Fiona (00:07)
Hello and welcome to unDUBBED the podcast where data dreams get real. I'm Fi Gordon.
Sarah (00:12)
And I'm Sarah Burnett. Thanks for tuning in. unDUBBED is unscripted, uncensored and undeniably data from AI breakthroughs to everyday reporting drama. We keep it real and a little bit fun.
Fiona (00:24)
we started this to share raw, real stories from the data world, lessons learned, screw up survived, and breakthroughs worth celebrating.
Sarah (00:32)
And hey, if you like what you hear, give us a rating or share or hit that subscribe button.
It helps others find the pod and keeps good data vibes going.
Fiona (00:43)
Today's episode is the Consultants Toolkit for the Modern Analyst. Whether you're already consulting or just want to think like one, this one's for you.
Sarah (00:55)
With so many analysts stepping into consulting or leveling up their internal game, it's the perfect moment to talk about what makes a consultant mindset so powerful.
Fiona (01:06)
Joining us are two data legends from the UK, Chris Love and Carl Allchin
Sarah (01:13)
Chris is a Tableau Hall of Fame visionary, Alteryx ace Emeritus and a technical architect at the Information Lab. With years of experience solving complex problems using Tableau and Alteryx, he's also worn hats as an account manager and has won the Alteryx Grand Prix.
Fiona (01:32)
Woo! Carl is the global head coach at the data school where he trains future data consultants worldwide. He's a hall of fame visionary too and the author of many books such as Tableau Prep Up and Running, Communicating with Data and Data Curious.
Chris Love (01:33)
Hmm.
Fiona (01:50)
so if you've ever heard how to clean or explain your data better, Carl's probably helped you.
Chris Love (01:50)
Yeah.
Sarah (01:56)
And if that wasn't enough, they've launched their own podcast, Love Data, Love Growth Conversations, which I've been loving. Definitely give it a listen.
Fiona (02:06)
With their combined experience, Chris and Carl have helped teams across industries unlock serious value. And today, they're sharing what tools, mindset, and magic make great data consultants.
So welcome to the podcast guys.
Chris Love (02:21)
you. Thanks for having us.
Fiona (02:23)
So to kick us off today, I'd really love to hear a little bit more from both of you about your journey from corporate environments into consulting things like what prompted the change? So Chris, why don't you get us started?
Chris Love (02:39)
Okay, I'll get started. Yeah. So I was at a place called Experian Marketing Services I was there for 10 years. So I joined there as kind of junior data analyst. I applied for this SAS job, SAS programming, expecting to use my maths degree and do all this programming. And in the end, I was just kind of editing code that other people had written and stuff. And I found Alteryx during that journey. I dismissed it to start with. It wasn't really...
where I wanted my career to go with coding and everything else. Then we started finding it really useful. So I got into this thing, Alteryx, and started using it. Wind forward about eight years and I started managing a team. I progressed to managing about 10, 20 people.
I really enjoyed the leadership and the management, but I wasn't using the more technical skills. So I was blogging a little bit about Alteryx and enjoying that. But I've got this conflict between leadership and the doing. I remember speaking to my boss at the time and said, what would be perfect is a small consultancy where I can just lend my expertise. And then
I found the information lab. think it was people like Matt Francis, Craig Bloodworth on Twitter that I joined. started using Tableau because Alteryx added a version eight connector into the product. And I contacted the information lab and said, it's a job application of sorts I'm not really looking for a new job. We would love to chat to you and.
see what you're doing with Tableau. I'm starting to use it. Not really keen on moving anywhere, but you should see this product called Alteryx that I'm using. They happen to be using Alteryx and had just signed the partnership and we're looking for someone's train. So that was a real almost perfect match that suited both parties. And I ended up just going to meet Tom who founded the company and it just worked out.
I made the leap across and did some consultancy and the rest morphed into sales and everything else.
Sarah (04:40)
Love it when something just falls into place like that, right?
Chris Love (04:43)
I know it was so serendipitous. You couldn't have planned it better.
Fiona (04:49)
Mm, Kismet, definitely. And what about you, Carl?
Carl Allchin (04:54)
I probably shouldn't be here is how I always feel, especially when hearing Chris and his background. So I was a history and politics student. I wanted to keep my options open and just give myself a few choices and chances in life to see what comes up and what happens. I ended up joining Aviva so a life insurance company, who really allowed me just to go and explore things as a graduate. I was very lucky having some great mentors.
both at the head of and director level. And I stumbled into data. I saw it being used. I saw how much people didn't want to touch it with a barge pole, but it seemed super, super useful. So using that little bit of flexibility, I gravitated towards it. And since then, data has become pretty useful. It's kind of grown and I've kind of hung on to its coattails in so many ways and how it's regarded and just rolled with it. That ultimately led me to kind of working at Barclays where
There was myself, Peter Gilks and Lee Mooney, known as the Data Ninjas. We had access to Tableau. were some of the first Tableau users in Barclays and certainly in big corporates. There's really us and Tesco, a supermarket in the UK who were the early users of Tableau. So we were in a very small market in so many ways of just grabbing Tableau's focus too. And the more that we tried doing stuff, the more Tableau reciprocates with it.
But it got to the point where Lee went and off and started football analytics for this small football club called Man City. They've done all right since, one of a few things. Pete went back to the US because his wife was a New Yorker. So went and had a great career that way. And I was left holding the baby, basically. I just really wanted to be good. I wanted to be as good as I could be. the team that we had in supporting Pete, Lee and I were this lot called the Information Lab.
So I wanted to be as good as these guys. And this is when Chris and others were joining too, but had the likes of Craig Bloodworth and I could just reel off names from people in the early Tableau community. They were all there. it was my chance to dive in, having access to these folks at conference, managed to convince Tom to give me a job. And that's how I jumped over, more just wanting to be as good as I could be.
to see if I could kind of like hold onto the coattails then of these absolute gurus and try to survive and just explore life outside financial services after doing lots of banking and insurance work up to that point.
