we're telling it that today's world is what we want. Right. And maybe we don't, you know and there's always a judgment in training these systems, we tell it what is a success and what isn't a success. Our unintentional biases can leak into this stuff in a million different places, even if you suddenly had God-like comprehensive powers to feed it, quote-on-quote, all the data, right. It's still leaky. It's still fraught. Jen Stirrup (00:17:13): Yeah and actually, I think it's an extension of their problem that we see just when we're building a data warehouse. Sometimes I'll go into a customer and they'll say, "you know, we want to see our data and see our latest vendor here," and then I'll say, "well, is it preserving the data or is it just, you know, been reamed out the other end, what you're doing with it? Where you're storing it?" And then the argument against the data warehouse as well. It's not going to capture everything in the possible universe of possibilities in my business, so I don't want to do it. And I find the argument goes something like, "there's an edge case that it won't cover." Others, "this edge case, it won't cover here." And then you have to say, "well, you know okay. So it's not going to cover all the possible edge cases, but it will cover 80% of what you need, and the rest, can go to shadow IT or shadow data systems or wherever they happen to be." Jen Stirrup (00:18:03): And I think we're still trying as it's a bigger picture perhaps trying to control everything that happens around our business, but we have to be flexible enough to cater for these scenarios. We haven't seen this before. I think that's what makes the AI so difficult actually, as we have more than one type of AI, we have a general artificial intelligence, which is more like Terminator, you know, these kinds of things. Rob Collie (00:18:29): Innocuous stuff like that. Thomas Larock (00:18:30): Harmless. What's the worst that could happen. Rob Collie (00:18:32): Yeah. I mean. Jen Stirrup (00:18:35): Well, I think as humans, we do enough damage to ourselves, most of the time we don't need a Skynet. Thomas Larock (00:18:38): That's true. I agree. That's often my reaction to, well you know, like self-driving cars, like what if it makes this mistake? Okay yeah but the human being track record behind the wheel, we're not trying to be perfect, we're just trying to be better than people, which is a little bit more achievable perhaps. Jen Stirrup (00:18:56): Exactly and it's all a bit context, which is how to program. You probably remember a few years ago, at SQLBits say Tom, Steve Wozniak visited. I don't know if you were there for that SQLBits but Steve Wozniak is one of the team that founded apple. You must know who he is, but he's talked to us about the Wozniak test for AI, the testers will have an artificial intelligence sought of robot come into your house and make you a coffee from scratch. Now that involves a lot of contextual knowledge. They have to find your kitchen, they have to get your ingredients and get a cup, you know all that kind of thing and that requires context. And that's more general AI, that's more difficult to program. But if we're to think with CEI being more successful for businesses automation productivity, and it's just trying to do something, one thing really, really well, something that will help a human to make better decisions faster. Jen Stirrup (00:19:51): Such as perhaps parceling out x-rays, which don't show any presence of a tumor as an example, but we then get the 10% of x-rays that makes sure something and passing those onto a human to look at. So there's plenty of rooms for defining what success looks like for us for artificial intelligence I think. With business intelligence, your right, we should have one version of the truth. People are still living so much in Excel and Google sheets and things of empires away, and that are sitting in their laptop. How do you move that to the cloud? So you move them perhaps to office 365 or a Google work space, and then you're trying to encourage people to rethink the processes about, Hey why do we save stuff in the cloud? Or why do we make our decision making more apparent? And it seems a bit difficult to ask AI to make its decision-making more apparent, when actually a lot of people spend time hiding or umpiring the knowledge anyway. Jen Stirrup (00:20:49): I don't know if you think this, but I often think business intelligence problems are change management problems in disguise. It just happens to be showing up in the data that there's a problem. Thomas Larock (00:20:59): Yeah. Rob Collie (00:20:59): Ultimately it's not about knowing, it's about improving. Knowing that there's a problem and even knowing what's causing it is really just the beginning. Very often it's like okay, now what? This is going to be a really difficult problem to address operationally. Jen Stirrup (00:21:16): I think we forget the process of optimization and business intelligence. And I wonder if that's the reason why AI is becoming so prevalent at the moment, because it is much more clearly talking about optimizing and improving processes and automating. I think in business intelligence, we have almost stopped talking about optimizing business processes. I don't see it quite as much, I wonder if we get sort of caught up in data visualization, you know Tableau came along and then power BI and everyone started chasing after that. We're perhaps forgetting that actually we're doing all that for a purpose, which is to make something better somewhere. I don't know if you find this but, I obviously run [inaudible 00:21:54] business and it's very hard to get customers to agree to a case study because they don't want to show that actually they were in a bad place and they don't want to show the competitors that they were in a bad place. Everyone's ashamed of the data. So it's really tough. Rob Collie (00:22:07): I've seen sort of multiple facets of that. So first of all, yes, everyone thinks that they are uniquely broken, everyone's organization that they feel a level of sort of like discomfort and shame about where they're at today or where they were yesterday. They feel like they're the only ones, but we see so many organizations per year, especially the kinds of projects and the pace at which we move the world is very much uniformly broken. No one's really behind, everyone's way behind of where you'd sort of like as a dispassionate observer, you'd expect people to be a lot further ahead than they are, but no, no, really the basics are still not sorted out universally. We're still kind of in a dark age, in a way. Jen Stirrup (00:22:51): Yeah. Something, I see really basic issues of one customer example of talking about where they were calculating the mean incorrectly for two years. And then two years before that, for another two years, they were calculating the median incorrectly in Excel. What they were doing was it were taking the middle value of a column. So of course, if you sorted the column next to it, the value changed. And they said that that was the median. And I said, "okay, so you've got a column of 20 items. Are you telling me that whatever's a number 10 is the mean?" And they said, "well, yes, that's in column B." What happens if you change the order in column E from perhaps alphabetical order to reverse alphabet order, the values can be changed, right? And they looked at me and I said, "why did you calculate it like that?" Jen Stirrup (00:23:41): And they said, cause we can calculate the mean using Excel formula. So eventually I said, "why are you using the mean," because it's quite sensitive to outliers the median's better. and then they said, "well we've tried that but we couldn't calculate the median either." I said, so okay "for four years you've been trying to calculate the mean and the median incorrectly in this one spreadsheet. Can you tell me about the rest of your spreadsheets? How often are you trying to use the median or the mean all of it incorrectly?" And I think it's probably the only time in my 20 plus year career, I've seen a customer actually punch himself in the face and it was just absolutely stunning. And he said, "I'll go and speak to the statisticians." And I thought, you've got statisticians working here. I'd love to meet them. Jen Stirrup (00:24:26): I wonder what they're telling you. And that was my second deal in sight, I was on the on and off for six months. And that was just the first problem I found. So I know we talked about data literacy. I'm not a fan of that phrase. I prefer fluency or something along those lines. So I don't want to assume people are data illiterate. Because I don't think that they are, I think we're born naturally within us an innate sense of numbers in a way, we can tell more from less, right? My dog can do it, right. So if I got five treats in my hand, he knows I've got others. If I just give them one, he's not stupid, he has a sense of quantity. And I think it's about, we need to get better in industry, perhaps explaining results, findings, conclusions, and context to people instead of just throwing dashboards at people and expecting them to understand it. Jen Stirrup (00:25:16): If somebody recently sent me a scientific article which was all about COVID and some testing that they did in mice, and I could read it, but I couldn't understand it because I don't have a background in medicine. I read the abstracts and I read the last paragraph and the first paragraph, but I didn't read the rest of it because I thought this is way beyond me. I don't understand what they're trying to say. But I think for me that highlighted a problem with data literacy, I could read it, I couldn't understand it, and I certainly couldn't act on it. And I don't want to give other people who are trying to consume business intelligence products in some way, whether they're dashboards or even dumps from Excel, that they just don't understand what they're getting. How we do that, I think is perhaps focusing in data translation. Jen Stirrup (00:26:03): How we do that, I think, is perhaps focusing in data translation. I had a woman who worked for me, she actually was a qualified librarian. So, her insights about information retrieval were very interesting. I learned a lot from her, because that was a little bit the data. And she would say things like, "Jennifer, Google is not the only search facility in the world. We can use so much more," because she's accessed all their library systems around the world. And there's so much information we don't access because we can't, usually. But the point being that what I learned from her was about translating things, where they were easier to understand for other people. And I think it's an incredibly valuable lesson, and the world needs more librarians. Rob Collie (00:26:43): There's a lot here, right? Business intelligence was always a means to an end, but because it was so difficult, it was just so incredibly difficult to even get a halfway-competent