#110 Live from MeasureCamp London 2024: analytics in 2034

[00:00:00] Dan: Welcome to The Measure Pod. I am Daniel Perry-Reed and I’m here with Bhav. 

[00:00:05] Bhav: I am Director of Analytics and Experimentation at LeanConvert. 

[00:00:10] Dan: And I’m a Principal Analytics person at Measurelab and we are here to discuss live at MeasureCamp 2024 in London. We’re here to discuss, what does analytics look like in 10 years?

[00:00:34] Dan: So I’m going to start us off on the topic and anyone feel free to jump in. And especially I’m going to direct this towards you to begin with. Are we finally going to get over this need to collect a hundred percent more accurate data in 10 years time? Or do you think we’re still going to have the same goddamn conversations over and over again?

[00:00:50] Bhav: I think we’re going to have the same conversations over and over again. Welcome everyone to the room. Dan and I are recording this live. This is a podcast. Missed the opening. Almost weekly podcast now, right? Going into Dan’s question, so Dan’s question was really around, are we still going to feel the need to collect 100 percent of data in 10 years time?

[00:01:12] Bhav: Do we collect 100 now? Well, people think we do, right? Yeah, exactly. So we have to now respect consent, which is, ooh, that was 10 years ago. Yeah, yeah. I mean, I think, I think if we think about, you know, the current state and extrapolate it further, that will give us a good enough answer. Certainly, I’m going to, I’m going to go straight into the swear jar word of GA4.

[00:01:35] Bhav: GA4 right now, for those of you who use GA4, might or might not be aware that it’s actually not collecting 100 percent of your data. It’s actually using something called a, what’s the model called? Something plus plus, I forgot what it’s called. Hyperlog plus plus model 2. Hyperlog plus plus model 2. kind of model your performance of your data.

[00:01:55] Bhav: And it does that when you put high degrees of hard analysis. So it tries to estimate it. And I have been very fortunate recently to have actually come across, Hyperlog, log plus, plus, plus, plus, plus, in the wild. And it’s pretty good, right? It’s about 2. 5 percent difference to the reality when I compared it to BigQuery. And you start to realize, actually, You don’t ever need 100% of data. 

[00:02:21] Dan: You’re assuming BigQuery, right? You’re comparing it against bigQuery, right? So who says that’s the source of truth that you should be comparing it against? 

[00:02:26] Bhav: Okay, fine. It’s not a model, at least. That’s what I’m saying. It’s not model data. So will we still need 100%, you know, 100 10 years from now, the answer is no, and we never really needed it for what we want to do.

[00:02:38] Bhav: But if we start moving to like one to one levels of communication, personalization, then we’ll be able to swallow it. 

[00:02:45] Dan: Of course, but we know that, and we in this room know that, that we’ve never had accurate or clean or perfect data, but who’s still here is still having those conversations around Google Analytics doesn’t collect every visit and every page.

[00:02:56] Dan: People are still talking about that. They’ve got stakeholder. Yeah, yeah. And, there’s only a few of them. but it’s the thing with that is that we’re still having conversations now. And this is essentially, especially in this part of the world. This is something that technically came into effect in 2018, you know, with GDPR like we’re 2024 with six years after when this happened, we’re still talking about it.

[00:03:16] Dan: I was wondering if given another 10 years, are we still going to have people that are like, how can I trust the data if we don’t collect every single bit of everything? 

[00:03:23] Bhav: Yeah. And I think this is where we’ve fallen into the trap of thinking data first. One of my favorite questions, and I hate quoting billionaires, but Jeff Bezos asks this question to his team when they’re thinking about what future planning is going to be.

[00:03:37] Bhav: And he doesn’t ask what’s going to change, he asks what’s not going to change. So maybe my question to you Dan is, if we step out of the world of data collection and thinking like data nerds, What’s not going to change 10 years from now? 

[00:03:51] Dan: I don’t know. Does anyone have a view on that? What would be the same in 10 years time? What’s the kind of shit that you’re doing now that you’re still going to be doing in 10 years time, assuming we’re in the same role? 

[00:03:58] Bhav: I’ll give you a quick example of what Amazon do. So what Amazon don’t think is going to change, people still going to want fast delivery. They’re still going to want cheap products.

[00:04:06] Bhav: They’re still going to want quick turnarounds and good service. So from Amazon’s perspective, They’re not thinking like, Oh my God, advanced technology. They’re thinking the core basic needs of users are not going to change six months from now, 10 months from now, 10 years from now. So I think that’s how they think that.

[00:04:22] Dan: So I’d love to hear from. Yeah. From in the analytics space. Got Charles Hills. We’ve got a microphone. What’s not going to change? I

[00:04:27] Audience member: think data silos. Around the business, we’re going to have different tools collecting the same data for different purposes. And to what you just said about business users expecting to be 100 percent of the truth, they don’t understand what data silo they should be using for which purposes. And I know there’s a lot of products now talking about how you can join and create data, see that third party perspective, but I don’t think we’re ever going to get over that. Especially in large legacy organizations, those silos in different parts of the business. 