Sarah (07:20)
Nice, and I'm just curious as to what versions of Tableau you both started in.
Fiona (07:21)
Mmm.
Chris Love (07:28)
You were before me, weren't you, Carl? I think I was like version 8.1 or something like that.
Carl Allchin (07:33)
Yeah, I'm about to do a horrendous moment, which I don't normally admit to on camera. I downloaded 5.2 of Tableau Public. So really, really early days when I was actually at the Insurer, I tried to get them in or tried to get Tableau in and was basically told by IT, no, you can't have any different software from what we're using. So I actually brought in two grads from a local university to try and build me a data visualization tool to use in my reporting team. And they did a good job, actually. They created something pretty cool.
especially just over a summer. But yeah, because of Aviva was saying, no, that's pretty much one of my main reasons why I went to Barclays was it seemed to be like the right thing to do. So try and and back myself into a place where I was getting more support and more access to go and try different things ultimately.
Sarah (08:18)
And that would have been pre dashboards. β
Carl Allchin (08:21)
Dashboards were in, there were certainly lots of stuff that weren't. But really my everyday usage came when I kind of really started the Barclays role and joined Pete's team. That was version 7. But we were probably six months away from version 8. So yeah, Pete, Lee and I got to do lots of the early road shows for Tableau showing what you could do in version 8, even though we couldn't use it in the bank.
So we're doing all this cool stuff at home, which we then couldn't actually utilize in terms of functionality in the day job for about another year afterwards, because corporates and packaging software can be really frustrating.
Sarah (08:57)
Yeah.
Fiona (08:58)
Yes, especially banking. Definitely me too.
Sarah (09:01)
Been there.
Carl Allchin (09:04)
to play outside of work and kind of almost know what's coming. I think that's what I've always loved with that opportunity. So I could actually look really good in the day job because I've been doing it for a year before it suddenly appeared in the day job when everyone else is like, oh, this is really cool. I'm like, yep, yep, brand new. I'm picking this up really quickly when I'd actually made loads of mistakes in the background.
If you see my tablet public profile, scroll down to the bottom, you'll still see loads of those mistakes. I leave them up there intentionally. But yeah, that willingness to explore and play around my own time definitely made huge benefits.
Fiona (09:38)
One of the things that I really love about both of your stories, you've both gone chasing something that you really wanted or that you really believed in. So both of you have approached Tom about, "I'd like this opportunity to be able to come and work with you." And now you've obviously worked with him for years as well. Really interesting to see that. Do you notice the same thing in...
other consultants approaching your business.
Carl Allchin (10:05)
I think it was a bit different because in the early days of Tableau, there was kind of like a set group of weirdos who would happily sit at home in their evenings. And this is across the world. This isn't just in the UK, but in the UK, there was a small pool of us. And I always remember one of our team meetings that we'd kind of have quarterly catch-ups because we all worked remotely. We didn't even have an office at that point. And we would catch up and we're like, right, the company's growing. Who else is there to hire? And we literally could name one other person.
in our community that basically wasn't working for the information lab at that time. So yeah, they approached us a bit, we approached them. It's kind of weird that we were just in that world where of course we would all work together. So is it different now? Massively. It's changed hugely and thankfully the talent pool has grown much more diverse and wide and loads of people at loads of different levels of experience and expertise with this stuff. But yeah, it's...
Fiona (10:39)
my goodness.
Sarah (10:57)
It must have been quite
fun being like super niche at the time and all having such a shared passion.
Carl Allchin (11:05)
Yeah, and getting recognized for it as well. I think I'd spent so much time with information is beautiful and things like that from David McCandless sat on the coffee table, proverbially. Wasn't actually, literally sat on the coffee table, but occasionally on my desk. And people come over and talk to you about it and think there was this really cool thing, but it was never acknowledged as a job where all of a sudden there's this company that's doing pretty well. It's growing off the back of it. You're getting to bounce ideas around with people.
where it's not just going, that's cool. And that's kind of the end of the conversation. It's like, have you tried this or looked at this, there's some really cool ways to do this. that really, it landed not just benefit in terms of improving what we were doing, but I think it was also us starting to challenge each other in a really nice way of almost like the intellectual challenge that we'd all come to this company, I think, because we wanted to get really good at this stuff.
And that all just kind of kept us ticking up and become this really virtuous circle, which created a really great foundation where it's been super easy for both of us now to stay at the Information Lab for over 10 years.
Chris Love (12:13)
I think we were also riding that wave of customers as well. Those early adopters or certainly mid-level adopters of Tableau at the time all came to Tableau because there was this crest of excitement. when you're working with people like that, it's relatively easy to show off and do good things. you've got this great group of people at the Information Lab supporting us who
could help that if I didn't know the answer, could just call on Carl or Craig or Tim or someone like that just to answer the question. it was a really fun time to have that. The market's now changed, people's expectations probably through its own success to have those now set the benchmark and people expect something quite different and the late adopters or the laggards are...
now just come along and have a very different expectation of what a BI tool can do for better or worse.
Sarah (13:06)
I've been using Tableau for 10 years now and the evolution has been incredible in terms of what was an acceptable dashboard visualization 10 years ago to what is acceptable now. Like you look at some of the Tableau viz of the days and they're phenomenal. Whereas before it was like what's going on here, you know, but now it's incredible what's what's being produced.
Chris Love (13:16)
Yes.
It's really interesting though, you know, just reflecting on that. I don't think what's changed at customers has changed massively. Seeing a good number of customers that we do, I think the benchmark for what's used in the real world hasn't really changed. I think what we see in the community is massively accelerated. But aside from a few people doing amazing things and building some really beautiful dashboards internally, I think
because people are still learning and still adopting, I don't think the quality is necessarily risen across the board.