system instilled, built, configured. When something is that hard for that long, it becomes its own goal after a while. It's easy to habituate to the idea that this is the goal, intelligence is the goal, knowledge is the goal. No, no, no. Improvement was always the goal. What's really been fascinating for us is, when we see our clients, the people we work with, when we see them start to get the BI problem under control for the first time ever, their gaze immediately sort of zooms back and they start thinking completely unbidden by us. We don't have to seed this conversation. It just happens. They start looking at the bigger picture now and going, "Oh, okay. So, now this information needs to feed into better decision loops and optimization and things like that. And how do we facilitate that?" Rob Collie (00:27:53): And from the beginning, we try to counsel everything being built around that "taking action" thing. You can build an incredibly informative dashboard that is intelligent, it's a work of art in many ways, on many levels, and it can be useless. It can be factual, it can be impressive, and it can be useless because you can't use it to make any improved decisions. I've been guilty of this. I have built things like this, like, "Ta-da." And the client doesn't even have the language to push back. Jen Stirrup (00:28:30): It's something I've tried to keep in mind now is the utility of what I'm actually doing, because people just want data for the sake of data, and they get that. I think, sometimes, they don't know what to ask for, so they take something because it's better than nothing. And they'll say things like, "Right, I want the last five years of data and 191 columns, I want it all on the same page, and I want to be able to print it." And then you have to say, "Well, let's think about how feasible that is. You'll get five years of data, it's not going to fit in one page. 191 columns is going to be really small. So, let's have a..." People ask that because they don't know what they want. Jen Stirrup (00:29:06): About a dashboard recently, a health and safety dashboard, it was using power apps as well. So, the company, if they saw a health and safety priority issue, they could use the app, if they were health and safety professionals, and the app would record data, you could upload a photograph, and then that would go into a system which you could then see in Power BI. And the nice thing about that was you could see improvements over time because people could get their health and safety issues resolved more quickly, so things like boxes stacked against fire exits, slip and trip hazards. Jen Stirrup (00:29:43): Now, it may not seem very interesting, but actually, the reason that project had happened was because someone that had been in a health and safety incident and it had not been tracked properly, and the idea being that they were trying to improve the process. But sometimes, I think data problems and data solutions happen because of two things. One is you need an executive sponsor, and the second thing is a crisis. And together, the executive sponsor and the crisis will engender change somewhere. And that change management process so often turns into a business intelligence solution. And nothing is an industry. It's something I'm personally trying to always keep in mind is: what's the purpose? What's the optimization? What problem am I trying to solve? Rob Collie (00:30:30): Yeah, one time, I was asked by a client to help debug a report that was really slow. So, this is great because this is an example of a report that I didn't build, right? I can use an example that wasn't one of my own families, but I'll tell my own as well if you want. But I go, "Okay, I'll take a look at it." I'm expecting some sort of DAX or data modeling problem or something like that. And they show me the report, and it is a 100,000-row pivot table. The pivot table has a 100,000 rows in it. There's DAX behind it. It's a DAX data model behind the scenes, but the report itself, the output is 100,000 rows. And before I even engage, I just turn and look at them and say, "Oh, my God, who was using this? You don't have a performance problem. It's..." And they're very insistent. "No, no, no, no, no. This is the thing. We need this." I'm like, "All right." Rob Collie (00:31:21): So, I start looking at it, and it's crazy how many columns there are. And it was a list of every employee and every location that they have in the country, which was hundreds of locations and thousands of employees. And for each employee, their scheduled time-in and their scheduled time-out, and their actual time clocked in and actual time clocked out. I turned back at him again and I go, "Okay, really? What are we doing here?" And they're like, "Okay. So, we have all these regional managers that are looking at this multiple times a day, probably eight times a day or more, to try to figure out if any of their stores are empty, aren't staffed because people didn't show up." And I just smacked my forehead and I go, "You don't need the timecard report," which is what they called this thing, the timecard report, "You need the empty store detector." Rob Collie (00:32:18): And I mean, there was no way to make this thing faster. I mean, this thing was such a gross misuse of technology. I just went to the whiteboard and I sketched what the empty store detector could look like, and they're like, "Oh, that's great. We'll never get our managers to switch over to using it, so let's just go back to fixing this other piece of junk." Jen Stirrup (00:32:37): Yeah, because something that I struggle with, personally, is the idea of surveyance reports. It's something that really bothers me. I've pushed back on a few customers to see, "Are you micromanaging or are you surveying? What is it you're trying to do?" On occasions, I have escalated it to say, "Look, this report is probably been used to hit people for the head, and I'm not comfortable with this because I think this has gone beyond micromanaging." And we had set the scope of the project of the thing we were supposed to deliver. So, I'm going to escalate this because I want to understand better the purpose. And if I'm wrong, we will deliver it." Jen Stirrup (00:33:12): And normally, when I go back and see that, even in that particular instance, I showed the senior management and I said, "Your middle management want to do this." And they said, "No. We are not spending time doing that. We need to understand the wider context. If there is any issues going on with staffing, then this is probably a symptom rather than the cause of the issues, if people are being watched like that." So, I think some teams escalating, as much as I don't like to do it, sometimes is the best way forward. Rob Collie (00:33:44): It takes a lot of professional courage to do something like that. For example, have you ever taken one of those principled stands and ended up no longer working for that client because they basically fire you for not staying in your lane? That's a risk, right? Jen Stirrup (00:34:01): Yeah. It is. I've never been fired for that, but I have said, "Uncomfortable, and I'm we going to stop delivering services, and we need to decide on an exit strategy." There's different ways you can do that, right? So, you deal with the current project. You then say that you're busy for the next century when we come back to you for other work. I don't like doing that because I often feel like you should give them an alternative to say, "Well, here. I can't deliver it, but I know someone who can." And then I recommend one to my network. But the thing is, when I make these quite principal stands, people back down often, or they back down and they just asked me to do it. But when I've gone back to people like that customer, who come back to me for extra work, I've done some investigating work and I've found that they have not implemented a thing that I've been worried about or concerned about. Jen Stirrup (00:34:49): So, I think, sometimes, if you do speak up, people are maybe surprised by it. It's maybe different who it comes from. And I think, perhaps, even a soft Scottish accent, smiling sweetly at them and saying, "Can you explain to me a bit more about the reasoning behind this? Because your team want to do this thing, but I have some discomfort because it's outside scope." And they're not telling them, and they're very direct. Wait at first, but they start to get their message. Jen Stirrup (00:35:16): A former boss of mine years ago, he said I had a soft rein approach. I actually think that's a nice way of putting it, where, as much as I might be tempted to go in all guns blazing, I'm trying to gently bring it up and then bring it up again a bit more firmly, and then, suddenly, people are starting to understand better. But that's me having to probably, sometimes, exert a huge amount of self-control as well. But I think that's part of the consulting game. It's very tough. But I think seeing something like that happen, I think the reason it happens is because people aren't thinking about it longer-term. And me as a consultant, it's easier, perhaps, for me to think about it long-term and also a bit more closely as well, because you are thinking about the consequences of what you're trying to do, the purpose. Rob Collie (00:36:04): Yeah. If you're good at data and you're experience with it, you spend a lot of time with it, that allows you to put some of those things a little further down in the subconscious, and the rest of your human faculties can resume working, whereas, I think, for people who data is still this arcane thing, it's not the thing that they've spent their lives with, it's just really easy to get target-fixated on the data, data, data, data, right? "It's not about the people, we're trying to figure out the data," right? "And inform me," and all of that. Rob Collie (00:36:33): And I think it's like when you're first learning to drive, I couldn't have the radio on. The radio was really distracting. And you certainly couldn't have a conversation with someone next to you. So, all you can do just to make sure that you're turning the wheel the right amount and all this kind of stuff. It's just overwhelming. But once you internalize all that stuff and you build the muscle memory and all those sorts of things, now your brain is free to do some other things. Like this data fluency thing we were talking about, it's neat how, as you climb that slope, you're never there, it's a perpetual journey, the other parts of the equation like the human things, right? They can come back. Rob Collie (00:37:12): An example, even just from our own business, we do a lot of internet advertising. And sometimes, when people at our company are thinking about this, now the wrong way to do it is to go and like, "Oh, let's go look at the ad words API and let's get fascinated by the tech around this." And I'm always trying to remind people that, no, no, no, we're trying to scale a human interaction. That's