[00:04:58] Bhav: But again, you’re approaching this, so I’m going to challenge that one. If you’ve listened to the podcast, you know I’m the dick from the podcast. I ask, I ask the podcast. On the podcast. but that’s still thinking of, from a data person perspective, we’re still thinking of this problem as data people. And I think if we think about actually what is, Data’s role 10 years from now, it’s, it’s still going to be exactly the same using data to make better decisions, give customers more tailored experience and a personalized product range and ensure that they’re only seeing things that are relevant, blah, blah, blah, whatever that means.

[00:05:33] Bhav: But what’s a better decision? Well, this, you, you, I challenged Dan on this. He also might have better decisions. 

[00:05:40] Dan: Okay, fine. 

[00:05:43] Audience member: I think it was actually saying from a human point of view, we’re still going to organize ourselves in silos. And the data is flowing from the silos, but we’re still thinking in that kind of way. I’m part of the marketing team, I’m part of the HR team, I’m with the logistics, the data is vital. 

[00:06:00] Bhav: Ah, okay. My apologies, Charles. I missed something. This is the pitfalls of doing a live recording. do you want to take this one, Dan? 

[00:06:07] Dan: Yeah, well, it’s politics, right? Like, politics ain’t gonna change, and company structures. I mean, we’re still organizing teams, we still have job titles that we put weight in, right? Like, we still have, like, a ladder to climb, and we’ve still got teams to report into, and stuff like that.

[00:06:19] Dan: And actually, the bureaucracy and the politics of business is, I think, is It’s a system that kind of supports itself. Like if it didn’t exist, it’ll be fine. But also a lot of people wouldn’t have anything to do. so I think that there’s an element there, which is like, okay, me to get shit done in the marketing team, I can go and just do that.

[00:06:34] Dan: But as soon as I involve HR, finance, IT, then it slows down. I can’t do it. There’s governance. I have to have meeting after meeting. And actually by that point, the requirement would have changed. And I think there’s just an element of like, He’s maybe part of least resistance. I don’t think that will change.

[00:06:50] Dan: I don’t think the whole business will be like, you know, woke on data, you know, like they’re not going to understand that we’re not going to have like, Oh, yeah, sure. We’ll all just react really quickly. I mean, generated AI, especially in the last couple of years has changed. The scale of change has increased rapidly.

[00:07:03] Dan: So, like, the fact that, you know, it might still take us 6 to 12 months to implement the data warehouse by the end of that, we might, well, we don’t need a data warehouse joining all the data because the data is not there or, yeah, it’s in sparse spaces. But, but it’s not about that. It’s about, like, the purpose we’re trying to solve for is today.

[00:07:19] Dan: And actually, by the time we’ve got it, it might be different. And I just think that things change so quickly now. We need to, you know, Rather have quicker fixes than longer term. 

[00:07:28] Bhav: I know, I know we’re on an accelerated timeline. It took you all eight minutes to generate an AI. But you brought up GA4. Oh, I mean, I see. Yeah, so actually maybe you guys are right. So the question isn’t so much around the user side of things. Actually, how is the organizational structure going to change that’s going to affect the need for these silos? And I think maybe if we think about it, I hate having a dystopian view on this, but are some of the roles even going to exist?

[00:07:54] Bhav: I don’t know. Right. So are we going to find ourselves in a situation where the data is being collected, but the need to make a decision of that on that data no longer sits in the, in the team or a person, or maybe not in the way we think about it now. So actually, I know 20, 30, 35 will be actually does a certain team even exist?

[00:08:15] Bhav: Like for example, the role of the acquisition market, is that even going to exist? Are we going to end up in a world We’re, you know, going straight into automated bid management. Well, we are. Well, yeah. But I mean, even further to the point where it’s, it’s not like manually, because I think so many organizations are still manually changing bids and they’ve got Excel sheets and Windows sheets.

[00:08:33] Dan: Oh yeah. But that’s, I mean, God, that’s just keeping ourselves busy again, isn’t it? It’s just like, I refuse to acknowledge changes happening and the tools are changing and we keep doing it. But cuckooing a little bit of like how search marketing works, but. But essentially, like, things have changed so much, and we can’t really do the same as we used to do.

[00:08:49] Dan: But I want to bring it back to analytics, right? And whatever we mean by the term analytics. Like, what does analytics mean now is actually a question mark. But like, we all have a slightly different interpretation of what a role of an analytics person is. What would that analytics person do in 10 years time?

[00:09:04] Dan: What would be different? Like, I want to put it out there. Like, do you, who works in analytics right now? Okay, pretty much all of us, right? So, so what do you think you’ll be doing in 10 years time? Do you, do you think, oh, let’s get, sorry, sorry. Need the hashtag content, sorry. Hashtag content. 

