#103 Modern marketing (with Ken Williams @ DiveTeam)

Intro | Topic | Outro

Intro

[00:00:00] Dan: Welcome back to The Measure Pod with me, Daniel Perry Reed. I’m a Principal Analytics Consultant and Trainer at Measurelab. And as always, we’ve got Bhav Patel, Director of Experimentation and Analytics at LeanConvert. This is episode 103 of The Measure Pod. We’ve just finished recording with me, Bhav, and our guest Ken Williams. and we recorded this on the 15th of July, 2024 for all those people that might be listening in the future. So, Bhav, you went into this saying that you had. It’s not your thing. It’s not your area of expertise. What did you think? What did you make? Well, let’s go from there. 

[00:00:48] Bhav: Yeah. I love this recording. It was really nice being Ken. he’s, he’s got so much, so much valuable information to share with the world. And he says, I know we talked about it towards the end of the episode. we are kind of like the first ones through the wall talking about some of these things. I, he calls it modern marketing, I call it measurement strategy and it’s, and it’s effectively as two sides of the same coin.

[00:01:07] Bhav: So as soon as he started. Talking about modern marketing, what he means by it. Also, like, you know, the transformation from like marketing from 2020 to 2020. Sorry, 22, 000 to 2020. You do realise we’ve come a long way and actually the advice we would give 10 years ago is not the same we would give now. So, I’m, I’m sold.

[00:01:28] Bhav: I, I think this is a fantastic episode. I was telling you before, I think this, this, this episode is, has been really stimulating for me personally. Both for the mind, but for the soul as well. So it was really, really great to record. 

[00:01:41] Dan: I’m great to hear because I thought I was just kind of like, getting, getting someone in my world to come on the show and talk to me about stuff that I already agree with.

[00:01:48] Dan: So I’m glad you’re coming at it from that perspective. I’ve, I’ve been thinking about the same stuff. Like I’ve never labelled it as modern marketing, although I’ve started calling that now. I know that Ken jokes about trademarking it. We’re not allowed to use it unless we pay him a dollar, but I think I might owe a lot of money in that, that, that swear jar as it were.

[00:02:04] Dan: But yeah, I, I find it fascinating. labelling it is really important actually to brand this, to call it something, otherwise it becomes a fluffy floaty kind of like just get with it. Things have changed. So I really liked that. For everyone listening and watching, all of the links to what we discuss are in the show notes, especially Ken’s posts around modern marketing, what they are.

[00:02:23] Dan: You don’t have to have read them before listening to this episode, but feel free to jump in and read them afterwards or beforehand, that’s entirely up to you. we’ve got lots of stuff to plug. there are again, all in the shadows that will come out throughout, stick through to the end to find out some really fun, interesting questions that we ended up asking Ken, and see how that went and before we jumped straight into the episode, is there anything you want to plug anything going on in the 

[00:02:45] Bhav: There is nothing major going on. I will, I will continue with saying sign up to CRAB Talks, you know, come along to our events. It’s really nice meeting people in the industry. And if you want to speak on one of these podcasts, come speak to me and Dan. We’re, we’re, we’re not that scary, I promise. Amazing. 

[00:03:04] Dan: I’ve got absolutely nothing in the world of analytics to share, but I have just been put in. I’ve just been sharing this artist that I’ve just been listening to nonstop on Spotify and that’s the James Hunter six. And it was really only shared with me the other day. And I’ve already been listening to it nonstop. easy listening. Lovely music. If you’re interested, give it a go. and just an extension of that.

[00:03:25] Dan: If you are in the world of analytics and you’re not already on the measure slack, jump on there. It’s measured dot chat. And within there, there’s a, we’re doing a music league. And so people are submitting and sharing songs each week. And we’re kind of rating and ranking them each week in different themes.

[00:03:40] Dan: So, the whole music analytics crossover continues over there. So feel free to join in. I have a habit of submitting heavy metal songs each week and they get very little love. So. join if you’re into heavy metal and give me some love, please. I think I need it. All right. So today we’re joined by Ken Williams of the DiveTeam.

[00:04:01] Dan: I just want to say thank you for joining us, Ken. Thanks for joining us on the podcast. I know it’s a little bit early for you. So, where are you joining us today? 