what we're trying to do. We're trying to reach people with our humanity- Jen Stirrup (00:37:43): I think that's so true. Rob Collie (00:37:43): ... and we're using a technological system to do that. It's a tool for the other thing. Jen Stirrup (00:37:50): You're so right. I think we should be using technology empower and enable. And I think my personal mission is about helping people. I find that rewarding, personally. I like things with a purpose, so that's why I do charity work with organizations like DataKind, because when you get someone crying because you've solved a problem for them and you've helped them, you know how incredibly grateful they are. But I think, for me, that's why diversity and inclusion, equality, and intersectionality more recently has become really important to me. Jen Stirrup (00:38:21): I'll just give you a few examples that's in my head. I did a project recently, and there was a woman of color in my team, and I felt that she was being talked over. I'm used to being talk over, softly spoken. But I could see it with her. And I just made a conscious effort to say, "I'm sorry, but I don't think she's had the opportunity to speak, and I can see she's tried to have some input." So, some of it's a bit like that. But some of it is directly saying, "What do you think? Sorry, we haven't heard from you," and pulling people out. And you know what? She was and is still incredibly insightful. And sometimes, the best data scientists I work with are people who can't code. And I think about her and I think about another woman of color as well that I work beside. Jen Stirrup (00:39:06): Fantastic data scientists, they both know Excel, but they can't write a line of code. And the reason they're so good is because they are such fantastic questions. That means the rest of us who can code have to then go and get the answers. And I think the knack of asking the right questions is such a gift, it's such a skill, and it's something that I am consciously trying to improve myself on. And I think diversity, inclusion, and equality is really important, but we wouldn't get anywhere with any of that if we're not allowing people space either to talk or we're not able to give them the space to ask the right questions. Jen Stirrup (00:39:42): Now, I am constantly learning every day. And to do that, I'm having to learn to get better at asking questions. And it is a skill to ask, but I think, when we're dealing with data, it's about helping people not to feel stupid if they're asking questions, because I think, with these particular cases, it's very easy to feel diminished in a conversation where other people are understand the technology, they can code, you can't, but you've got an insight. I know we talk about data-driven, but I like the term "insights-inspired," and I wish we had more of that because that, I think, gives us room for other people who perhaps don't understand the technology but do have business insights that I would never get, because they help me interpret the code or the data to make it better. Thomas Larock (00:40:28): So, you said data-driven, but you prefer insights-inspired. I think those are still two different things because, when I think of data-driven, I actually think of that in terms of, "I'm going to make a decision based upon what the data's telling me, not upon my feelings." The insights-inspired, to me, is how I get to the question I want answered, right? But I'm still data-driven. I think there's some overlap, but I also think there's a lot of space there where they are distinct, because I do believe in data-driven because I've been in those meetings where somebody's like, "Yeah, I don't really care. We're going to do what I think is right." "But the data says something completely opposite." "Yeah. That doesn't matter to me." And lots of those cultures exist. I love insights-inspired, and I'm going to steal that. Jen Stirrup (00:41:16): That's fine. I think we need both, actually. I'm sorry if I wasn't clear. But you're right, there is a good impetus for people to think, "What does the data say?" And I like that. I think the "insights-inspired" piece will help us to understand if the data's right. And I'll give you an example of something that I did. So, I was doing some work for the national health service and there's some data missing for a hospital, and it was not an insignificant amount of data. It was for about five years, the data. And I searched for it all morning, and I was just about to ,arch down the corridor to go and corral a DBA to ask him, Have we lost any data? Because I cannot find this." Jen Stirrup (00:41:55): And then, when [inaudible 00:41:56] was passing, she said, "How are you doing?" I said, "Oh, have you ever worked at this hospital?" I won't mention which one it is. And she said, "Oh, I was there until it closed for five years and it merged with another hospital." And I thought, "Oh, you've just answered my question. Right." Because I was sweating beads because I thought, "We've lost five years' worth of data." And I thought, "We've done that. We are in so much trouble," because it's a lot of data. It's a lot of patient data. No, no, no, no. They went somewhere else. And there was a very good explanation that I would never have got by the data. I could have hugged her. Jen Stirrup (00:42:31): And to this day, I still feel the palpable relief, because I was walking in the hospital, thinking we need a really good explanation for this. But according to the data, it was not there. So, I think, when I look at data-driven, I think they're two sides of the same coin, because insights will tell you what the nurse said, "Well, actually, it's like this," and they will add to the interpretation. Jen Stirrup (00:42:54): I just sat in a meeting once where one of the leaders said, "All right. So, we've got the data now?" I said, "Yes, everything's fine." And in front of four of his team members, he said, "So, we can get rid of the business analysts then, because we've got the data now." And even when I mention this, I still, at this point, feel my blood pressure rising, which is not good for me. I am well over the age of 40. And actually, I was stunned. I said, "How are you going to understand the data if you don't have your business analysts. Who's going to tell you what it means? "Oh." I said, "Are you really thinking that you can just throw your data at a wall, see what sticks, see what's left, and that's going to drive a business? Because, pretty much, that's what you're doing, if you are not involving the people who understand the business." Jen Stirrup (00:43:43): And after the meeting, I mean, some of them were crying, saying, "He was talking about me losing my job." And the people impact was terrible. So, this is where I've got my principals coming in. So, I went and I escalated that afternoon, and he was taken off the project the next day. That was due to happen. That was just outrageous. And if any of you who are listening and this is you, I love that team, their insights were incredible and I learned so much from them. And to the leader in that organization, please listen to your team members. You will get so many many great insights. Rob Collie (00:44:23): Wow. Jen Stirrup (00:44:24): Sorry, this is very cathartic for me. I'm glad you've brought me on today. Rob Collie (00:44:33): I mean, just watching your face as you told that story, I can see the emotions that you're feeling, right? Jen Stirrup (00:44:37): He's going to get this. Rob Collie (00:44:38): And it's a mix, right? It's a mix of the beauty of some of these people that you worked with, right? Contrasting with like this horrible, horrible attitude, at the same time, from this one individual. When you have all those feelings at the same time, it's like you need a new name for it. It's like, "What is this feeling?" Jen Stirrup (00:44:56): And I think the industry is like a pendulum, so we go towards data-driven. And for some organizations, they need good data-driven, so Tom's given a great example. But sometimes, it goes too far and they say, "Yeah, I read that buzzword. I'm going to do that." And then, there's an expense, something has to give. And that, unfortunately, was his team. Like you said earlier, Rob, it's about the people. We should be there to help people by helping people do their jobs better, not necessarily replacing them. That was not ever on the menu. Rob Collie (00:45:29): Yeah. It's counterintuitive. Sometimes, when your data system gets better, the right move is to have more analysts because there's more ROI in having them. Even just hiring a data professional services firm such as yourself, the reason to do it is because the ROI can be massive. Jen Stirrup (00:45:51): Yes. There's lots of unseen costs. I worked with an accountant last year who spent four out of five days a week merging Excel together. And I sat with her, I got to know her pretty well, I mean, remotely because of COVID. And eventually, she said, "Oh, I'm looking for a new job." And I said, "Oh, really?" And she said, "I did not incur a graduate debt to sit and do something that I could have done without my degree." She'd put a lot of effort and, same in the US, lots of student loans to do a degree. And she said, "Technically, my job title is accountant, but I'm not accounting. I am munging data around in Excel." And one of the projects I had recommended was data integration, right? And they wouldn't go forward it. They kept saying, "No, no, no. We've always done it this way. So-and-so om accounts does all that." But they never asked her what she wanted. Jen Stirrup (00:46:43): So, she left, and I was not a bit surprised because she said, "I want to be an accountant. I want to account." And I know that it's not my personal lifestyle. It wouldn't be my choice of a job, but for her, she just loved that, and she wasn't getting to do. So, sometimes, the causes are quite unseen if you're not looking after the processes or the data, because that incurs hiring costs, then, on staff onboarding costs that don't get included often as part of these business strategy projects. When I'm doing a data strategy, I try to include them, to say, "But what happens if you change? But what happens if you don't?" And you're going to lose people because your people, very often, want to be skilled in the later technology. Jen Stirrup (00:47:25): And I'll give you an example. One customer I worked with said to me, "We need your help with reporting services, SQL server." So, "Okay, good. I like reporting services." Then, they talked to me and I said, "What version are you using?" And they said, "2005." And I said, "Why?" "Because the application that's using it requires SQL server 2005 and we can't upgrade." Said, "So, what was the application written in?" "VB6," which you may have heard of that technology. It was around in 1999. It was last century. So, the data state was antique. I had no idea that it was that bad. But then, the application came up, and Microsoft still