[00:09:22] Audience member: so do you, so like, I’m a product analyst and like, I think 10 years ago. Thank you. Yeah. 10 years ago, I think when, I was looking for like data related jobs, it was very much like data analyst, data analyst. Do you think these analytics jobs are going to be more specialized? They get more narrower? Or do you think potentially the opposite way? where a typical, like, data analyst has to have a number of strings to their bow, at least you know a bit of That’s the swear jargon. do you need to do a bit of Do you need to do a bit of, like, a bit like what you do at LeanConvert? You do, like, you do lots of stuff there. But I suppose, yeah, are we going through 

[00:10:04] Audience member: more specialization? Or will the average data analyst need to do way more, like, data science, maybe, like, analytics, engineering, and more broader things?

[00:10:13] Audience member: Can I take 

[00:10:14] Bhav: this question? As a fellow product analyst, I don’t know how else to do it. I thought it was a problem. No, definitely not. That was, that was, that was all natural. so I feel it’s a really good question. we have seen certainly over the last few years, roles that were, that existed five years ago in analytics no longer exist.

[00:10:35] Bhav: And we continue seeing this evolution. I think five years ago, four years ago, whatever, data science was the sexiest job of the 21st century or something. I wrote a post about product analytics being the next sexiest job. I don’t think we’ll ever reach the heights of data science. but then we saw data sciences now Transition and split into multiple different areas.

[00:10:54] Bhav: You have like a metric, machine learning, you have all of these types of things. And I think the short answer to your question is yes, I don’t, and this is what I was kind of like going back to the point of like, do roles that exist now, you know, will they exist 10 years from now? And I don’t think they will.

[00:11:08] Bhav: I can’t see a world where the roles that we have now exist 10 years from now, which stuff, which begs to ask the question, what skills as an analyst, you need to stay on top of and should you be. specializing. If you choose to go down the route where you’re specializing, you’re kind of putting your eggs into, you know, all your eggs into one basket.

[00:11:26] Bhav: And I think a good, the core fundamental skill sets of a good analyst is adaptability, being able to look at a problem and go and find a creative solution for it. So I spend a lot of my time specializing in product analysis, but You throw me into a situation where I need to build an ML model, I will. You try to, you know, if I need to do some attribution modeling, I will.

[00:11:47] Bhav: If I need to do, you name it, I’ll find a way to do it. And I think that focus on, you know, too much of, you know, I’m a, Solutions engineer. I’m a GTM engineer. I’m an implementation specialist. I’m a marketing analyst. I think if you continue going down that path, you’re going to find that the roles are shrinking.

[00:12:07] Bhav: And actually the best way to potentially stay relevant is to continuously Broaden your skills, not to the point where, you know, you, you become the jack of all trades, the master of none, but definitely understanding like actually there is a need to flex your analytical skill sets. And I think analysts are in the best position to be able to do that because we’re so adaptable.

[00:12:25] Bhav: I think more so than probably any other part of the organization. I think the other parts of the organization are adaptable. Way too specialist. 

[00:12:33] Dan: You mentioned that, like, I know I brought up generative era before, but I think that’s going to shift the focus away from like the technical skills. You mentioned data science there.

[00:12:40] Dan: Like that, that, that job will still exist, but it will be like the Googles and the methods, right? Like they need that to maybe to kind of do the heavy, heavy lifting side. Yep. Yep. You know what we see even now, 10 years before this time, like what data science is now will become rudimentary, right? And I think we’ve seen that shift as we go on, like, even just doing like machine learning now is essentially a click of a button in a lot of platforms.

[00:13:00] Dan: And that used to be like, you know, hours and days worth of processing and people’s jobs dedicated to it. But it still is in some parts, right? And still some areas. But I think in the world that if you are looking into the world of like application analytics, like marketing products, whatever, like It’s just going to be there.

[00:13:15] Dan: And I think maybe the generative AI will become a point where it’s like, okay, well, it doesn’t matter what the coding language is, you know, I can utilize these tools. Now there’s still going to be specialists that go above and beyond, but. Would the average company need that? Maybe not. Like, maybe that’s something that you just need someone that knows how to use the tools and it’s the business application and all the, a lot of the talks that are going on today is around like analytics, going from a cost center to a profit center or proving the value of analytics.

[00:13:40] Dan: And I think that’s going to be the focus. It’s not going to be that the role it’s going to be like, cool. Can we justify five people, you know, in a, in a job using data when a lot of the time, and we’ve talked about this before, but like we, we aren’t as analytics or analysts. We aren’t the ones making the decisions.

[00:13:56] Dan: We provide the decision makers with data to support a decision or to, to influence a kind of a path. And so we can’t pull the trigger. And so that’s not going to change. 

[00:14:04] Bhav: Can I challenge you on one thing? I’m going to. All right. So, I’d love to hear from you guys. Dan said that nowadays you can click a button on machine learning, right?

[00:14:15] Bhav: Which to some extent, hands up. If I said to any one of you guys, go into organizations, click a button and set up machine learning, hands up. If you would be able to do that, as analysts, competently able to do it, or politically able to do it, or competently able to do it. 

[00:14:29] Dan: The lowest barrier of entry here. Pardon? 

[00:14:31] Audience member: And how long do we have? 

[00:14:33] Bhav: Take as much, give yourself two weeks. Give yourself two weeks, right? 