[00:04:09] Ken: I’m in Atlanta, Georgia. So it’s 11 AM. It’s not that early, not bad. 

[00:04:13] Dan: So me sipping my beer is going to be slightly odd for you, but you can have your coffee.

[00:04:17] Ken: It’s all good. 

[00:04:19] Dan: Well, can I, we’ve crossed paths a couple of times in the past and I feel like this is something that I may have remembered more so than yourself, but this is back in the days when you’re at search discovery. And, we were on a couple of Google partner panels trying to give them feedback around their product suite, in the very classic Google way where they listened to the feedback and ignored it completely and did what they wanted anyway.

[00:04:40] Dan: but tell us a bit about your background. Who are you? What’s your experience and what do you do today? 

Topic

[00:04:46] Ken: Yeah. So, Like most people in like marketing analytics world, I had a completely different past and I went to business school, got interested in what I would have called data science at the time, like around 2009, 2010, we actually wouldn’t have used the word data science back then, but that’s what we would call it today.

[00:05:09] Ken: and then I took a job working in, A BI team actually originally, and, over time switched over to an analytics role and then worked at search discovery for about 10 years as a consultant. did all kinds of stuff. So, you know, I was a technical architect. I was a project manager. I managed the relationship with Google for a number of years.

[00:05:32] Ken: I was in charge of Google Analytics 3 60 sales for a while. So lots of very focused Google product specialisation, but really kind of like broadly marketing, Consulting focused. So how can marketers take advantage of data to make smarter decisions is really what we were, what we were selling. So yeah, that was my background.

[00:05:57] Ken: And then, you know, last year, I left my former employer and started my own company DiveTeam. So, We’re really just kind of focused on bringing in very senior people. So everybody that works with us has at least 10 years experience and we’re focused on data engineering, data science and analytics primarily.

[00:06:22] Ken: So we’re really trying to bring some sort of specialised senior experience people in to solve sort of custom problems as opposed to. You know, a lot of the prior work we were doing was just kind of like tag management. So, yeah, we’ll get into this a little bit more, but I think there’s a little bit of a timing game that we’re playing.

[00:06:44] Ken: Like, I think this is becoming more and more important. And I believe that enough that I quit my job to do it on my own. So 

[00:06:53] Bhav: How’s the journey been so far? 

[00:06:57] Ken: It’s been great. You know, we’re six months in and or so we officially launched in January and I was, very. Very scared and everything fell into place pretty quickly for us.

[00:07:10] Ken: And I feel extremely lucky. Someone told me once that starting a business is like 50 percent working really hard and 50 percent being very lucky. And I definitely had both of those. Going in our, in our favour. I would say there’s another element of just having the right people around, you know, and, and getting advice from people.

[00:07:30] Ken: And a lot of the people who are working with me now just had things in their life that freed them up to work with me. So just lots of things worked out and it’s really going great. So. I’m very excited about it as 

[00:07:42] Bhav: someone who I saw last year. Ken, I’m not sure if you’re aware. I actually did start my own consultancy.

[00:07:49] Bhav: And then like six months in, I kind of lost my co-founder and I decided to go back into full time work, but I couldn’t agree more. I think, I think 50, I think it’s actually 50 percent hard work, 50 percent luck and 50 percent who, you know, Because it’s, it honestly, like I tapped into my network so hard, that it was really helpful just to be able to leverage that.

[00:08:10] Bhav: So, I think there’s probably another 50 percent in there as well. which is something I, but I’m, I’m fairly certain doing a startup is more than a hundred percent, work, especially when we, I used 100 percent as the anchoring point in comparison to when you’ve worked in an organisation which is quite stable and what have you.

[00:08:30] Bhav: So I think, you know, so if anyone says to me like, but that’s not possible, I use, you know, I say compared to like when you’re working from nine to five, 200 percent is easily possible. 

[00:08:39] Ken: Yeah, they’re not thinking about percentages in the right way. No, exactly, exactly. I’m with you. 

[00:08:44] Dan: Well, nor has Bhav, according to his 50, 50 and 50 maths right there.

[00:08:49] Bhav: Right. 

[00:08:49] Dan: Well, I can’t wait. You, you go on, go on. 

[00:08:52] Bhav: As the person who is newest to this topic and, probably the least experienced, I’m going to ask the easy question, right off the bat, which is what is modern marketing and how does that differ to more traditional marketing? 