do a version of a Visual Basic. You can go to the site, the latest version... But the point being that the staff and that place had settled for VB6, they'd settled for 2005. That doesn't mean that you're getting the best team members. And when we worked, it was recommended an architecture. Said it was not touching it with our [inaudible 00:48:30]. Rob Collie (00:48:30): I'm still very fluent in VBA6, so maybe after we finished this show, can you give me the information of this organization? I might go apply. The last place on earth that VBA6 fluency is... Actually, that's not true. It's still being used everywhere. It's just not being used centrally. Jen Stirrup (00:48:53): Yes. I did say to them, "I am not touching any software that was not built in this century. So, if it's in the last century, you've no chance." So, re-architected, actually, we're using the Azure Cosmos... Thomas Larock (00:49:04): It's a good rule. Jen Stirrup (00:49:05): ... and dot... Yeah, it's a good rule. It's a rule to live by, you can quote me on that. I use no software built in the last century. In fact, I'm going to make that my new company advertising strapline. That's great. I like that. So, they're happily in Cosmos and .NET. And we used that because the developer said, "Hey, does that mean we get to modernize?" I said, "Yes. And you will either modernize or I will leave. Your bosses are going to have to modernize." So, they did. But again, that soft Scottish accent comes up. "Well, why don't we use software that's built in this century?" Rob Collie (00:49:42): It's a devastating maneuver. If we were making a card for you in a trading card game, that would be one of your two power moves, right? Soft Scottish accent. And the description of the power is something like, "Removes all defensive screen cards from opponent." Thomas Larock (00:50:07): Disarming. Jen Stirrup (00:50:10): Absolutely. Yeah. It's just funny how the data problems are really throwing up what's wrong with the organization. Obviously, they did that, but two years ago, I went to visit them again, just before COVID last year. They'd implemented a data science team and they just wanted some strategic consulting. And I was really pleased with how they turned around. So, sometimes, if you just find a problem like that, a small success, building those small successes, and they were allowed to up. I don't know if you see this, but big thing of what I'm doing when I'm in organizations is change management, but also a lot of that's people. And people tend to align themselves with success. So, if you can just show one small success, people get on board with it. Rob Collie (00:50:53): Yeah. I mean, it's everywhere in humanity, right? We're fundamentally pattern-matchers. And if you haven't given a population any positive patterns to match, no examples, it's amazing how stuck you can be. But one success, right? We have an infinite percentage increase in our population of successful examples. We went from zero to one. Like you say, the dog knows that there's five treats in your hand, right? We're not dumb. If there can be one success, there can be more. But if there's zero successes, that's powerful. Jen Stirrup (00:51:25): Yeah. And I don't know if you see this problem, but it's something I see a lot is people think maybe Tableau or Power BI, they buy this, it's going to give them a success. And it does, until the data starts to get hard. And then they either have to scale up in DAX, which is fine, but sometimes they don't have room or bandwidth to do that, so they get almost a bit depleted because they realize, actually, data's hard. We've never really nailed data as the human race. Rob Collie (00:51:55): It's always hard. Unfortunately, to sell software, to a certain extent, you have to sell the lie. If you're a software vendor, you have to se... Rob Collie (00:52:03): ... have to sell the lie. If you're a software vendor, you have to sell the lie that this tool is the magic fix, that it's going to make data easy. And I do actually, in a weird way, I kind of like blame Tableau for making this worse, but while at the same time, being very grateful to Tableau that they made interactivity a must have. Jen Stirrup (00:52:24): Yes. Rob Collie (00:52:24): I think they were actually, more than any one entity, responsible for us breaking this notion that reporting services and similar tools were it. Jen Stirrup (00:52:34): Yes. I remember the first time I saw Tableau. I had been hired as a developer for SQL server [inaudible 00:52:40] services and my boss said, "I think this is a future, this stuff, Tableau. Here's the download link. Tell me what you think". 10 minutes I was completely hooked and it changed my career because otherwise I would have probably stayed in the database reporting world and I suddenly thought there's a whole world here with stuff. So I love what they did. I really, really think it was groundbreaking. Thomas Larock (00:53:01): At what point did a report just become synonymous with the word "Tableau"? I have a limited experience and maybe it's an outlier, but to me, I always hear people say, "I'm going to run a Tableau report". I mean, it's just a report. I worked with Crystal and BusinessObjects, same thing I guess. And do people always qualify the type of report they're running as if that makes it more special or do people always say, "I'm going to run a power BI report"? Why is it always a qualifier? And in my case, I always hear, "I'm going to go run the Tableau report". I'm like, "It's just a report. It doesn't really matter what's the software that's doing it. It's just data. It's just a report". But I hear that a lot. I just figured I'd ask you two if that's the same experience? Jen Stirrup (00:53:43): Yeah. I think I'm hearing that more and more and I actually think it's almost going the other way, where people are only wanting interactivity, they're only wanting things they can click and tick. And what they're not wanting as much is a SQL server, mahogany red, forest green, slate gray, corporate template, because that was the what, about four templates you got with reporting services. So I see that more and more apart from the finance world. They still very much want it. But what I'd still see is a big need for tables. People still want to export to Excel. And I think it was you, Rob, who actually said this years ago, that the third most common button in Tableau is something like "export to CSV". Thomas Larock (00:54:26): Yeah. Rob Collie (00:54:28): Yeah. The third most common button in any data application is "export to Excel". Thomas Larock (00:54:32): Yeah. Rob Collie (00:54:32): Behind "OK" and "Cancel". That's the joke. And what it is, is an acknowledgement of, again, the human plane that this report, this app, does not meet your needs. It's in a way like if you could instrument your organization and find all of the "export to Excel" buttons that are being worn out, those are like the hotspots for you to go and improve things. That button being, click, click, click, click, click, click, click all day long, is telling you that there's a tremendous opportunity for improvement here, both in terms of time saved, but also quality of result. Quality of question that's even formulated. You mentioned questions earlier, asking good questions. Here's the problem. The ability to execute on answers and the inability to execute on answers, the friction, the inertia, that works its way upstream into the question- forming muscles. The question-forming muscles atrophy to a level where they fit the ability to execute on the questions. And so when you suddenly expand the ability to answer questions, it actually... You've got to go back and re-expand your question-asking muscles to be more aggressive, to be more ambitious. Jen Stirrup (00:55:52): Yes. I think sometimes the data-driven piece is trying to, in a way, subtly bring that back into play. It's okay to admit that we don't have all the answers and it's okay to admit that we need to ask questions. I think there should be more of that. Something that, certainly earlier in my career, asking questions was discouraged. It meant you didn't know it. It meant that you were vulnerable in some way. And I think as an industry, we need to encourage people to ask questions. I think with the diversity inclusion piece, try and make a conscious effort. If I think someone in the meeting is being quiet, regardless of the background, but at least I'm trying to watch out for that now, whereas maybe 20 years ago, I wouldn't have realized it, but sometimes people do sometimes need that extra help to speak up and speak out. They often don't know what to say or how to beckon to a meeting and say something. It's quite difficult. Jen Stirrup (00:56:51): Especially if you were being measured in your performance. I think sometimes people see things very confidently. And actually when you start to pick it apart, you think, "I need to as a person, stop believe in confidence and maybe thinking is that right, not how it's being delivered". I think they're stolen for quiet voices, hopefully like mine, who are trying to say things but I do find that harder to get heard. I think it's good that you do podcasts like this because I think it gives people the opportunity to talk about different ideas and how they impact people because that is important. There's loads of vendor podcasts that will talk all about the technology but we need to know better how to apply it. Rob Collie (00:57:31): When we were talking about starting this show, it was pretty clear we did not need another tech show. People who are working in tech, but are human beings, like yourself, and who are focused on helping other human beings. We weren't sure if it was going to work. It was one of those like, "Are people are going to listen?". Thomas Larock (00:57:45): We're still not sure. Rob Collie (00:57:50): We knew that we were going to like it, but yeah, it's building an audience. I've enjoyed it. And plus, it's an excuse to get together and talk with people such as yourself. If we just pinged you out of the blue and said, "Hey, you want to get on a two hour Zoom call with us and just catch up?". That's going to get pushed and pushed and pushed and pushed, but, "Oh a podcast? Oh, well, yeah. That's exciting". Jen Stirrup (00:58:14): Yeah. I know what you mean. It's good to, I think, to try and translate data and technology into something people feel is within their reach because I think there is still an element of people who are almost being scared of working with data. I deal a lot with CTO's, CIO. I was busy CTO and some way reports sent to their CFO because their CFO is over all of it, keeping costs down. The CTO has to work really hard to justify them. And I think what they want, ultimately, is not to appear stupid or not to know what they're doing. So some of these leadership conversations I have are about people saying, "Explain these terms to me. I don't know what a data lakehouse is. Do I need