[00:14:37] Dan: Or a month, right? Alright, a month. Who can set up any machine learning model in any platform they want within a month? 

[00:14:43] Audience member: Do I have to delegate it? 

[00:14:44] Bhav: No, you’re not allowed to delegate it. 

[00:14:45] Dan: You’ve got to figure it out for yourself. 

[00:14:47] Audience member: Can I ask ChatGPT how to do it? 

[00:14:48] Bhav: You can ask ChatGPT there was no restrictions in the way that I was approaching this, right? You can ask whoever you want. You can use whatever technology you want. The challenge with this, it’s kind of like, you don’t have to spend money, come on, we’ll figure out how to do this, right?

[00:15:04] Bhav: But it’s, it’s still a, it’s still an uphill battle to understand, like, okay, what are the underlying mechanics of the machine learning model, right? Like, what data do you didn’t say I’d be a good model. 

[00:15:12] Bhav: Yeah. Okay. Data. Yeah. Data quality is going to be an issue. Right? So if we think about the fact that data quality is going to be an issue, hands up if you guys are hardcore SQL coders.

[00:15:25] Bhav: Okay. Only if like half the room, that was a safe assessment, right? Half the room. As an analyst, a prerequisite is not SQL. More and more and more understanding how to write a SQL code to be able to extract that data, connect it to some type of tool that will allow you to plug it in. Google Cloud Platform has machine learning models built into it.

[00:15:46] Bhav: Yeah. But fuck if I can navigate Google Cloud Platform, right? I have no idea how to use that platform. I can get in and out of BigQuery really quickly. I can just about give people access to the, you know, to be able to access BigQuery. But beyond that, it’s an absolute minefield. do I want to know? I think this is the question we have to ask ourselves is actually if the roles, again, going back to your point, is who’s going to be able to be, who’s going to press that machine learning button?

[00:16:12] Bhav: what else needs to happen around it. And I think this is, this is the hard, this is the thing that’s really hard to predict. And I think, you know, we just come from a talk which talked about, which was really interesting around, the pitfalls of attribution modeling and, and models and last click and DDAs.

[00:16:28] Bhav: And I really liked it. I, if you’re the speaker of that talk, I’m really sorry. I’m about to challenge you. You built a solution without thinking of the end user. It’s a really smart solution, but Not how many marketing people do you know are going to go and connect every single one of the data sources into this model then sit in there every day, teach it, train it, learn from it, adapt it, and pay for it, and pay for it.

[00:16:50] Bhav: And then when I, when I was looking at the comparison, I asked the question, I asked this, I asked this chap, how different is your model versus a DDA model versus a Lozgrip model versus a Buzzscript model, the answer is not that different. Yeah. So from a marketing perspective, like, you know, you’ve built a solution.

[00:17:08] Bhav: As an engineer, without thinking of the end users, because it’s a complicated platform. I think things like machine learning is There’s still going to be someone who needs to push the button. We just need to know how to push those buttons. should we go?

[00:17:19] Dan: If you set up a Google BigQuery export from Google Analytics 4, and you’re not really sure what to do with it, it’s your first time using the GCP, then give us a shout.

[00:17:30] Dan: We’re here to help. We’re GCP certified, GMP certified, and of course I can’t stop talking about Google Analytics 4. So we’re here to help you make the most out of that data and show you exactly how powerful it can be. The world of cloud databases in the Google cloud platform can be complex and overwhelming at times, but ultimately it’s there to serve a purpose and to make the most out of the data you have, you can reach out to us by scanning this QR code, or you can click the link in the show notes to find out more, or you can just head over to measurelab.co.Uk and find out a bit more of what we’re about.

[00:18:04] Audience member: yeah, I think being a bit apocalyptic, going like these jobs won’t exist anymore, I think they may have changed to the point where if you think about. Cloud migration. When I was working in the 90s, the DBA rule existed. Like, you didn’t get a server provisioned, it was hardware on premises at the time, and the old guy who knew how it all worked was typically an old guy.

[00:18:31] Audience member: Now, I quite happily run most of my clients, and I don’t have any kind of DBA type skills, but I can spin up a server, oh no, I need double the size. Double the size. Well, anymore, the DBA rule doesn’t exist anymore. for most places, until you need it to.

[00:18:51] Audience member: And at which point, when you do hit a problem that can’t be provisional, you can’t spend your APACs, somebody can come in and optimise the table, rerun the indexes, plan out the scheme that needs to be done so your data warehouse can actually work, rather than being a big mess of a data lake and actually weary.

[00:19:06] Audience member: Turns out, hang on, there used to be a DBA, but now they’re a cloud specialist, and effectively they are still doing what we would recognise as DBA type tasks, but we don’t call them a DBA anymore. So there’s no database administrator. That is, that is Push a button, but somebody who knows what it should look like still has to exist.

[00:19:25] Bhav: I think that this is it. And actually, I can’t remember where we started. I think it came up potentially on a pop up that we were, my memory fades me in my ripe old age. someone mentioned, we spent ages trying to connect data sources. We’re trying to build this single place where all of our data needs to go into.