[00:09:07] Ken: Yeah, you know, maybe the way to answer that question is to give like. The one minute background on where the ideas came from, because I worked, I think, you know, like you guys, and most of the 2010 to 2020 period, I was doing consulting or I was working in house like for marketing teams. on using their data. So that’s kind of like what I was doing that time. And then actually Daniel, you and I had a conversation about this previously, but there was a period where I realised what I’m doing now is really different from what I was doing before.

[00:09:50] Ken: That’s really what started this concept of modern marketing. It was just simply based around the idea that. The things that I would have told a new person starting out in analytics in 2015 are not the things that I would tell them in 2025, just the skills are different. The things we’re doing are different.

[00:10:10] Ken: You know, the way that we focus on, the, the way we think about the problems we solve is really different. And so I started trying to figure out like, why is that what’s happened? And. I ended up spending probably like four or six months working through a blog post, before I finally published it, where I kind of landed on, I think there are environmental shifts that we’re going through.

[00:10:37] Ken: And I like to use the word modern just to say currently, you know, I don’t, I’m not the word modern to me is like, this is what is happening now, as opposed to what was happening before, which is really before the barriers to cloud computing were so low, you know, they’re really It’s really pretty easy to break into cloud computing now relative to just a few years ago.

[00:11:02] Ken: Of course, generative AI has changed a lot of things. We didn’t even know what that was not that long ago. The other one is privacy. I think that there’s been enough of a shift in demand coming from customers to privacy that the things that we thought of as successful analytics In 2015, which had a lot to do with building a 360 view of a customer.

[00:11:26] Ken: If you’ve been in those conversations, they’re just like, they’re not aspirations that we have anymore. You know, 

[00:11:32] Bhav: I’m getting PTSD from a 360 customer view. 

[00:11:37] Dan: my, my, my colleague, Mark, calls it the 330 customer view because there’s always 30 you can’t see. So he’s always like the three 30 is the best you can get.

[00:11:46] Ken: Well, that’s pretty wise. I mean, I, I, I would say, I don’t know if that really clearly answered your question. The, the, the answer I’d say modern marketing is just like solving problems that are the same problems we’ve always solved, which in my mind, I’m a, yeah, I went to business school. So marketing is product place, price and promotion.

[00:12:06] Ken: So like how you do those four P’s there, those P’s haven’t changed. We still have to do those things, but like the tools that we use have changed. So. Modern marketing is really focused on privacy. It’s focused on incrementality techniques as opposed to, you know, the past where we were focusing on amassing customer databases and things like that. I really liked that. That’s a short answer. 

[00:12:28] Bhav: No, I think it’s a great answer. I think it’s actually, I think it’s a fantastic answer because so often when we think about when we hear the word modern is associated with tech stack or data stack. and it’s kind of nice to hear about modern being described as current, which is a hundred percent true.

[00:12:44] Bhav: That’s all, you know, it’s, everything’s always modern at the time or current at the time when you’re doing it. but I think what’s more, more, it’s more enjoyable to hear and kind of nice to hear from my point of view. we’re talking about sort of like, you know, it’s, it’s iterative improvements and not like groundbreaking shifts.

[00:13:02] Bhav: You know, like, especially if I understood that correctly, because often people think about modern things and think, Oh my God, this is going to be some new disruptive thing. And we have to completely rip up the old playbook and create a brand new one. It’s like, no, it’s just, you did this last, you know, like five years ago, you did this now you have to do this, you know, and five years from now we’ll have to do something else.

[00:13:18] Bhav: So, I think that was really, really nice to hear. And I think the other thing that you’ve kind of sparked in me is that 2010 to 2020 are now what I’m going to forever start calling the golden years of marketing analytics. 

[00:13:34] Ken: Yeah. Yeah. What I think of it, I would, the way that I, in my mind, frame it up, I actually think 20, 2000, is sort of like the, Key year that I like to point to because that’s when Google Ads came out and then it’s not like a specific year, but there was a, there was some period or AdWords at the time, there was some period where we made a transition from what before that I would say was very focused on.