one? How's it different from a data lake? What about the warehouse? Is that going away or is that rebranded as well?". I know Microsoft talked about data hubs recently. If you're a data vault person, a data hub means something quite specific. It's been a term around for 30 years to mean something else. But I think sometimes people get very confused with the terms. Rob Collie (00:59:16): Like for example, the noun "dashboard" in Power BI, right? It's just a head clutching frustrating mistake. I mean a Power BI report is probably best described as a dashboard. The multi-visual, interactive experience, lowercase D dashboard is what I always want to describe it as, but no, no, no, no. We repurposed that word. Jen Stirrup (00:59:41): I know, and customers don't always understand it because they say, "Well, actually my report looks exactly like the dashboard. So I don't understand this publishing thing". So I have to try and explain that actually, we can take data from [inaudible 00:59:55] here and you can extra things. I'm interested to know actually, how much Power BI users spend actually making dashboards as opposed to making reports. And I just wish we'd ever the answer to that because sometimes you just want to get reports that they can run in their desktop or not always sometimes use a browser and just have the reports and have them open on the actual dashboards higher up. So I feel that's a bit of a separation that maybe wasn't required to have. But Tableau does something similar, doesn't it in a way? But I think with Tableau, it's a bit more clear that you're putting these things together. Rob Collie (01:00:29): Well, we were talking at the beginning about the importance of comprehensive training sets. Well, let me just tell you, we only need one data point here. I, as a Power BI user, have never once created an actual Power BI dashboard. So let's just conclude that that's it. No one uses them. But yeah, I've never felt compelled to need one. I tend to put together, what I need in the report. Jen Stirrup (01:00:56): Yes. And that's what I do because I'm trying to get the customer from A to B. I'm trying to do it quickly and I can see that they've reached on that tool ceiling of where they want to go and then they've got this other thing they need to do and they don't understand why. So sometimes it's a battle I just don't have because I just think, "You know what? These often been through so much to get to that point in the first place, cleaning data and getting access to the data and all the things that are hard and even understanding what they want in the first place". I try and work out where the fatigue is. Rob Collie (01:01:28): Yeah. I think there's a certain hubris just in the idea that a user will go around and then harvest little chunks out of other reports and take them completely out of context. Anyway, we didn't come here for cynicism today but- Jen Stirrup (01:01:43): I have plenty of that. Rob Collie (01:01:43): But it's still there. We can't really help it. So it's come up a few times and I want to make sure we actually make some time to talk about it specifically. So you've mentioned a number of times, inclusion and diversity and already a few anecdotes within your own professional organization, within your own firm. Outside of your own data relish organization, what are you up to in this space around the diversity and inclusion as a cause? You're very active in the community in this regard. Can you summarize for us what all you're up to? Jen Stirrup (01:02:15): Yeah. I've started there to talk more about intersectionality. There's a lot of data, which I don't have to hand, which is terrible, that shows that it's the intersection of people's lived experiences that can sometimes work together against that person. So for example, we know that women are paid less than men, and regardless of the stats that we use, that's the number that comes up. There is some data that talks about how for black women, it's even less and for Latin women, it's even less again. So the idea being that, for people of certain ethnicities and backgrounds, it's interacting with the fact that their background, their ethnicity, their race is interacting with the fact that they're female, and both of these characteristics together are interacting to produce an adverse outcome for the individual. Jen Stirrup (01:03:06): Now this is quite an interesting area, is something that's been part of academic research for about 30 years. And there's some great researchers out there that talk about this. I want to say [Christy Reynolds 01:03:18], but I need to double check that surname because I'm not very good with names. It's age. Which is another characteristic as well. I was called an interfering old bag by somebody. Rob Collie (01:03:29): What? Jen Stirrup (01:03:30): I thought was really funny. There was a community person. I don't know if they realized it would get back to me. And I said, "Okay, so diversity and inclusion, obviously it's there with the bag aspects, but old as well. Okay. Thank you very much". Rob Collie (01:03:41): Great. Jen Stirrup (01:03:43): And yes, I am interfering. You know I am. Rob Collie (01:03:46): Triple word score. Jen Stirrup (01:03:50): Exactly, and it's... For me, I just thought, "Okay, so it fits". It was actually somebody I know that said that behind my back and they told somebody else and they said, "So-and-so has said this about you'll and I said, " Okay, that's fine. I am interfering, absolutely. If I see something I don't like, I can realize that being a staunch supporter of certain initiatives, of causes, can make me appear interfering". If I feel strongly about something, I will speak out and that can be cast fairly quickly into, "She's interfering again. Who is she? She's an old bag". So I think that phrase for me, made think about those two things. One is my age. I'm 48. Jen Stirrup (01:04:28): So there's that and female as well. I thought whatever I was interfering about, does it matter if I was right or not? Because whatever I was interfering about has got lost somewhere and it's got lost in the fact that my other characteristics were brought up as well, rather than seeing, "Well, actually Jennifer is wrong about something", which I could have accepted, I think. If someone had said, "Actually, you've got that wrong", I would rather know that, because then I can rethink or change my mind or perhaps give an explanation and then it turns into an adult conversation then, rather than back and forth. Which I don't want to do, I am not interested. So I didn't engage with the individual. I thought, "I can't change your mind, because if you're just going to bring up my age and my gender, I'm not getting a good starting point. Maybe you need to listen to someone else and see what they think". Rob Collie (01:05:22): How do you achieve that peace? I'm sorry to interrupt, but that's just killing me. How have you reached the point or have you always been like this? Teach me. Can you teach me that peace? Jen Stirrup (01:05:35): I think I got to the point of fatigue actually, where I realized that no matter what I said or did, there was always going to be haters who would always cast whatever I said or did in a bad light, and it doesn't matter what I do or say, that will be twisted. Someone wrote me a piece of hate mail recently on Instagram, which said about, "You said something about this blog post, about this person", and I said, "No, I didn't". And then I thought, "Should I go back and ask them what was I supposed to have said and who about?". Jen Stirrup (01:06:05): In the end I just wrote, "Thank you very much for getting in touch. I appreciate it and I think we need to close the conversation here". And I just left it because I thought, "Actually, I don't know what's going on, but whatever it is, I'm going to lose". And I think maybe I did start to lose in some ways, because I thought I am just going to lose and I can keep being dragged around by other people and it's spending energy. I don't know if you've ever read The Art of War? Rob Collie (01:06:31): I have. Summaries of it anyway. I'm not sure that I've read the whole thing, but it's not very long, is it? Jen Stirrup (01:06:36): It's not very long and there's a bit in it that really speaks to me. It's the bit about... So obviously about war, but it talks about strategy and it says that you should regard your enemies... stand back from them and regard them as a boulder rolling down a hill, and if you stand back, they will eventually run out of energy because there's going to keep rolling. And the whole point about you as that strategy person or a tactician, is that it's all about timing and it's spending energy. And I realized actually, the best thing I can do for myself is be very careful in how I spend my energy and be very careful about my timing. So I think I've been trying to ignore stuff online for quite some time. I'm not going to pretend it hasn't been challenging or that it hasn't been a hurtful. It's been both of those things. Jen Stirrup (01:07:23): And I think that, because it would speak out a refute to do anything, people tend to believe bad things and they just think, "Where did all of this come from?". I've come to the conclusion as well that I think we talk a bit about mental health in the tech community. I think that people are struggling and have sort of come to the conclusion that some people who are throwing stones and things are actually not in a good place, because if you're a happy person you're not behaving like that. Jen Stirrup (01:07:50): I'm not seeing people are mentally ill, and even if it was, it's not a bad thing to be a mentally ill or have mental health issues. That's not what I think, it's not what I mean. I just think it's coming back to asking questions again and timing to say, "Actually, I'm going to stop behaving like that because that's not a normal or proportionate way to behave, and I have to perhaps seek some help from the right person. Whether that's a friend who's going to give me a very honest answer or perhaps stepping outside my echo chamber or my [inaudible 01:08:22] chorus, so that I get different perspective". Jen Stirrup (01:08:25): And as humans we like to flock together in terms of who we like, and also we go off into groups, perhaps through a shared interest and that for a community, and that's a good thing. But sometimes it's a bad thing. It's something I should probably mention before. I've been speaking to the police for a long time, telling them about online harassment, because some of it is really unpleasant and some of it I have had to report to the police. I've been dealing with the London Metropolitan Police and [inaudible 01:08:52]. We all have. I've handed them access to things like my Facebook account so that they can see some of the stuff I am being sent. I don't mean people seeing my Facebook messages. They are deadly dull and boring. I do not live an interesting life. It tends to be things like, "What would