[00:19:47] Bhav: Does it though? Does the marketing team or the product team need all of that in one place to be able to build a 360 customer view, blah, blah, blah. All of these things, do we need that? I think the answer is no. We spent, you love talking about, one of your favorite topics I know is, modularization And I love this as a topic.

[00:20:05] Bhav: And I think, I think you’re absolutely right. we, we have historically, we’ve come from these places where there’s been one platform that ruled them all, right? there was an Adobe suite or a Google suite or something like that. And actually more and more companies are like taking and shifting and putting pieces of the puzzle together, but it’s, and it’s great, but you don’t always need to then rebuild the puzzle.

[00:20:29] Bhav: You can have the pieces of the puzzle, without having to it. I think you can have like smaller puzzles instead of one big puzzle. So I think maybe 10 years from now, this concept of like having everything in one place for all of us to be able to go into. What’s the point? 

[00:20:41] Dan: It’s the data mesh. Yeah. Yeah. I mean, this is the new buzzword, isn’t it? like, or like the new things, like actually it’s all about having access to data. Then it’s about centralizing data. Then it’s actually, we don’t need it centralized. And then we can just connect it all.

[00:20:57] Audience member: I do hope we’re going to be talking a bit more about ethics. Because it’s one thing to sort of push buttons and produce a model of X, Y, and Z. But do we actually need to be doing that? Or what are the government actions to do that? And to say that we’re compliant legally is Sort of cat’s way out. You can say, yeah, you know, I’m doing this and legally we’re all good and that’s all fine.

[00:21:19] Audience member: But to say that it’s the right thing to do, it’s slightly harder. Bring everybody on board. And then to do that at global level where ethics may be, you know, be variance with different regions is, is a pretty hard task but you don’t hear an awful lot about it.

[00:21:35] I love this question because I put to bath a couple of days ago that this topic for this exact conversation should be. because we don’t, we don’t thank you for that. It’s such an important thing. I, I, I, I shout to anyone that will listen about like people build products and they create an MVP, but that’s not where you stop. Right? That’s the minimum viable product. And then you actually build the whole thing becoming legally compliant with anything.

[00:22:01] Dan: He’s an MVP. And it’s like, okay, well, I’m working with a, with a, with a marketing team right now and they’re questioning, like, what do we, what do we have to do to become compliant? I was like, well, what do you, what do you want to do to be good with your customers? What’s the ethical approach your company is putting on top of your, your customers?

[00:22:15] Dan: Like, I can tell you not to set a cookie without consent. Sure. But like, what do you want to do? Like, what’s your line in the sand? And I think it’s such an important part of that because. Who knows? I mean, the whole legal compliance is just going to be button clicks. It’s becoming easier and easier to deploy a cookie banner.

[00:22:31] Dan: It’s out of the box. Google consent mode, theme consent mode, Meta are doing the same thing. That will all consolidate and be automated at some point. But like, what’s that step above? And I think there’s some really interesting examples. I really like IKEA’s example of where they went above and beyond and they built privacy by design into their platform.

[00:22:46] Dan: And every single time they’re asking for a bit more consent and what you get for it. I like all that. Oh, are cookie banners not the way to go. 

[00:22:52] Bhav: I love the fact that we’re talking about privacy by design and most organizations can’t even do measurement by design. No. Yeah. So from a product analysis perspective, when we think about what good product development should look like, the code should be bug free, like a wall of code.

[00:23:08] Bhav: It should be solving the user problem. And the third most important part is it should be measurable by design, but product teams still not even doing that. So to get to like privacy by design, it’s just, it feels like the moon for a lot of companies. And. I guess until, until there is a legal requirement or there is a monumental shift in having a very ethical approach to data collection, I just don’t think we’re going to get much more.

[00:23:31] Bhav: And I hate this as an answer. We spoke about this before actually, the whole ethics around data collection and people need to challenge companies to be more ethical. So until. People caring enough enough about their own data. Companies are never gonna take that first step. They should. Yes, of course. Is it gonna be a differentiator from a, from a, from a commercial standpoint?

[00:23:53] Bhav: Probably not. You could potentially be the first one there and, you know, give yourself a bit of USB, but it’s never, unless there is a, you know, the people need to rise up and say, look, this is my data. How do you own it? And you do see it in some cases. But, and I think about GDPR as the biggest, like everyone waited for this wave of requests that come through around what data you’re collecting, data deletion.

[00:24:16] Bhav: And I remember I was working at Moo at the time, we were ready, like the entire team were like, guys, we’re ready to delete, we’ve got all of our processes set up, we’re ready to give people access to their own data. We had like 10 requests in the first week. And it got me thinking, like, who, like, why was this such a big deal?

[00:24:33] Bhav: Right? And of course, some companies got penalized. And fine, they’re not having proper laws in place, but beyond that, it’s just Actually, of course, I see it religiously. 