Sarah (14:08)
I was at a client just last week and they have like a chart of the year and they showed it to me and I was just like, well, like, yeah, we can better that for sure.
Carl Allchin (14:08)
I thought I said...
Chris Love (14:21)
Thank
You
Carl Allchin (14:24)
I think the thing that has changed is the buzz around the technology and people seeing this stuff. I think it's become a lot more normalized. I know lots of what we were trying to do in the early days at Barclays, we were kind of told no in a way because what you're going to show a customer that you have that data on them, where that kind of stuff is now sitting at the heart of lots of the FinTech applications that are challenging those banks, that there's that usage of data, that expectation of what you could do with data. And also in the early days, this stuff almost was like magic.
We had a massive 15 inch touchscreen that sat at the end of our desks and one morning it was gone. So basically the CEO had come down and saw some of the tableau stuff that we had up on it and it had a little computer attached to it and literally unplugged it, taken into the lift and took it back up to his office because he wanted to explore stuff. So we actually had to go and buy another one to go and take it back. you know, that just, I don't think that would happen today in the same way. I don't think we were walking on water with what we were doing.
I just think there was a lot more buzz that this is new and this is different and this is useful. Where, to Chris's point, I agree that the skillset hasn't raced on across the board, but I think more people are used to seeing more good quality data analytics in this space. So it becomes less special in a way. And I think that's what's changing the market somewhat, as well as Tableau's maturation and growth of other tools around it.
Fiona (15:47)
Mm, I think that's a really good point around what else is in the ecosystem as well and surrounding it. I mean, with the advent of Gen AI and GPT, Claude, everything at your fingertips now, it's so much easier to find information, learn, evolve and go through it. So, you know, we're just at this.
point in time where things are so much faster for us to actually do, I do think it is a little bit different.
Carl Allchin (16:23)
Question to you both, what have your guests been saying in this space? Because I definitely have thoughts about AI and where it fits into and how it's kind of changing our roles as analysts. But what's a quick summary of what your guests have been saying about what they're experiencing on the ground?
Fiona (16:44)
There's a lot of money that's out there, a lot of budget that's available for doing development and AI. Everyone seems like they've got some budget, but a little tentative and knowing where to spend it, how to spend it. We find that there's a lot of AI anxiety, especially with developers.
So people are really worried about their jobs and where their jobs are going. So a Tableau analyst, for instance, what is my job going to look like in five years time, given when you look at some of the advancements that they're making on product, whilst they might not be the most beautiful yet, they're certainly changing quite quickly and evolving. If you think back,
a couple of years when we had Ask Data, for instance, and to now what's happening with Pulse Insights, soon with Pulse Discovery, really digging into the whys behind things. That's a real shift in how analysts will be working and operating, going from taking requirements and delivering everything, perhaps more into a translator type space.
Sarah (18:04)
The other thing I'm noticing is it's that shiny ball again. we've seen it with big data. We've seen it with cloud, all the execs like we've got to get on this train without really thinking how it's going to affect them and improve what they're doing. And I think there's some, some rationale that needs to happen in that space as well.
Chris Love (18:26)
Mm.
Fiona (18:32)
but some amazing things that we're hearing people doing. Not that I really want to share any specific use cases because obviously people share them within confidence, but the way that people are piping data into large language models and other modeling tools as well to really understand
how they should evolve their processes or what changes they should be making. It's actually mind blowing.
Sarah (19:02)
We are seeing the odd rogue person in the organization that's just really leaned into it and become a subject matter expert in AI and is really kicking it out of the park.
Chris Love (19:16)
There are some massive opportunities for individuals to do that and really grab the bull by the horns and lead on this, probably as much as there was 10 years ago with Tableau and producing the BI and thing. So I think it's a really exciting time for individuals to make a name for themselves.
Sarah (19:35)
in my mind, they're almost taking some quite big risks because they're kind of having to step out of maybe what management is expecting and kind of going down a rabbit hole and hoping that they'll come out the other end and impress the leadership.
Chris Love (19:45)
Mm.
ask for forgiveness, not permission. In some instances, you've got to be really careful with that, particularly in the AI world, could cost you your job depending on what industry you're in. But the risks around AI are massively over exaggerated in a lot of cases. And you've just got to work out what risks you can take, I think, as an analyst and dare to push things with your business.
Sarah (20:17)
and the thing that surprises me at the moment, and maybe it's to come, is I'm sure there's been some really big fuck ups.
Carl Allchin (20:28)
Yes.
Chris Love (20:29)
100
percent.
Sarah (20:33)
Yeah, because you know, AI loves to hallucinate with conviction.
Fiona (20:33)
Absolutely.
Carl Allchin (20:40)
And I think that is the world where the analyst is going to. This is a topic that keeps coming up in mine and Chris's conversations, especially as we dig into data engineering a little bit more and look at analytics versus engineering as such. So as Fi mentions, in so many ways, it's that flow of these additional data sources that we can now get hold of is so powerful. You've still got to know what you're doing with them though. So all of the skillset, as much as we say that...
job market's changing and yes there's lot more excitement around AI and rightly so it's the new shiny toy you've still absolutely got to understand what's going on behind the scenes more so that data education now is wanted or needed more than ever but at the same time that's not the end point for once is then going further into understanding generative AI to work out how to use it and to really leverage it and make sure it's right to your points Sarah because otherwise yeah
people are going to make some pretty costly mistakes and that's not going do anybody any favors.
Sarah (21:37)
Yeah, exactly.