you like me to bring you back from the shops?". Jen Stirrup (01:09:12): And what I have found is that there's much more sympathy for victims of online bullying than there used to be, and it's too much to the point. I've speaking much more closely and going through a sort of community resolution process at the moment with someone. I don't know who they are due to protection. Isn't telling you that. But I have been speaking to the police where someone has been caught, questioned, and they're going to write me a letter of apology. And I think that is going to be tremendously huge for me, because all I want is an apology that it's just not going to happen again to me or to anyone else. So sometimes speaking up and speaking out, you don't see results right away, but sometimes if you stick at it, collect the data, collect the evidence, speak to the right people, sometimes you can get a result. Jen Stirrup (01:09:59): In the minute I feel full of nervous energy about it. I'm not pleased yet because I think it's going to take some time to uncoil from all of that. But I think the point I'd like to make is, when people are seeing things by other people online, there are consequences for the victim, but to also see that for them personally, there can be consequences as well. You know so somebody has now to spend time in the police station, which is taking them away from their work, they have to explain to the family, that kind of thing. So I think what I'd like to see in the tech community is more proportion, a sense of proportion. If I have upset somebody, they need to talk to me directly. I will hear them out and if an apology's due then yes, absolutely I would give someone an apology. Jen Stirrup (01:10:44): I had to apologize to someone a few weeks ago. I said something that inadvertently offended somebody. I had no idea. I was very upset about it and I learned how that had been taken and wrote them an apology in Facebook actually, which they accepted. But I wanted to give them the opportunity to say their peace and I talk a lot about people speaking up and speaking out, and I need to take it when other people take the time to speak up and speak out against me as well. And I did the right thing and I learned something about how something came across. I'm really actually grateful that they took the time to do that. Rob Collie (01:11:17): Just hearing you say these things live, it brings it home viscerally in a way that really no other medium does it. It's the cliche now in this increasingly hyperpolarized world, there are people quote unquote, on your side of these issues who are incredibly combative, unfriendly, non listening, right? There's people on both sides, right? That are like this. I'm not saying like, "Oh, it's the inclusion and diversity people that are so terrible". We've got those people everywhere, but it's just so hard to imagine talking to you, anyone characterizing you as a villain. It's just jaw-dropping. If I disagreed with you on every single thing in the world, I still... there's no way that I couldn't get along with you. Jen Stirrup (01:12:07): Thank you. Rob Collie (01:12:11): I really, really, really don't get it. Jen Stirrup (01:12:13): I think some of it is perception and some of its proportion. I think, as I've got older and I'm divorced... I've been through a bad divorce, and I think it had been true so much in some ways that I've suddenly developed a lot more empathy. And I think maybe you develop more empathy as you get older. So maybe it's that. Maybe if you spoke to 22 year old me, you would find a very different person because I am opinionated about lots of things and I do interfere, but I think maybe I've just got to learn a better balance between knowing when people don't want to hear it anymore. And me realizing that actually, you do need to speak up and speak out but when does that stop? I've been thinking more about this in the past few days actually. You can speak up and speak out, but maybe at some point I have to get better at understanding some people are just not going to listen. Jen Stirrup (01:13:05): And then that comes back to your previous question. How do you walk away from that? And I think you kind of have to, to protect yourself and maybe think about family members who are affected by seeing you upset. I think the community needs a lot of healing. I think past is appearing is not below any healing and I think [inaudible 01:13:24] over topics which are maybe not constructive and going to help anyone, is going to help. But I'm happy to talk about things like diversity inclusion, intersection and equality and inequality where people feel that perhaps there's something they could apply or maybe help them to think. And I should add that I am learning about these things all the time, and I can and I do get it wrong because I talk and ask questions with people because I am learning. And I'm very fortunate that people have been quite patient with me I think. Rob Collie (01:13:55): I just want to, almost sarcastically, go, "Wait, wait, wait, wait, wait, wait, wait. We're on the internet now. What is this humility and 'maybe I don't have all the answers'-stance? Like you can't do that. You've got to go out there and just bluff that you know everything and you don't understand why everyone else can't figure out something so simple". It's so easy on the internet to curate one's own profile. When you're up close and personal with someone, it's hard to hold it together. The human flaws that we all carry, they just kind of leak out if you're in the vicinity of each other and you're watching closely and if you're like working together in person or whatever on a daily basis. But if you're an internet personality or just really honestly, like everyone is becoming an internet personality in some way, right? Rob Collie (01:14:45): If you have a social media account, you have the opportunity to start broadcasting to the world a curated picture of your life. That's very, very, very seductive. Like, "Ooh, I can put out there only what I want people to see". You don't even need to consciously think about it. And then you get all these examples of other people who are doing it. It's almost like the prisoner's dilemma, right? If everyone else is doing it, but the thing is, you don't know that they're doing it, right? You see all these people that appear to be so together and so you need to go out there. The pressure to go out there and be the same, pretending that you have it all figured out, is super high. It takes a lot of courage to go out there and say, "Yeah, I'm figuring this out. Work in progress". Jen Stirrup (01:15:28): Yeah. I'd rather give an answer that's got integrity than pretended expertise I don't really have. I wish I had better answers. And maybe if you ask me in a year, I might have changed my mind. I think people are in a bad place in many ways and I've tried to be more thoughtful about actually maybe people are not accessing or getting the support and help that they need and that is coming out online. And maybe you just need a bigger sense of proportion that I don't think we get online on social media. Rob Collie (01:15:55): When there's an audience, everything gets orders of magnitude more toxic. For a while there, at a point in my life, operating my blog was a big part of my professional existence. I would get trolls that would come and attack out of left field for assuming just sometimes like incredibly bizarre reasons that you wouldn't even understand. I had a guy emailing my manager. They had tracked down who my manager was at Microsoft because I was still working there at the time and was emailing him that I had done something like I had not done at all. Like, it was crazy. He thought I hacked his PayPal or something. It was just totally out of the blue. He wasn't part of this tech community at all, right? That's the most egregious example, but I had an amazing, in the end, like an amazing track record turning these trolls. Rob Collie (01:16:39): Engage with them one-on-one privately behind the scenes and don't lead off with a punch. Lead off with, "Hey, hey, hey. What's what's up here?". And it was amazing. My expectation going into all these interactions was that this person is just unhinged, you know? And they turned into great people. I think like 9 out of 10... I think there was like one that didn't out of all the top 10 trolls from my history of operating the blog, nine of them turned into people that if I'm in their city, I'm going to say hi to them. Jen Stirrup (01:17:09): That's nice. I think you can try and give people an opportunity, but I think the other side of it is me thinking I have to look after my own mental health too. And I think it's better just to say, "Hey. My email address is online. I'm here on LinkedIn. There's always lot's of ways to get in touch with me. You can come and talk to me, I know". And I think maybe as you get older, you're better at picking your battles. Maybe it's a bit of that. Rob Collie (01:17:34): Yeah. Jen Stirrup (01:17:34): So I wish a [inaudible 01:17:37] face for people who are going through that. I've just decided just to screenshot, block and move on. And I feel that every blog post right now, I have to preface it with "This is not about aimed at any individual. This is just some observations about intersectionality". So I've got a blog that I need to publish about that and the last paragraph is about, " I'm not picking on anyone. I'm just trying to highlight an issue that-" Jen Stirrup (01:18:03): I'm not picking on anyone. I'm just trying to highlight an issue that is our various characteristics interact, and it's something that for me talk about diversity and inclusion, we focus on women in tech. And not that that's not a normal thing, because as it tends focus on white women in tech, and I'd like to see a mixture, women of color included as well. And a woman of color wrote to me last week, she said, "I'd like to share some of my experiences with you because women in tech is not including women of color." And I've actually got a meeting with her next week. I don't know what I can do or say, but I'm going to hopefully use it as a learning experience. It's not that I've said anything bad. She just said, "We're not talking about enough about women of color." And I feel it's probably fair, and I think it's important message. Thomas Larock (01:18:52): So I want to make sure, Jen, that you understand that I want to thank you for interfering. That's an important role that we all have to play at times. It's one thing if you're just always interfering for the sake of interfering, but when you see that interference is necessary to advance, to make something better for everyone, at the end of the day, we should all be looking to just simply do good things for ourselves, but for each other. So sometimes that interference is necessary. And I know you and I have interfered with things and tried to get things to a better place, and it's necessary and I don't want