[00:14:02] Ken: If you’re a marketer, you want to get where people are, like get your brand in a department store or on a billboard or on a TV show, you know, like those are the ways you got in front of customers. And when AdWords came out in 2000, it shifted to, Oh, wait a minute. Now I can target specific individuals.

[00:14:20] Ken: And I think that from about 2000 to 2020, we went down a trajectory. Very focused on building this customer data, thinking about things like multi touch attribution, which is very much kind of based on a digital fingerprint for lack of a better word, like following somebody’s digital presence.

[00:14:44] Ken: and then trying to make decisions about how well we’re performing. At 2020, that, around 2020 ish, that changed, and I think probably Safari was the biggest player to kill it with, like, the elimination of third party cookies. There are a lot of other things that happened. I think, you know, we like to talk in the 2000 to 2020 age about digital transformation.

[00:15:08] Ken: I think COVID sort of killed that. Like there’s not that many people who haven’t transformed digitally anymore. but you know, yeah, so in my mind, 20, 2000 to 2020 is about 20 years. We were on a path that we’re really not exactly on anymore. I think some people are struggling to make that transition. So really that was really the whole concept. 

[00:15:29] Dan: So, do you think we’re regressing then? And I, and I don’t mean that in a kind of negative way, we’re going backwards kind of way, but do you think it’s reverting to pre 20, sorry, pre 2000 levels? Or is it kind of, are we in a different kind of paradigm again, where it’s kind of unexplored territory?

[00:15:46] Ken: that’s a tough question because there are definitely some things that we’re going back to and a lot of people point at media mix modelling is like the example of that, you know, like that was a thing in the 90s and then we stopped doing it and now it’s becoming important again. I think that the way that I would answer that question is that we’re not necessarily reversing, but our priorities are changing.

[00:16:10] Ken: And one way you see that play out. is with the way that we measure success. So it used to be socially acceptable to say, I got a return on ad spend of, you know, whatever from my paid search or display advertisement. I don’t think that moving forward, many people are going to hang on to that. At least the way we’ve Calculated return on ad spend as a, just based on like total revenue and total like investment in a marketing channel.

[00:16:42] Ken: I think now. People are starting to realise that we really probably were cannibalising other channels, and we weren’t really talking about the incremental impact of flipping on that marketing. So, that is something that’s driving us forward. There’s been a lot of studies done on the way that advertising over-promised itself in the last 20 years.

[00:17:05] Ken: I think there’s a Freakonomics about this that’s interesting to listen to. We kind of were lazy about it in the past, and it’s really easy to use a cookie based approach. It’s harder to do a geo test, right? Or a media mix model or, you know, something that actually is a statistically significant way of Looking rigorously at your data and figuring out, wow, is this actually benefiting me in any way to spend all this money on this marketing channel? So in that way, I think we’re improving.

[00:17:34] Dan: I think this, this, this is one of the things that I keep coming back to, and I think this is something that touches on stuff that I deal with an awful lot is that I wonder if it truly did, shift priorities. the kind of change that we’re seeing now, because ultimately the priority is to be a profitable company, to give people value, good customer experience and not piss them off or do anything illegal, right? I mean, fundamentally, the kind of bottom line for a lot of businesses, all businesses, one would argue, is the same. And that doesn’t matter if it was pre 2000 or now, or in that kind of the golden age between 2000 and 2020. What I’m thinking then is I’m just wondering if.

[00:18:05] Dan: The, the adoption of this kind of prolific technology and monitoring and measurement that we had, especially when it comes to things like attribution modelling, where we’re tracking individual people. And back in the days when cookies and browsers were allowing this stuff, it was people, less so nowadays.

[00:18:21] Dan: I’m wondering if that shifted and skewed our perspective of what the metrics are, the KPIs are. So when we were looking at things like total sessions and bounce rates and impressions and review through conversions, actually that skewed the perspective, which now we’d be in almost. Push backwards to remind ourselves that actually the bottom line is what’s important. And actually the methodology of measurement is actually the thing that’s interchangeable. 

[00:18:43] Ken: Yeah. Yeah. And, you know, I think that one of the things that a lot of people, if you were going to analytics conferences, a couple of years ago, a lot of people were talking about was value. How do we create? And this is something I was thinking a lot about when I was writing that blog post and starting to publish about this concept for a long period of time, made some assumptions.