Fiona (21:37)
I mean, I
remember I remember when I started to learn to code which was in SAS as well, Chris. So same SAS then Alteryx I remember going into Telecom New Zealand. This was many, many years ago and the way that the guy taught me how to code on the tools was, why don't you look up your own phone number, Fi? And there's your birth date and there's all the calls that are
originating from this number, all these external numbers to your number and who you're calling as well. And thinking about that, in today's sense, the amount of data privacy issues that would be out there. I mean, I actually recall credit card information even being available in some of the databases that I had access to back in the past as well. And so, you know, when you think about that, and where we got to, let's say,
four years ago even in terms of data privacy and how we are looking after people's data. Now with how data can be used at such a wide scale and so prolifically and so quickly as well to be doing these decisioning, prescriptive analytics, which are really where are we turning next? Agentic AI, which is actually taking the action on behalf of you as well.
Data ethics that come to light are just astounding. And I actually saw a really good post today on LinkedIn. I'm gonna post up something, a PDF, and the show notes, which was from Charles Sturt University. It's a secure framework for GenAI use that they use for their staff. And I just thought, I'll throw that out there. It's always good to take a look at what people are doing in academia.
but also thinking about how you can apply that internally in your own organization, whether that's a consultancy or a corporate organization as well.
Sarah (23:36)
And it's good to see that that stuff is starting to come out, right? Because data governance and AI feels like a massive missing piece right now.
Fiona (23:48)
Mm-hmm.
Carl Allchin (23:48)
Yeah, I
saw a beautiful thing from, I think it was MasterCard as the kind of payments provider. They are now allowing cards to be written in the name of the agentic AI. So it still ties back to your own accounts, but basically the agent can now spend your money. And that moment of going from secure banking, where we're just literally trying to understand that every single time where fraud's happening and how you can control that.
to we're not even touching it anymore. It is mind blowing. that sent all kinds of shivers down my spine of where is this going and how is this gonna be ultimately controlled and levels of risk and the modeling around it. And then my head got into the data space and that's why I love the consultancy piece of this of diving into these problems to try and work out how the hell would I solve it? And I think the answer is on that one. I think I need Chris.
Fiona (24:38)
Okay guys, β why don't we get into some more questions. good chat, so I love it. Really, really fascinating. Carl, you've helped hundreds of trainees become excellent consultants. What's the first thing that you tell them to do when approaching a completely new client or data problem?
Carl Allchin (25:02)
Yeah, it's an interesting one because I think what's fun for those who don't know, I run a thing called the Data School which was started by Andy Kriebel and the information lab together to go and train, people up to be that next generation of great analysts is the idea. And it's really interesting the way that we get people to apply is they just build something in Tableau public. So showing their visual design skills, their analytical skills, and then they come and talk to us. So we sense check that they are human.
And that poses that really interesting question of when I get them on day one, who have I got in front of me first of all? But then what do we need to do to kind of get them ready for this weird world where we work with every different industry at all different stages of kind of data, literacy levels and, and sophistications. And I think the thing that always comes back to me is that piece of getting them really comfortable to ask any question. And it takes a hell of a lot of bravery to do that.
if you're asking a question, it's because it needs to be asked. If you don't listen to the answer and you ask it five more times, then that becomes what is a dumb question. But actually, if you're going to go and ask absolutely anything because you're curious about it, you don't quite understand it, you need it for the context to help make better decisions, that is absolutely the right thing to do. And I think that's one of the things that I've seen in the corporate world of people run away from that.
because that's going to be held against them for their reviews at the end of the year or Janine asks very, very basic questions and she's always asking stuff. Whereas actually, as life as a consultant, ask more, ask more. We're not supposed to be the gurus in the subject matters that we deal with. We're just bringing data to the party and making it super accessible to go and support those decisions. So for you to be able to work that out, you've got to ask a boatload of questions. So really it's, it's actually kind of
talking to the team in my first hour with them is saying, I want you to feel really comfortable going and asking our founder, Tom Brown, a question as much as the person sat next to you as they're on their first day too. And if you feel comfortable with that, there's nothing stopping you doing exactly the same at clients. And that's what will make you successful in the long run.
Sarah (27:13)
And I think it's really important to ask questions, which is great, but a lot of it also comes to listening to the answers.
Carl Allchin (27:22)
Yeah, that's it's something that we always ask in interviews of like, what are the great consulting skills? And everybody says listening. It's then really interesting to see who does in the day job of are they really engaged with what someone's saying? And again, being super nice to Chris, cause he is amazing. But Chris watched some DS presentations in the early days of the data school and I turned to him, I was like, what did you think? I thought it was a good presentation. And he was like, shouldn't have been a presentation. Should have been a conversation.
Chris Love (27:31)
Thank
Carl Allchin (27:50)
And that's one thing that's always really stuck with me that we often fall into these kind of formal roles that we would have expected to clients of, you asked me for a presentation. So I'm going to present this to you, which often becomes a tell where what Chris has kind of really astutely found was the better way to get a better reaction from that client and, and to really go and push on the subject was to turn it to a conversation. Cause if we're talking about something, I want to get feedback from you. It's going to be in the moment. We're then in this together to go and build it.
And that becomes really attentitive And I think it's that growth and that ability to ask those questions, listen to those answers, but also go and ask questions back to them to get them to talk and get them listening as well. I think that's a really neat skill to be able to do and where somebody really masters that. We can make some really interesting improvements and quite often what gets asked for in terms of our roles isn't what somebody actually needs.
So to go and tease that out, those conversations become really important. So yeah, somebody can be conversational. I think that's a really neat skill to have and what makes for really some of the best consultants.
Sarah (28:56)
Yeah. And I prefer like, when you've got a presentation that almost gets derailed because, you know, the client is talking more than you are. I think that that's when you really get the juice, right?
Chris Love (29:11)
Yeah, I've seen some big mistakes there where people have carried on with presentations, particularly in the sales environment, because I moved from the consulting world into more of a sales role. I'm never going to be a real salesperson in the same way that some of my colleagues are. But I've sat on presentations and watched them delivered where people have, despite people trying to interrupt and have a conversation, they stick to
I've got a message to deliver, I'm going to stick to it. And it really derails the conversation and you're just all trying to get through the slides then. What's the point? know, the slides are there to jump around them, help, you know, go off in a different direction, ask questions, be prepared to bring things up, know, Tableau is this great tool. We can even dive into it live and answer questions as things happen. That's mine, the tool exists.