[00:24:41] Audience member: You don’t have people acting and trying to create problems. They haven’t actually got a problem with it. But they’re doing it just to troll the way out. Pay control. The United States, literally, that We’ve got a solicitor’s business. He’s finding people who have been breached with California law. And then just targeting them, but basically holding them to ransom. We’ve charged 15 million, we’ll have 75 grand now. 

[00:25:02] Dan: Who’s, who’s had, who’s had a conversation in the last couple of years about legal compliance with analytics? Okay. Everyone. Who’s had a conversation about ethical compliance in America? I want to say a third. A third of the people that put their hand up. I think that’s an interesting perspective, right? Like, we’re always clawing to become the minimum legal compliance as possible, but then, anyway, it was just interesting for my benefit. Who had the point? Yeah. Where is that? 

[00:25:30] Audience member: To that point around privacy, would you guys agree that, people say they care about privacy more than they actually do? Because you, you say a lot of talk about it, but when people see this cookie banner on websites, they just have to have, et cetera, et cetera, et cetera. They’re more than willing to give Facebook or Google or anyone all the data they want. 

[00:25:49] Bhav: You know what? You could replace privacy with anything that could be personalization. It could have been whatever. And the question is like, do people care about it? Sometimes they do. I don’t know how much we want, and I think there’s a difference like we care about because we’ve seen it right like I’ve been there man Like you weren’t there man.

[00:26:10] Bhav: Yeah, we know what it’s like. We’ve seen the like buckets and buckets of data we’re collecting Shit, we have a lot of data here, right? And of course, we’re ethical people. We’re not going to use it in unethical manner. 

[00:26:24] Dan: You know what would make us more ethical? Collect more data. Yeah, this is true. Go on, Charles. 

[00:26:28] Audience member: I don’t think it’s the people who don’t care about it. I think we’re in the 1 percent of people who understand it. Yeah. Most of the people clicking on your cookie banner have no clue what that means or what’s happening behind the scenes. 

[00:26:38] Bhav: Or if it’s even working. My wife doesn’t care. I asked her. We’ve spoken about this and she’s like, I don’t know. 

[00:26:43] Audience member: That was an interesting question about ad blockers and getting around that. And someone asked the question, isn’t this just futile? And I immediately started thinking of Black Hat SEO. And they basically answered me, well, you know, if people, if clients are coming up to us and asking for this, then who are we to say no? They want to give us the money, then, Are we going to turn down work?

[00:27:09] Dan: Maybe we should be saying no. I will be, yeah, absolutely say no, yeah. I’ve said no to working with companies before that didn’t respect a cookie banner before and they knew what they were doing and didn’t want to do it and I didn’t work with them. I’m not saying you should all do what I’m doing, but I’ve been there and maybe we are the 1 percent or maybe less than the 1% that can do that, that knows how it works, but they’ll find somebody who will.

[00:27:31] Dan: Of course they will, they’ll find someone, because there’s going to be someone on, you know, Upwork or something like that it for them. 

[00:27:37] Audience member: If you do black hat SEO and someone’s willing to pay for you, you’re probably not breaking the law. It is a difference. 

[00:27:45] Dan: But that’s it, that’s the legalities versus ethics, right? Yeah, in terms of like, who would be willing to take the money for that, like, once there’s a legal aspect involved, I think it does change. 

[00:27:52] Audience member: Yeah. So just because we’re talking about a continuous time and you mentioned the 1%, so where the 1 percent of people who knows what’s best for the business and for our customers, do you think that’s going to change in terms of like, who is like responsible for the data?

[00:28:07] Audience member: Because that is like, you know, it’s easier and easier to get access to that data. Are we still going to be the special 1 percent of the business? 

[00:28:13] Bhav: So yeah. I recently wrote a blog post about tech layoffs. I don’t know if anyone saw it. I did an analysis on the likelihood to be laid off, right? And I used engineering as a anchoring point. And as a product manager, you have two and a half times, as a user researcher, you have three and a half times. As an analyst, it was one to one, right? You were just as equally likely to be an engineer. And I think this is going to come down to skills, right? Because Who’s like, are we still going to be, I think the 1 percent is going to merge with another percent in another part of the organization.

[00:28:47] Bhav: And actually what we’ll have again, going back to kind of being able to do multiple things. Imagine you’re a product man. One of my biggest problems with product is that in that sort of like research design engineering, analytics is not even part of it, it’s like Holy Trinity. And what, and this is partially one of the reasons why I think product has been so impact, is because they fundamentally not been able to prove the value of what they’re doing.

[00:29:12] Bhav: Why? Because data isn’t at the heart of what they do. So you’ve got, you’ve got a problem here where it’s like, okay, they’re not a, they’re not hiring people because they are analysts as because they don’t see it as a core part of what they do. B, they don’t have the skill sets to do it themselves. So one of two things can happen.

[00:29:28] Bhav: Analysts, CROs, who are people who are analytically minded, are going to transition more into it. Product, and then vice versa. What we’re going to start seeing is like overlaps, because I think the traditional model of, like, or view of an analyst is like, Oh, we’re just narrating, we just sit down.