you to feel that you should stop or that you're not supported. If you ever need a pick me up, just call me and Rob, we'll spend a couple hours telling you how great you are. Jen Stirrup (01:19:41): Thank you. Thomas Larock (01:19:41): But interfering is necessary. The thing about the online, like Rob was talking about, I wanted to say to Rob, I think part of it is that when you're online, there's a tendency to feel seen. And what I mean is if I make a comment, and I've learned this over specifically the last five to six years, if I just make some comments about a behavior that I've witnessed that I think is bad, inevitably somebody online thinks I am talking about them, specifically. Like you probably think this tweet is about you, don't you? And it's true. I can track things out. You've had those trolls and I've had my share of trolls and abusers. And inevitably it always comes down. I feel you're speaking to me about this and therefore I'm going to stand up for myself. And I'm like, "How did we get to that point where you see something online and you think that person must be talking about you?" If that's the case, then you need to do some inward reflection about what you're doing instead of attacking the person for calling out a bad behavior. But anyway ... Rob Collie (01:20:51): It's like the old joke, if you go to work every day and there's like one or two assholes you run into, that's just normal. But if you go to work every and everybody's an asshole, you're the asshole. Jen Stirrup (01:21:00): I think it's a mix of confirmation biases. We've got these sort of bases in our heads. One is I want to say Dunning-Kruger, but I don't think that's right. And where we think we know more about something that we do. And the think there's a related bias, which I've forgotten the name of that, which is if we don't understand something, we think it's going to harm us. And I think that the bias is there if I've maybe put up a tweet or a blog post, they haven't understood what I'm saying. So then some way of thinking, she's out to harm me. And the two things together, I know a lot about this and I don't understand what she said, she must be able to harm me. And I think people are jumping and mixing biases. I've got blog posts in my head and it's really something I try and practice doing is, what do I see if I see something that I'm uncomfortable with? I've got better over the years. It's saying, excuse me. Jen Stirrup (01:21:59): And then, this is a Scottish thing, and I know people on the podcast can't see it, but pointing your finger is actually very Scottish thing. So I remember it being at [inaudible 01:22:08] a few years ago, when some of their guys were making comments with one of the female presenters and I pointed my finger at them and I said, "I'm speaking to you, lot. What do you think you're saying? Stop it." Now, I'm getting better at calling people out as I see it, then I used to be. And I think maybe it's partly because I'm more aware of it maybe but also I think maybe I'm seeing more of it as well, so maybe there's that. And then it's just confidence to go up and point fingers at people. Jen Stirrup (01:22:37): But I think something, a line I cultivate is, are you kidding me? And that works everywhere. So anything that you've seen ever. And I want to sort of blow post an axe at something I'm trying to get better at. And the reason for that is I was in [inaudible 01:22:54] a while ago and somebody said something against the Jewish person. I live in an area with a lot of synagogues. And they said something which I will not repeat and asked him if he was okay afterwards. But I just didn't know what to say. And I thought, I now have to practice situations where it will prepare me better for speaking out when I see something which makes me uncomfortable. And it's a sort of indicator of the world that we live in, that we have to practice and have a stock series of phrases. And I love to blog about this. And the reason I mention it is just because I'm still figuring it all out. And I wish I knew a better way forward, but I think that reacting to trolls, reacting to people online, reacting when I see something, I think if you're a nice person, you're not expecting something. Rob Collie (01:23:42): Yeah, it's surprising. Jen Stirrup (01:23:43): And it surprises you and it catches you off guard because you think, "Well I never think like that. Where did that come from?" Rob Collie (01:23:48): Just as a humorous aside, you mentioned that are you kidding me is proven to be very effective. I have not found this to be very effective in my marriage. It is exactly the wrong thing for me to say, even though it is a go-to. I keep thinking that this will be the time that this sentence works. No, it doesn't. Jen Stirrup (01:24:08): No I wouldn't recommend it to a spouse. I'm divorced. Rob Collie (01:24:13): So am I, so am I. This is marriage number two. I want this one to work. It doesn't mean that are you kidding me is the move. Don't do that. Jen Stirrup (01:24:25): Just if I see something. So if I saw that incident again on the train where that man said something to that gentleman, it's pier capping things on, I would be best to prepared now. Thomas Larock (01:24:30): I'll try it with Suzanne to get some more empirical evidence. We'll see how it goes. I'll report back next podcast. Jen Stirrup (01:24:37): Suzanne is lovely. I don't think you'd ever have to say that to. I remember meeting her and we discussed Scotland because she had favorite TV programs. Thomas Larock (01:24:45): Yep. Rob Collie (01:24:46): Oh, I don't have to say it, either. It's not like it's justified. I'm just, anyway, I think it's fair to acknowledge that my nine out of 10 trolls story. All 10 of these trolls were male, shocking, I know. I definitely benefited from being a guy while handling them. There's a guy language and a guy code, and there's a lot of subtext going on in an interaction between two men that is very different. Whether we like it or not, it's very different when a woman is the target of it, and the way that you deal with it. It's weird. Even over the internet, there's this almost primal behavior going on. There's this threat that's been signaled, and then if there's a man on the other end, he turns around with some version of, "Oh yeah." Jen Stirrup (01:25:37): Yeah. Rob Collie (01:25:38): And there's sort of a deescalation that happens at that moment. And I don't really think that those tools are available to women in the same way that it's available when it's a guy to guy interaction. That's tricky. Jen Stirrup (01:25:54): You're taught to be fair and move. Rob Collie (01:25:55): Yeah. If you use exactly the same words that I do in an email response, it's not going to come across the same way. It's not going to get you the same result that it gets me. Jen Stirrup (01:26:08): Yeah. Rob Collie (01:26:08): I just basically said some version of, "Oh, okay. Oh, come on. There's got to be some sort of ..." There's no doubt that your success rate with those same words would be different than mine. Jen Stirrup (01:26:17): Yeah, it can work either way, I think. Sometimes it's just like the old bag is speaking. I'm not going to listen to her. But the other side of it is, "Hang on a minute. She actually said something to me." And that can cause a reaction. So sometimes it can work either way. It's quite hard to know. And I think trying to codify that into a checklist is quite hard. But I just thought, how can I best get better at dealing with these situations? Because I think as we come back to the world after COVID, we'll probably see more of this similar misunderstandings. And a few years ago, I tried to set up a decency charter and a code of conduct in [inaudible 01:26:55] . And I was surprised because not that many people adopted it. I think only two organizations did. And I thought with a decency chart or say something like, "We will be inclusive. We welcome everyone regardless of color and age and everything else." I put that together. And people wouldn't sign up to it. And I thought, "How can you not? How can you not" I don't know. Rob Collie (01:27:17): I want to go read it. And I want to go find all of the incredibly controversial content that I object to in this charter. I'm just kidding. Jen Stirrup (01:27:29): I know. Rob Collie (01:27:29): I'm expecting it to be a hundred percent obvious innocuous. I can't wait to say, "Oh hell no, we're not adopting this." Jen Stirrup (01:27:44): I'm going to throw it out. That's getting a red pen through it. Thomas Larock (01:27:47): Rob, our blog posts will just be, "Are you kidding me? Are you kidding me?" Rob Collie (01:27:54): I've even, by the way, on Facebook a long time ago, I made a fake dashboard of all of the buttons and levers available to me in a conversation in my marriage and giving them all labels. But there's this one giant button in the middle of the dashboard that's, are you kidding me? It sort of represents me just failing over and over again. Thomas Larock (01:28:21): Put that in the show. Rob Collie (01:28:22): The completely stupid rob dashboard for ... anyway. Jen Stirrup (01:28:31): Yeah, I was just trying to find stock phrases that I could just have in my head to respond with really quickly. Where I could just, if I see something. There's that and another one is which I find works if I'm getting harassed in the street or something, is I just say, "Oh, grow up." No one can say anything when they're told to grow up. I've never, ever had a good answer. So say I'm just walking around, minding my own business, walking with a laptop and going back to my car. And somebody wolf whistles. I'll just say, "Oh grow up." And I've never had a good answer to that, ever. What I'd like to do is find these phrases. Where if I see something, it makes people stop and think. I don't know, I wish I had a better answer. I'm still trying to figure so much out and I think I'll figure it all out and then I'll day the next day or something. There's so much to learn. Rob Collie (01:29:17): I know, and all this acquired wisdom, you start to like really cherish it. Like I badly, badly, badly want to share as much of it as I possibly can with my children. It seems like such a waste to develop it and then it's all lost. Like Roy Batty said like tears in rain at the end of Blade Runner. You mentioned sort of having these stock phrases on speed dial. I read one time about this guy who had been in prison for a long time. And he says 20 years later after being out of prison, he still has this five punch combination that he memorized for himself on speed dial. At any moment's time, he remembers exactly what it is. He had to have that to survive. Jen Stirrup (01:29:56): Yeah. Rob Collie (01:29:57): Not a good sign. Jen Stirrup (01:29:58): No, it's not. Rob Collie (01:29:59): When you need these things. Jen Stirrup (01:30:00): Yeah, and we shouldn't need them, but it's unfortunate that we do. And [inaudible 01:30:05] do self-defense