[00:19:07] Ken: I don’t think we have anymore. We definitely used to assume that more data is good data. So let’s collect everything. You know, I would sit in meetings with clients and I would say like, what’s What are you, what do you want to see here? And they would be like, just collect it all. And then we’ll just analyse it any way we want later.

[00:19:24] Ken: And the insights will fall out, you know? And so I think that part of what is shifting is a little bit of burnout on that. Like I think everyone. It’s kind of hitting this area where we’re like, you know, there’s a lot of money that’s been invested in standing up analytics tools or CDPs, or, you know, you name it tools that are very expensive that then nobody goes on to use.

[00:19:50] Ken: So how are we going to make sure that we’re getting something valuable? Return on our investment for these things. Because of your point, businesses need to make money. And I think that we’re a little bit moving in a direction towards, I don’t want to capture everything. I want to say what the business is trying to do.

[00:20:07] Ken: You know, maybe we’ve got a product and it’s brand new and we just want awareness. So that’s fine. Like we don’t have to push revenue. Like maybe we’re just trying to get people to watch our. Videos we’ve put on YouTube about our product. That’s fine. That’s a business objective that you’ve decided on. So then let’s figure out how do we measure that?

[00:20:28] Ken: And then what data do we need to measure that? And really laser focus on hitting some goals with specific targets attached to them. Learning what’s going well, what’s not, and changing things. And that’s where we get into that, like, iterative improvement concept that I think is really starting to emerge significantly in this period in a way that, in the past, we would just dump a bunch of money into things, at the end we would try and show the analytics in a way that made it seem like it wasn’t a waste, and then we were already moved on to something completely different.

[00:20:58] Bhav: And I saw a blog post, or a post on LinkedIn recently, which talked about how click through rates are, are meaningless and, and people shouldn’t, and execs shouldn’t be chasing things like CTRs and impressions and things like that. And then, you know, the author of the post was talking about how we’ve lost the ability to do brand marketing and things like that.

[00:21:18] Bhav: And I disagreed to a certain extent in the sense that, we, we, we gave up on those metrics a long time ago. They were the metrics that got us, you know, into the door and the seat at the table in discussing analytics at a very exact level. But I think very quickly we realised that those metrics are meaningless unless we attribute them to things like ROAS to ROI to CPA to, you know, to whatever it might be.

[00:21:40] Bhav: So when we, you know, we talk about, and you mentioned that, you know, you mentioned that ROAS, certainly in the way we traditionally measured it has, you know, it’s, you know, it’s probably moved on a little bit. I don’t think it ever really will. I mean, There will be, there will be some type of return that we need to recognize through spending and, you know, on marketing channels.

[00:22:05] Bhav: I think what’s probably changed, and then, you know, to your point about iteratively, it’s just that we’ve gotten better at targeting the right metrics and looking at the right metrics. I think that’s the first evolutionary step. As we move away from being able to do CPAs or cost per clicks or, you know, where it’s directly attributable to a keyword or a campaign or an ad, we will still need to be able to show some type of ROI.

[00:22:28] Bhav: Don’t you think? it’s just that the tools we use are going to go back to something that, you know, the tools we would have used 20 years ago to measure TV campaigns or, or out of home, campaigns. And it’s, Not, you know, I don’t think we’ll ever really be able to move away from showing value because I think ultimately there’ll be people who care about the value.

[00:22:49] Bhav: I think for some companies, you know, for some products like Nike, you know, where it’s about brand perception, there will be that, you know, ongoing pursuit of excellence in terms of brand awareness and brand recognition and recalls and things like that. But for anyone who operates in a pure digital e com space, getting that ROI number is always going to be valuable.

[00:23:13] Bhav: I think the only thing is now that we’re starting to see the bigger picture and how all the pieces get to fit together. you know, fit together and, and data scientists, you know, whatever you want to call them these days, marketers, analysts, we’re all working together to fit all these pieces together.

[00:23:26] Bhav: So, and you’re right, you know, we, you know, the CDPs and things like that, we’ve moved away from collecting everything to just collecting better things. So that’s my kind of view. I don’t really have a question in there. I just, you guys said some interesting things and I wanted to throw, throw my thoughts in.