[00:29:42] Bhav: I’m fucking cool. Right? I could do a product job, right? You can put me into any situation like that, and I develop skills outside of just being a SQL monkey, right? and I’m cool. That’s my choice. But my point is, we’re going to start seeing gray areas. 

[00:30:00] Audience member: And then just as a follow up to the, you know, justifying your value as like an analytics team. So like, I think like in the present day, we are basically taking very simple systems and making them more complex. So like an example is like, you know, marketing attribution or a marketer wants to know is like, where’s my money being spent? And then we’re adding all these like layers of complexity, like attribution modeling, just because it’s the easiest way to explain to a marketer that it doesn’t work.

[00:30:23] Audience member: So we go DDA or now like, do you see a world where, where they’re actually going to go from simple to complex back to simple? And then, like, you can’t really hide behind your data. 

[00:30:31] Dan: Well, the complex might just be simple by then. Like, we might, we might be more familiar with, like, you know, modeling and machine learning. And if everyone and their grandma’s using generative AI, maybe the concept of AI in our workflow doesn’t make any difference anymore, becomes normal. I mean, the whole concept of cookies was like a mapping, and it’s still kind of unknown, what is a cookie and stuff. But, I don’t know. Maybe it just becomes normal. We just talk about it as if it’s kind of expected. 

[00:30:55] Bhav: I think our ability to translate complex problems into simple solutions, this is where the magic is going to happen. Right now, you know, again, it’s great talk, but trying to translate this person that built this product into like simple enough times, it’s really hard.

[00:31:13] Bhav: So, and, I did, I, I, I’m, I’m not a natural mathematician. I did my degree in math, but I’m not a natural mathematician. I used to go to school with guys who just, you know, you just look at math from, they just, it clicked, right? It took me a lot longer to understand the problem, but when I understood the problem, I could explain it way better than those people.

[00:31:34] Bhav: And I think one of my. One of the talents I bring to the table is my ability to translate complex forms. So if we’re talking about progression into sort of like the future and what it’s going to look like, of course, I think that typical view of like, yes, I’m a data person, that’s gone, right? But the next step as data people is to be able to talk human language to people, right?

[00:31:54] Bhav: And explain all of these things. Very clearly. And they could be complex problems, to your point, or they could be simple problems. Sometimes you can’t evade, evade, avoid simple, complex problems, right? Like, I used to wrestle with this, like, why should I lower my bar? Why can’t someone come up to me?

[00:32:11] Bhav: Stop being a dick, Bhav. My CFO said this to me. He’s one of the best CFOs I’ve ever had. 

[00:32:17] Dan: We’ve nearly out of time, by the way, so if you need to leave when you’ll get some food, go ahead. We’re not going to judge you for it, but we’ve got two more questions and then we can start wrapping things up. 

[00:32:26] Audience member: I’m just wondering if, Analytics is waiting for its Black Cab moment and what I mean by that is that, There’s a thing called knowledge, a black cab driver basically successfully sort of prove what they can do. And the knowledge is something kind of, now, that feels a little bit archaic and unnecessary because now you’re just interfaced with the knowledge in some way, like you’re just interfaced with the system that gives them the 

[00:32:52] Audience member: knowledge. And I’m wondering if, we talk about one percent, we talk about two. Do you feel like when you talk about the skills that you feel like you’re talking about, there might be a tipping point? Analytics has its black cab moment. The black box. We’re just talking about the black box.

[00:33:13] Dan: Maybe the, maybe the whole concept of it, like, changes. I mean, the thing about cab driving is that, like, everyone can see it and can see it work. And it’s like, like we work in, I think, data here, it’s safe to say, and I, I take a lot of value in like cooking or making something with my hands because you can show it to someone and say, Hey, look, look what I made.

[00:33:31] Dan: You can’t do that with the work that we do. No one’s going to look at you and be like, great model, you know, or like good spreadsheet, like that’s the thing. But they, I think with us, no one can, it’s only us that are judging ourselves and proving it. Like if I say I know Google Analytics, I can’t prove that to someone that doesn’t know it.

[00:33:51] Bhav: I don’t think it’s, I don’t think the comparison, I think it’s, it’s fundamentally saying, can you drive this car? Can you drive it around the city? Whilst you’re doing it, can you talk to me? What the, in this example, if I have to really think about it, I would say that the role of the cab driver is not just, or the analyst is not just to be the cab driver.

[00:34:14] Bhav: It’s also to say, this is actually where you’re going to go. Right. Rather than having the passenger tell you where you want to go and just help them get there. I think it’s going to be actually, you think you want to go this way, but this is where you really need to go. Or you don’t need to know because it’s 5 p. m. on a Tuesday. I know because I’ve got the knowledge that going this way is the best way. 

[00:34:36] Dan: No, I’m not talking about the route, I’m talking about the destination. But the destination is always the same, I think that’s the key thing, right? It doesn’t matter how you get there. I think that’s the whole point of the question, right?