Fiona (29:40)
you
Chris Love (30:03)
And I don't think people should be afraid of that scary. You've got to do it a few times and you've got to be prepared to make mistakes. β but yeah, it's, it's a great skill to have if you can, if you can do that.
Sarah (30:14)
because I hear you've got an actual framework. So the big list of questions. So how do you structure? Yeah.
Chris Love (30:21)
I did blog about this. Yeah,
I blogged about this because it was something that I started doing just to almost prepare for conversations, going into new places, just write down as many questions as I could think of around this thing. And it just gets you into that frame of mind of I'm there to ask questions. And if you send it in advance and then use that as a framework to start talking about things.
then it just opens things up and opens your mind up to what you should be knowing. think, know, to Carl's point earlier, think people are afraid to, when they hear an acronym they don't understand. As a young consultant, you're like, everyone else knows that and I don't, I'm going to keep quiet and I'll look that up at the end. And you lose a little bit of the train. But just writing down all the questions and...
thinking about all the stages of a development. What's the plan to get this live? Who are the stakeholders that are involved? Who should I be speaking to? What frameworks have you got to do this already? Et cetera, et cetera, et cetera. You don't have to answer all of them in one go, but it just helps guide the conversation and gets you used to that question mindset.
Fiona (31:39)
I was really hoping you would share your big list of questions.
Chris Love (31:43)
I know it's really difficult. It's really difficult because they're always β of in a moment about a specific project because I tried to write up a blog post and trying to pull across different ones I've got was so difficult. Maybe I need to find a generic set that I can share. But yeah, it's really difficult because they've deliberately got to go into a lot of detail around what you're doing, why you're doing it.
just to get that out.
Sarah (32:11)
And I think just playing the devil's advocate, when you ask a lot of questions, you can get a lot of detail. how do you pull it back and stop it becoming overly complicated as well?
Chris Love (32:24)
I think you can focus on different areas and break it down for the first thing. You don't have to do everything in one meeting. Let's focus on these questions and get this stuff covered. Take it offline. A lot of questions just need a bit of thinking about and someone can go away and just answer your questions on a bit of paper. Then you can pick up the answers in a meeting. And that's sometimes why I send it in advance because some people are just a lot happier just to say,
Have you got an established project framework, know, folder structure that you build something in already? Yeah, it's here. We've got the templates in SharePoint. Great. That doesn't need a big question on a meeting. The other thing that I've really started doing, is recording meetings. If you've got permission from whoever you're having the meeting with, just transcribe it. And if you do get into detail, you can...
it's just really easy to go back to and learn from and use in future. And if you are in an environment where you can use AI to summarize those things, you're building up a really good, RAG (Retrieval-Augmented Generation) database of content that you can just go and use in things like Notebook ML, ask questions of in the future. Things that you've left behind in three meetings ago, what were we doing there? Okay, it will tell you.
It turns you into little bit of a superhero sometimes where everyone's floundering over what you discussed three times ago. You go into the AI and quickly ask it and everyone's kind of, wow, he's got a great memory. It's just a retrieval mechanism using AI.
Fiona (33:56)
you're not a menopausal woman then, if you've got a great memory.
Chris Love (34:00)
No, I've got a terrible memory and I'm also rubbish at taking notes because I really want to engage with questions and the longer you're there writing notes, you never have time to go back to them. So AI has just solved that for me. It's fantastic.
Carl Allchin (34:15)
I think one of the things we teach at the data school to not take away from Chris's big list of questions, because that is absolutely what we prime some folks with as well. But it's more around just almost getting into the five whys and finding that right level when you're starting to dig into the requirements to keep questioning why. Don't ask it like a toddler, because we've all had those moments with toddlers where it's just like, just because.
That's not why you're trying to get to with a stakeholder. That's not going to get you the love as Chris mentions. But having that ability just to kind of dig into something with somebody to find that real purpose of what you're trying to do. That is the piece where every single first kickoff, let's have that as a conversation. Because once you've really helped somebody else work out what is it that we're really trying to solve, life becomes so much easier to then wrap your head around the solution and
Chris Love (34:40)
you
Carl Allchin (35:10)
That is where, to Chris's point, being really prepared, thinking about what some of those might be. So you're not suddenly taken aback by that. You've done a bit of research on the industry. First of all, you know where a company is at. So are they suddenly trying to pull back spending if their finances are struggling or sales are struggling a little bit? Are the wider challenges in the industry you need to be aware of? As somebody who had a massive data leak, strangely, they're not going to want to go and open up loads of APIs, et cetera.
So it's that basic research, but the five why's becomes really powerful to just go and find that right level of understanding of somebody else's situation so you can help them solve it.
Sarah (35:47)
Yeah, I really like that.
Chris Love (35:48)
I find the five why's really cliche. I'm going to be controversial. just, I just don't like it. It just feels a bit like everyone talks about it and I've never met anyone who does it well. Maybe that's the nuance of, of what people should take away from it. Not actually asking why five times
You know, you're asking because you want to really dig into the question and find out where the value is coming from for that individual. That's what we need to get to. Not the framework doesn't really matter.
Carl Allchin (36:17)
I'd like to point out to Chris, don't ask why five times because yes, that's bloody annoying. That's not doing the toddler thing. I've used it with you subtly many times and it's been so subtle you haven't even seen the benefit of it. There we go. So yeah, I think that natural curiosity though to Chris's point is something that's huge. Have you wanting to understand and generally solve something for somebody which without that understanding you can't do and also that curiosity to
Chris Love (36:22)
Yes. That like, yeah, I know.