[00:23:41] Dan: I’m just going to jump straight in because I, I get your point, but I think there’s, there’s a, there’s a key thing which I think is happening, which is defining these metrics. You mentioned ROI and ROAS interchangeably there. And I think when we’re looking at ROAS, that’s very much an in product thing.

[00:23:55] Dan: And that’s return on ad spend. That’s Google Ads telling us how much money they’ve spent and how much money they’ve made us back. But actually ROI is looking at, okay, what’s the cost of selling this good? And what’s the profit we get back and how much on top of the ad spend did it cost? I think these are the things where.

[00:24:10] Dan: I think it’s been going on for so long, even since 20, well to the year 2000, as you said, when Google Ads came out in the Google Ads example, where they are defining metrics that they are, they can measure that they would like us to look at. And actually it’s such a big, and it’s mad. It’s such a big revolutionary thing now for Google Ads to take in profit.

[00:24:28] Dan: And it’s this modern, crazy, awesome thing. And you’re like, Jesus Christ, we should all be looking at profit. Like we should all be looking at ROI. And actually we’re now in the year 2024. 24 years after this tool was introduced and now it’s a crazy new novel idea to look at profit. And I think this is where we are slightly blinded to the products and the tools we’ve been using.

[00:24:47] Dan: Like every dashboard I’ve ever seen, a website dashboard is looking at sessions, users, page views, bounce rate. But I don’t think I’ve ever met anyone that knows the definition of them, let alone understanding the relevance of them. Or as you said, Ken, which is the most important thing, which I never see.

[00:25:02] Dan: It’s one thing having KPIs, but nothing’s having a target for them. How much is good? How much is bad? And this is a thing that a lot of people don’t have. And we are, we’re all suckers for marketing in our own way. We’ve been fed these data points, these metrics from different tools, and that’s not our fault.

[00:25:17] Dan: That is just lots of technology feeding us lots of data that we perceive to be important. But actually, I don’t know if it is, there’s always going to be indicators of did the content do what you want it to. And it’s not always possible or easy or even real time or frequent enough to be able to measure the Impact on the bottom line for sure.

[00:25:34] Dan: So things like click through rates and impressions are still going to be interesting to some extent, but, yeah, sorry, that’s my rant over off the back of your rant, Bhav. 

[00:25:43] Ken: No, I, I love what you guys are saying. I, it, I put together when I was working on the modern marketing idea that I think there are these four different principles that like sort of define it.

[00:25:55] Ken: The first one is this outcome based measurement strategy concept. And I always put a short sentence that says this is defined as abandoning a single source of truth in favour of being flexible and being focused on what you’re trying to do as a business, rather than really just taking the things that are easy.

[00:26:17] Ken: So to your point, I do think that a lot of the Reason that we would look at things like sessions in the past is because it was just right there. It was provided. It’s very easy. You could say something like, Hey, we have X number of sessions this month and last month we had less. So it feels good. But I think that.

[00:26:37] Ken: Whenever a client asks us to design, a dashboard for them, or even if we say we’re going to implement Google Analytics and then we’re going to provide you one, even without you asking, the thing that we always do is land on what are the two or three things you care about as a business. So I use that example a minute ago, like maybe we want to build awareness of our brand.

[00:27:01] Ken: Maybe we’re trying to improve a conversion rate on a checkout flow, whatever it is short term, this quarter, or maybe this year, we care about these two or three things. Then we say, how will we know if we’re successful at that? And don’t let the data that’s available in Google Ads Drive how you measure.

[00:27:21] Ken: If you’re doing that, think about like, what are we ultimately trying to do? What data will, if we look at these data points, do we feel like best approximate that thing? And then when we deliver these dashboards, we always deliver them in a way where you can see how metrics roll up to some objective that you’re trying to accomplish as a business.

[00:27:40] Ken: And every time we do that. 100 percent of the time the client’s response is, could you reorganise this? So I can have one section about paid search and another section about display and another, because that’s what everybody wants to do. And it’s just a different way of thinking that, you know, I like a lot of my premise.

[00:27:58] Ken: To this modern marketing idea is that this is not something I’m pushing. This is something I’m observing and I’m observing it in small pockets. It’s not everywhere. I think most people are still in a mindset of just show me how many sessions I got through, you know, my display campaigns last month, yeah so that being said. 