[00:34:44] Bhav: So if you have to have soft skills, and you’ve got plenty to operate, which is good and excellent, but then you’re just pushing as far as you can. That’s us now, isn’t it? It is, yeah. it’s, I don’t know, it’s a good question. Should we take this one? 

[00:34:56] Audience member: You both, loop back to something I was thinking earlier, which I think is partly why we still have black cabbies, and also partly why we still have paid search analysts.

[00:35:07] Audience member: It, in some ways, when you were saying, Oh, what we have a button where you just generate a normal, my day to day job is not analytics, but I work with people who take the job. It is analytics and I have done my entire career. 

[00:35:18] Bhav: What is your job out of curiosity? So I’m curious. 

[00:35:20] Audience member: That’s a long story. My job title is senior search scientist. Do it that way you will. It means nothing. Or it means everything. No, it means nothing. 

[00:35:32] Audience member: I might just turn this off, maybe. The, what I’d say is like the good ones, the good, good analyst professionals that I work with are not the people that could spin up an ML model in a month or 35 seconds. They are, or not at all. They are the people who understand the nuts and bolts of the data that’s coming in, so they know when it’s broken or whatever. They understand the context they’re working in, whether that be the marketing channel or whatever, to some degree. But more importantly than any of that stuff, They can, they understand what something’s significance is to the business and they can communicate that.

[00:36:03] Audience member: And that’s all irrelevant to whether I’ve got the ML model go button or not. It’s all irrelevant to whether I’ve got the knowledge of the push of a button or not, it’s why I would still go for a black cab driver sometimes and not an Uber. It’s why I would still hire a paid search agency that has PPC analysts, even though they could just be using the automated, model. It’s because they can communicate those things upwards and the significance and identify when something’s wrong, which I’m not looking at. 

[00:36:30] Dan: So it’s a question to you then, is, do you think that’s going to change in 10 years time? I agree. 

[00:36:36] Bhav: I think this is, I have a, I worked, I’m going to tell a really funny story. I worked with this guy once. I didn’t directly work with him. He was a data scientist based out in India. And, he was given a task, I was like, what’s this guy doing? Oh, he’s working on some really cool data science project, on attribution or something. He went away for like a month and he came back and, his recommendation was to put more money in the PPC brand. and it, it was the worst, like, I couldn’t believe that after a month’s worth of work, that was the recommendation. 

[00:37:07] Audience member: It’s the story he could have told for that recommendation, right? 

[00:37:09] Bhav: But that isn’t it. No, I mean, no, it’s not even, how do you, how do you, how do you suddenly put more money in PPC Right.

[00:37:16] Bhav: PPC brand by its very nature, was there any money in that book? Yes. Oh, right. But he was saying like, we should spend more on PPC brand. It’s like, well, no, the market dictates how much PPC brand we put in, right? We’re not going to, so if we’re hitting market cap on people searching for our brand term, you can’t suddenly go and spend more on PPC brand.

[00:37:34] Bhav: You have to do other activities. I think that’s this part of this context was completely missed. And I think this is where you do need the main expert. But the problem with this. is when, when you look at that, and I know a lot of like PPC analysts, again, very bright, but when you ask them to look outside of the PPC space and that ecosystem, which is part of a much wider ecosystem, how do you start bringing in things like TikTok and Facebook and YouTube and understand actually how is that driving your PPC special, like your PPC analytics?

[00:38:05] Bhav: It could be doing like your PPC, But we think about top of the ton on bottom and BBC is largely a poll channel where people are coming to rather than having stuff pushed on you, but there is an element of those push channels like YouTube, TikTok, blah, blah, blah, that are driving to that and being able to understand actually, is this organic, like this, is this happening through organic means? Or is this happening through sort of like something that’s happening outside of the world? 

[00:38:32] Dan: This is the question which we’re going to have to end it on, unfortunately, because we’ve taken more of your lunchtime than we actually intended to. Before we do that, we’ve got three mugs and some stickers to give away. So, I want to say, first of all, my favorite question of all time, Epics, if you’d like a mug, please do. I should give you a mug and some stickers. Thank you for your contribution.

[00:38:50] Bhav: I am, I am going to pick the product analyst and this gentleman over here. Alright, you’ve given us enough of your time.

[00:39:01] Dan: Alright, well thank you everyone for being part of this. who knows if we’ve even clicked record or anything. So hopefully you’ll see this in the feed, like, subscribe, all that other jazz. Enjoy the rest of your day. And, yeah.

[00:39:16] Dan: That’s it for this week. Thank you for listening. We’ll be back soon with another episode of the measure pod. You can subscribe on whatever platform you’re listening to this on to make sure you never miss an episode. You can also leave us a review if you can on any of these platforms. We’re also over on YouTube.[00:39:30] Dan: If you want to see our lovely faces and our lovely guest faces while we do this as well, make sure to subscribe to the measure lab channel to make sure you never miss an episode as they come out. If you’ll leave us a review, that’ll be hugely appreciated. You can do that on most of the podcast applications or that is a form in the show notes, you can leave feedback directly to me and love, thank you for listening and we’ll see you on the next one.

• Spotify
• YouTube
• RSS Feed