Carl Allchin (36:47)
look at new ideas, be brave with it as well. To come up with a great solution is really where we're getting to with this. That without that great understanding, you're never going to find the right solution for somebody. But also without that curiosity of what's going on with tools, what's those new features, you're not going to be able to find a really great optimized solution for them either. So that's really where we're going with this, which is don't be afraid to pull on those threads and just keep pulling.
and really make that understanding. Cause I think that ultimately is consultancy. If you can do that piece, then everything else opens up really nicely.
Sarah (37:24)
And I think when you do ask the five whys, it's in that kind of listening. You can't just go, but why? It's like, string it together and go, okay, that piece and hone in and then ask, like you said, Carl, in a way that even Chris wouldn't recognize what you're doing.
Fiona (37:43)
Why do I feel like this is gonna come back around to me and suddenly Sarah is gonna be sneaking in these little why questions?
Sarah (37:52)
Why do you think
I did the human centered design, know, certification?
Fiona (37:55)
β yeah.
You were leaning in and I like the controversy and the tension points there because I think that's where things really get interesting because it's easy to get confused about, I ask them why five times, you know, and having that nuance in there. I'm sure that both of you have seen some data disasters over the years and we always like learning from problems that occurred and
what we would do differently the next time or how we got out of it. Do you have any data disasters that you could share? Obviously, you don't need to talk about specifics and clients or even previous employers, but it's always good to learn from something that happened.
Chris Love (38:48)
My favourite one that I like to talk about goes back to the early part of my career and rushing to get things out to people. You're often under a lot of pressure to deliver to colleagues, senior colleagues, to account managers, other people.
I just remember a few times we were really under pressure and people were saying, you've got to get out. You're there till eight o'clock. You send it out and the data is wrong. And there's nothing worse than that. It just undermined so much confidence in what we were trying to do. We then spent the next day trying to undo it. This was vehicle owner tracking for cars trying to.
to match two things together and send them out to car dealers. And find who owned the car now so that you could sell them new cars in the future. And as soon as that data's wrong and the end client gets it sent out, you get heat from the end client, get heat from the account manager who was delivering it. You look like an idiot. And you just need to take a step back sometimes. And I've learned so much about.
just stepping back and thinking in the moment and not reacting to pressure. It's really easy to react to pressure and get things wrong. And the biggest mistakes I've made throughout my career are always when I've felt lots and lots of pressure to do something and not had time to think about actually what are we doing? What is the right way to do things? So I always try and step back and do that properly.
Sarah (40:21)
And I think leadership likes to put like these almost unnecessary deadlines and into projects that that do cause this kind of chaos. So I'd like to see what's your thoughts on pushing back on some of those and making sure it's done properly and not just with urgency.
Chris Love (40:43)
It's really hard to push back against those, whether they're arbitrary or not. unless you're actually saving lives within healthcare or something like that, data is a discipline where it feels important in the moment to get things out. But we just need to remind ourselves that, most of the time we're not saving lives. Yes, there's maybe some revenue, some reputational things. But when I have pushed back,
things have never fallen apart like you expected them to. You only need to do it a couple of times to realize sometimes if you push enough pressure back, people will understand that you're trying to get it right and trying to do things properly. You just need to explain your reasons and explain what you're doing to meet the deadline. I think it's possible to do it.
But yeah, you're right. It's sometimes arbitrary, particularly If you're working directly for salespeople, their commission's on the line or they think their commission's on the line. They are trying to get clients and they don't understand data. And that's where most of the pressure comes from, in my experience. And you just need to be really firm with them and explain that you can't do it. And remember that ultimately you are peers. It sometimes feels like salespeople are.
in more of a position of power, but you are peers and as a data experts, we've got to to lean into our expertise and explain why things can't be done.
Sarah (42:08)
A favorite quote of mine working in my corporate banking world would be, is anyone going to die if this dashboard doesn't go out tonight?
Chris Love (42:17)
Yeah. Yeah.
Fiona (42:20)
Bluntly
Sarah (42:21)
Very blunt.
Chris Love (42:21)
Yeah.
I mean, it's when you get the answer you might, you know.
Carl Allchin (42:28)
Wow, Experian was tough.
I'm actually going to change my data disaster based on exactly what Chris said. So I once lost over 10 million customers for an organization talking about people happening. And it wasn't, they weren't real customers. It's just somebody had actually counted incorrectly and hadn't de-duplicated the numbers. And no one had ever thought about it because they weren't being told to rush an answer out. And someone was given the task rather than somebody given the challenge to actually determine what this was.
and I kind of always took the challenge more than anything else. And I probably got myself very close to being fired with this. I always said that if I didn't get fired two or three times in my career, I hadn't tried hard enough and I haven't been fired yet. So clearly I haven't tried hard enough. Or Tom's really, really forgiving of me either way. This wasn't the information lab, but it was a case of, nobody had actually really thought about what needed to be done.
But as I was articulating this clearly a nobody wants to hear that message that suddenly our customer numbers has disappeared through the floor. But also at the same time, it was really difficult articulating de duplication to a few people around the table who were the decision makers who weren't as data literate and were really struggling to understand that. So I think ultimately it comes back to kind of similar things as you've just discussed with.
Actually, if you can build trust with those leaders that you're working for, whether that's your direct manager, your manager's manager on the way up, when you have those moments of hang on a sec, I get that you're saying that you need this for then. However, if you give me X amount of time, what I'm actually going be able to go and deliver it back is all of this additional benefit. And that's going to help us really solve this thing. Cause that's what we're trying to get to. Right. And I think as I sit on the board of another company,
Sometimes, no, I just need that freaking answer. I just need that first little bit and you're right in the long term. Yes, that'd be lovely, but right now I need this. So I think there is reason why time pressure there. But if you've got that trust of that individual, sometimes they'll turn around and go, yep, no, take the extra couple of days you need, because we need to be right on this. And I think it's finding those moments and finding those people where you can have those conversations. That's really powerful in the workplace.