classes. They're so helpful. Rob Collie (01:30:13): So you're up to like a two to three punch combination now. Takes a little while to get to five, I suppose. Jen Stirrup (01:30:18): Someone tried to mock me in my local village. I'd got some sausage rolls for my son and had them in a little bag, in a baker's bag. And this guy came towards me and I thought, [inaudible 01:30:30] with his fingers. And I thought, "He's going to mug me to get my thing." I got nervous. And then I remembered all the self-defense training. One of my friends is a former Olympian for the UK in karate. And she's very much about, you need to be on your guard all the time. I attend her lessons and I'm not very graceful at any of it. But he tried to grab my bag. And she taught me, use your elbow. And all that stuff because it's muscle memory because I practice every week. I'm trying to lose weight. So I hit someone in the nose with my elbow and he jumped back because it didn't expect it. So he's twice my height and half my age and it's the last thing he expected me to do. And as he came forward, invited himself, I punched him in the temple and the side of the face. And I'm laughing now, but at the time I was absolutely terrified. And it wasn't worth it over the sausage rolls. I should've just given them up, but just the fact that we have to be on our guard all the time. Jen Stirrup (01:31:28): And if he'd said to me, "I'm hungry and homeless and penniless," I would have given him it. But just the fact that he didn't look any of those things. I think he just thought, perception, older female, I'm going to take whatever she's got in her bag. So when I punched him and he fell to the ground, actually, and I just walked away. And I guess that I never thought I would ever need self-defense, but I think these things that we see sometimes in our daily life are showing us that we do need some form of self defense against seeing things and empowering ourselves to see something when we do. BEcause I think that's one of the hardest things is speaking up and speaking out. And I know I talk about it a lot, but I find it tough. Jen Stirrup (01:32:14): And I guess my lesson here is can we develop some sort of self-defense when you look at community events where we have got a code of conduct and my decency charter. It doesn't have to have my name on it. I don't care if it does or it doesn't. I just want people to feel welcome and that we are doing self-defense and other defense if we see something. Because it's a shame because as Tom and I know, we spend hours if not days that pass talking about anti harassment. And we developed, I think, a great piece of work on that. I think that's something, a past legacy that we should be incredibly proud of actually, because we put so much work into it. Was it perfect? No, but we're back to 80/20 again. It covered 80% of cases, edge cases sometimes not so much, but that's why the red cases, because there was the things that you maybe, or I didn't think of, maybe tried to be as prepared as we could. And I think it's almost like we need one of those for everyday life. Rob Collie (01:33:11): You got in a physical altercation. Jen Stirrup (01:33:15): Yes, I did. Rob Collie (01:33:17): It sounds like you're knocked him out. Down he went. Thomas Larock (01:33:21): What I took away was she left him for dead. Jen Stirrup (01:33:26): Yeah. Rob Collie (01:33:28): She walked away to go after his family. Jen Stirrup (01:33:34): Walked away because it was me scooting away, because I thought I'm just going to scoot away because I've done this and I'm embarrassed. And I was full of adrenaline and I didn't want to run because then you look like you're guilty. Whereas he was coming after me. So I turned back about halfway down the street and he was dusting himself off. And actually a few weeks later I was in High Street and someone stopped me and they said, "I saw what happened with that guy. He's been doing that for some time and somebody needed to teach him a lesson." Rob Collie (01:33:59): So you interfered. Jen Stirrup (01:34:00): I interfered. Rob Collie (01:34:03): You interfered with the normal order of events, which is him harassing people and taking their stuff. Jen Stirrup (01:34:09): I mean, change not too much. What's your bio, actually. I think I've now made a life objective to be an interfering old bag. Rob Collie (01:34:17): So are there any sort of pithy, short tips that you would provide to organizations, sort of like easy things to remember, easy things to do, new habits to develop or old habits to discard that can advance the cause of inclusivity? Jen Stirrup (01:34:35): Be kind. The reason I say that is because I was on customer set a few years ago and I witnessed something and I'd rather not say what it was, but there was an incident I was unhappy with and nobody stuck up for that person. And it's just sometimes in technology companies, there isn't that culture of speaking out. And I thought, "Well, I'm going to speak out on behalf of that person." And I wrote a report, actually. They asked me to write up a report about it, which I did. And I said that, I should try and find it, but wrote this phrase about acting with kindness. Jen Stirrup (01:35:06): I think the other thing I think is something that Mark Sousa used to say a lot is that destiny that joins the passport and it has helped me so much is assume good intentions. So for me and other interfering old bags, that I need to remember to look for the good intentions. Because I think I can be quite a negative person and I'm not always looking for those. And I think that was an incredibly wise thing. Yeah, I've forgotten the actual phrase which I gave to that customer. But I did say something along the lines of, and I actually suggested even ambassadors of kindness because the culture was really unpleasant. I shouldn't have to tell companies this, right? Rob Collie (01:35:47): No. Jen Stirrup (01:35:48): Anyway, I left and I don't know if he did anything. Rob Collie (01:35:51): So many table stakes. Here's one, I haven't taught a class in a long time, but I used to teach all the time. And one of my observations from teaching classes in DAX and data modeling, power BI, whatever, is that the population of students in the room was always at least 50% female. On average, over time, it was slightly higher than 50% female. So the quantity was 50/50. And then quality-wise, my guess as to who the best student was in the class at the end of a couple of days was again, coin flippy, whether the person who I thought was probably the most promising half the time male, half the time female. And yet our company, the people who apply for jobs with us, overwhelmingly male. Rob Collie (01:36:36): I feel a number of things about this. First of all, I just feel that it's a terrible shame. There's something wrong here. You talking about women in tech and all of that, usually you don't start from a baseline like 50/50. We're in this place where we have an amazing baseline. We're already in a great spot. You mentioned things like asking the right questions and everything. I actually think that, might be unpopular to say it, but I actually think women are probably better on average at formulating questions than men. The talent base is there. Where are we failing? Jen Stirrup (01:37:05): Yeah, I think there's a few things happening. I tend to get lots of job applications of people speculatively finding CVs, and the sense you get a lot of women and people from LGTB backgrounds as well applying. And I think maybe some of it is about positioning the company. So if you position it as, "Yeah, we're all about the data. We're all about the technology. But there's our mission in there." Because I do work for DataKind who are a data science charity. They always get 50% new female inclusion and I don't really want to see 50/50 because some people are non binary as well. So there's a healthy representation there, I think. I'm not non binary, but we are attracting people from at least three genders and that's good thing. But I think the reason for that is I get some tacitly, technically astonishing women are doing fantastic things. And I think it's about the way that it's maybe marketed as being inclusive and secondly helping people and having a really good impact. Jen Stirrup (01:38:09): I think people tend to be incentivized in the workplace is that they're having a good impact or perhaps being salary motivated. And if you go in more with the mission and the purpose, then that's a good thing. I mean, that's just I'd have to look at your company website or any company website. But I think some of it is what do you blog about? Are you just maybe talking about technology or are you talking about making things better somewhere for somebody. And maybe that's quite general, but when I go into Datakinds data dive and I'm sat with at least 50% women, they are harnessing people of all different backgrounds. And I think it's about that we have a mission. We and our skills and technology expertise are part of us, but we're here all with a common purpose. And I think the tech community needs that as well. Jen Stirrup (01:38:57): I've often said I'd love to see Microsoft Ignite, for example, doing a Data Hackathon Day. And I think they've tried something like this sped would think it was dialed up enough. I would love to see them partner with Datakind as an example and do something really good over the course of a conference like that. I think I had the idea sort of after we left Pass. So don't think it was something I did as part of Pass. But the thing is, people will build a sense of community and the get better technical skills because they're interacting with people of all sorts of backgrounds. Jen Stirrup (01:39:29): One of the guys on my team is at Cambridge, University of Cambridge PhD doing a post-doc in genomics research. And he was doing location mapping for one of the charities that we were working for. And he was just phenomenally brilliant. I think when you deal with people like that donating their time for a cause, they tend to be nice people because they want to help. And I think technical community, speaking generally, it doesn't always have that. It's like, who's this serving? Is it serving our company? Is it serving the individual because they're building their career? I think when you talk with tech community, it's about, "Yay, come present and you'll increase your career skills and your technical skills." What about saying something like, "Hey, come along and you can help people. You can have a charity. You can have an impact in people you'll never meet." I think I'd like to see more collaborations. Rob Collie (01:40:19): Jen, thank you so much. I've really enjoyed, but I've also just really appreciated this conversation. Jen Stirrup (01:40:25): Thanks so much. Announcer (01:40:26): Thanks for listening to the Raw Data by P3 Adaptive podcast. Let the experts at P3 Adaptive help your business. Just go to P3adaptive.com. Have a data day!