I've always naturally gravitated towards those people that I've worked hard for, but then have also reciprocated that trust that when I've turned around and gone, hang on a sec, they've really respected that. And there was a director at Barclays who was absolutely great at that. He was under huge amounts of pressure whenever new features were released. But if I told him that suddenly 250,000 people can log into something that they should be able to because of that change that's been made, he would sort it.
and he would give me that space to go and work out the impact of it before he then took action. And I think that's really brave of that individual as well. But it created a great environment to work in where you could have those conversations. So yeah, if you're lucky enough to have that, keep enjoying that. That's not everywhere. But also at the same time, try and work out how you can foster more of that.
Sarah (45:35)
It's amazing when you get in that position where the leader has got your back. And I think unfortunately they are a bit rare and not so common these days.
Carl Allchin (45:49)
It's pressure more generally isn't it? That, you know, life is all about seemingly quick wins, shareholder and stakeholder returns, because that's what we've got used to of everything short-termism. But really the companies that make huge amounts of value and leap forwards are startups that have then generated over two, three, four plus 10 years before they come into market and they look like they're brand new. But actually, if you look at their backstory, there's a huge amount of investment that's gone into them over time.
trust giving people that space to develop, but we only see those short term wins, those new things that come out rather than really understanding the backstory and the work that's gone into it. So yeah, keep highlighting those.
Fiona (46:30)
think it's a really interesting point that you've reflected on there, which is the pace at things, the speed at which things are moving right now. If you think back to, I'm gonna use Chris here, if you think back to the 90s, the end of the 90s for instance, and the way that things were operating there, perhaps we were,
Chris Love (46:52)
Mm-hmm.
Fiona (46:57)
just starting to get into the internet being available, maybe a few chat rooms around the place. we had mobile phones coming up at the end of the 90s as well. But the pace at which we were working then compared to the pace at which we work now, perhaps because information is ubiquitous and at our fingertips and so easy to get a lot.
of context behind it.
It's the expectation that's there now from everyone that we work with.
Chris Love (47:26)
Yes.
Fiona (47:27)
How do you feel about that?
Chris Love (47:29)
I agree. think the, the expectations around being able to get data out. working on mainframes back in the, late nineties, you'd spend three days just waiting for a time in the queue before you could extract your data. And now people are so used to getting instant answers through Google or everything else. Data work seems easy. And that does increase the pressure. Back then data work was a hard
discipline that you had real experts and professionals doing. And I think there's a danger that we've dumbed down the industry and people kind of have expectations that this is all easy and it's not. It's just a, different time. I don't think it's necessarily any harder. The challenges back then were just different, but the expectations perhaps are.
are a little bit higher
Sarah (48:19)
I sometimes dream about if I had all the technology available today, but I went back 10 years. Could you imagine how amazed everyone in the room would be?
Fiona (48:32)
get that time machine mate
Carl, do you have specific techniques in your consulting toolkit that could immediately make a difference to any analyst's life?
Carl Allchin (48:43)
Do I have a magic wand? Is the question. I really don't think so. think that the biggest thing people can challenge themselves is with that curiosity and bravery that I spoke about earlier. So being naturally curious about stuff and allowing yourself space to do that. if I'm honest, and this put down well post COVID, primarily because it's challenging working hours, but
Chris Love (48:46)
Thank you.
Carl Allchin (49:11)
I developed my curiosity and analytical skills outside of the workplace. it was working on, Pete and I were both massive basketball fans, so working on NBA data, because strangely the bank didn't want us to take customer details and stick it on Tableau public. Fair enough. Not that we tried just before that steamroll starts happening. So that was really where I refined those skills and just kind of explored the technology. So...
I think you've got to, if you haven't got a workplace that gives you that, I think you've got to go and develop it. And it's that ability to then go and harness that curiosity and intuition with technology is what you can then deliver back into the workplace and make those big strides forwards and also being brave enough to use it and really go and challenge what the norm is in your workplace that that influencing is tough, especially in junior roles when you're in the early days, but that's also.
a great place to go and do those challenges and kind of show what is possible. So if you've got a portfolio that you can show off, people can buy into that a lot more rather than proverbial pipe dream. So be curious, be brave, see where it takes you.
Sarah (50:23)
And I think, having things like Tableau Public and the Alteryx, spaces available to be able to publish and build portfolios was like some of the amazing things that built the Data Fam and other areas like that.
Chris Love (50:33)
Mm-hmm.
Carl Allchin (50:40)
one of my big regrets from both Tableau and Alteryx is that they've never given us public spaces to go and share workflows, to go and show the logic behind it and what we can actually do in that space. I obviously running prepping data for five years or so I just, really wish people were able to publish their flows of what they did and show those nuance and tricks instead of just a screenshot.
Chris Love (50:40)
Yeah.
Carl Allchin (51:02)
of the flow that was on the screen. Cause I think that would have also taken people's learning as to what's possible so much further and generated more of that within the workplace. So it wasn't just the visual cool chart. It was people then working on what we need now, which is much more rigged control data sources and getting them ready and making them easier and more accessible. And realistically for people is I'm kind of sad that both of the tools didn't really take up that mantle and create that space.
Chris Love (51:27)
No one's interested though, Carl. No one cares about the pipe. That's the problem. We do, but actually, people care about the funky visuals, the immediate in your eye, eye candy. No.
Fiona (51:39)
That's really interesting.
would say down under here, we are seeing such huge growth in the engineering space where everyone wants to be a data engineer. No one cares about being a data visualization developer. It's very much gone back to pipelines being closer to that foundation. And that's because of the closer that you are, that's where they're going to get closer into the modeling.
Chris Love (51:51)
Yeah.
Fiona (52:06)
and AI.
Chris Love (52:09)
100 % and we're seeing the same. just don't think it generates the same clickbait digging into the detail if you share a Python script that does something cool. it's not got the same draw factor. I always liken it to the F1 race where, you know, the people changing the tires. No one cares how that's done. They just care how fast, Verstappen drives or Hamilton drives.
Fiona (52:36)
What about the weekly challenges on the Alteryx community? I mean, there was a bit of a draw card in there and sharing different workflows.
Chris Love (52:45)
I think that problem solving, what people really like about those is the optimization of the solution and trying to do it in less tools or trying to solve the problem. I'll take that point. Yeah. The, problem solving aspect and what Carl does with prepping data, just being able to get there, I think is really key and it's a great learning tool. think it just doesn't have the.
thousands of views that the dashboards used to get on Tableau Public perhaps died down a little bit now.
Fiona (53:20)
And just stepping into the prep space a little bit, the acquisition of Informatica announced yesterday by Salesforce, how do you think that's going to impact the Tableau prep space?
Carl Allchin (53:34)
What a wonderful question. I think they're two very different audiences, if I'm honest. I've had Informatica at Barclays, and that was very much more of the IT professional setting up very controlled experiences now. I'm 10 years removed from that world, so if Informatica has changed, apologies to those folks. I need to get hands on a little bit more with it, and hopefully we'll get the opportunity to do that.
Where Tableau prep isn't even really aimed at that Alteryx audience in the same way. It was very much for people who were getting that ability to go and grab a dataset that they wouldn't have got otherwise from their IT setup or whatever else, merging that spreadsheet into what they were doing to enrich their analysis of what's there and start to learn the skills. And so I kind of feel like there's actually going to be no impact in so many ways because it's just two completely different use cases.
Where at the same time, I'm hoping it does shine a light on the value of this and that actually the more we can open up data preparation skills, that really unlocks data for so many people. It's why Jenny, Tom and I ran Prep and Data for so long and Jonathan joined me in creating it originally just to go and give people the skillset to do it. Cause we were stunned when we first threw out the Prep and Data challenges that people just generally couldn't do it. It was such a different skill set, even though they're creating amazing dashboards.
cleaning, reshaping and joining multiple datasets together was just alien to them, which is how I created a whole load of blog posts to help people do that, which turned into my first book. So I'm hoping it shines a light on the skillset and how valuable it is and how much is needed and that people keep opening that space to what Alteryx used to class as those artisan data workers to allow people to go and explore and play more because that's really where benefits going to come from.
Chris Love (55:16)
I think it's very much needed in the Salesforce ecosystem that more formal ETL tool. We've got data cloud now. We're looking at preparing data for agentic analytics as well as then in the wider Tableau cloud platform. You've got Tableau prep, which is a end user tool. What they're missing slightly is that more formal ETL tool to do to get data into data cloud. So I'm really interested to see how.
how everything joins up there and pulls data together.
Sarah (55:47)
Personally, I'm hoping it's just going to elevate the whole semantic layer, which for me has always been missing in Tableau. I know they've started down that journey. I just hope with the acquisition of Informatica that the semantic layer model is going to become more important.
Fiona (56:07)
we've discussed how we're very keen on the zero copy going on from piping together the different solutions or at least bringing the data and being able to see it in one place, But how Informatica plays in that space, given that it's more of that transformation tool, building out those solid pipelines will be really interesting to see.
The one thing that I've been reading about online, which is quite fascinating, is that there seems to be quite a heavy weight towards the governance side of why they acquired Informatica over and above what I typically think of Informatica for, which is much more of that ETL processing.
Chris Love (56:52)
I agree. people have often converted Informatica to Alteryx and I've always kind of talked about the fact that Informatica is very heavy on governance and all those pieces. And if you're using Alteryx for that, are missing out on quite a large aspect of being able to build that governance natively because it's a different tool. I think it does bring a lot more of that to the party compared to something like, them buying Alteryx and putting that in there.
Sarah (57:20)
It will be interesting to see how quickly they've managed to integrate as well.
Chris Love (57:25)
Yes, yes, and what happens to it?
Fiona (57:31)
We have just realized the time and the guys unfortunately have a hard stop. We could continue this conversation for a very, very long time, but Chris and Carl, honestly, thank you both so much. I've got so many amazing thoughts swimming around in my head and so many more questions as well. So maybe we can pick this up a second time around because it's been such a
Chris Love (57:50)
Why not?
Fiona (57:51)
Brilliant masterclass for me.
Sarah (57:55)
Exactly, you know...
Chris Love (57:55)
It's been great
fun.
It's great to be invited on these things. love chatting about this. I love talking about it and love hearing other people's thoughts and, debating things. And it's great to be invited and just have the conversation.
Carl Allchin (58:08)
It's why we started our pod was because we basically were having the conversations over a pint and got to the point where it's like, we're actually coming up with some quite good stuff here. Probably blowing our own trumpets and maybe too many pints at that point. But I think it's that kind of flow that actually there are useful bits that come out of this as we talk it through, as we discuss it, it stimulates new ideas. And yeah, it's the stuff that all of us are facing literally as we're on either sides of the globe from each other, right? That it's, we're all facing those similar challenges.
Fiona (58:38)
All right, everyone, make sure you follow Chris on Substack and on LinkedIn, as well as Carl. And don't forget to subscribe to their podcast as well. It's well worth a listen and I really love their guests and they share heaps of awesome insights. There's Carl's books, of course, Tableau Prep, Up and Running, Communicating with Data and Data Curious. And of course, the information lab and the data school have some wonderful content out there, so make sure that you
Google it, take a look at it, even chat GPT it and see what it takes to become one of their consultants as well.
Sarah (59:11)
And listeners, if you've enjoyed this conversation as much as we did, please take a second to like, subscribe and leave us a rating. It genuinely helps our podcast. We're all about sharing these valuable insights with as many people as we can.
Fiona (59:26)
So thanks again Chris and Carl and thank you all for listening. Stay tuned until the next unDUBBED .
Sarah (59:33)
Over and out. Bye.