[00:28:20] Dan: Now for our long term listeners, it’s no secret that I like to run training courses all around Google Analytics, Tag Manager and everything in between. Check out a full list of courses over at measurelab.co.uk/training to see all the courses and workshops that we have available.

[00:28:33] Dan: Everything from learning Google Analytics 4 to Google Tag Manager, data visualisation, and short for workshops to cover small specific areas of interest, such as user provided data, generative AI for data analysis, as well as lots of other stuff. That’s measurelab.co.uk/training for all the details, or you can click the link in the show notes, or if you’re watching this, scan this QR code now.

[00:28:54] Bhav: I mean, I think this is a really fascinating point because I’ve just done a piece of analysis for a brand. I can’t, I wouldn’t say on the podcast, but it was the brand we were just talking about that me and Dan seem to be drinking every day of our lives. And the analysis really showed that actually This brand was spending X amount of money on, on Facebook and Meta X amount on paid search, but they were also investing in TikTok and, and YouTube.

[00:29:18] Bhav: And they pulled their TikTok spend and start pushing the YouTube spend. But actually the analysis that I did, I did, I did a regression analysis to understand two things. A, well, there was a direct conversion analysis, but I was also trying to see what drove and fed into the, into the brand and the, PPC brand and the SEO searches.

[00:29:39] Bhav: And actually TikTok was a much better platform. So if you have your direct conversion rate from a platform that you say, this is what we’re going to use classic, you know, conversion rates, CPA type metrics to spend on this brand. But we’ve now also had this, on this media. But we’re also going to spend X amount on these other channels, which we know drive brand awareness, but they don’t have the tools at their disposal.

[00:30:04] Bhav: They had the data. They just didn’t have the tools. and by tools, I mean, knowledge entails things like that to be able to say, well, actually does this drive, how do you drive, like what is the value of TikTok? What is the value of, of, of YouTube and people always directly use conversion rate because it is the easiest thing to measure.

[00:30:22] Bhav: And I understand why they want, you know, as you said, Ken, they want sessions. They wanted to have these numbers, which are easily, you can break it down, but trying to then take a study that is like a regression study of the impact of your TikTok campaign and putting it into a dashboard or something that says, Hey, look, it’s hard for me to say this is the number, but what I can say is that there is this correlation coefficient or this, coefficient in front of your X value that shows that the more you put into this will drive up your Y, which is your brand, which we know converts out.

[00:30:57] Bhav: A really high conversion rate, but trying to get that into a metric or into a dashboard format is so hard. and I, until we, it sounds really bad, get rid of the old, you know, the old God of the marketing directors and marketing leaders. We’re never going to see this transformational shift into this new way of thinking where we’re adopting better metrics and thinking about things holistically outside of this is my PPC campaign. This is my SEO campaign. This is my Facebook campaign. So I’d love to hear your thoughts on that. 

[00:31:26] Ken: I think. I think what we’ll find is that those people who are struggling with adapting to understanding things like how regression is different from what they saw in their, like, TikTok reports, those people will be replaced over time.

[00:31:43] Ken: And actually, I love the movie Wall Street. Have you guys seen this movie from 1987? So I put it, I put this in one of the blog posts I have about. Modern marketing, because I actually didn’t watch it until, maybe like 2010. And it occurred to me that this movie came out in 1987. It’s very much focused on how the finance and really like investment banking world was I mean, it’s focused on insider trading, but the huge, not overwhelming, theme of the movie is about technology.

[00:32:21] Ken: So everybody who’s really good at investment banking has computers all over their desks. And they’re like these old silly looking computers to our standards nowadays. But if you think about it, personal computers came out like 1980, 79, 78, around there. This movie was only nine years later. And a lot of it was making fun of people who just couldn’t adapt because before that investment banking was about looking for stable companies with a good product.

[00:32:48] Ken: And it became quickly about information, like who’s got information the fastest. And I think marketing is going through something similar where. We used to be successful by looking at information that you would get from a report provided by your ad platform. Now, it’s a much better decision to run a regression on your data and find out what kind of impact it’s having.

[00:33:12] Ken: And those people who can’t transition to that are like people in finance who weren’t. learning how to use a computer in a lot of ways. So I think over time, there’s going to be some people who are like, you know what? I don’t want to adapt. I’m happy with the way I learned how to do marketing between 2000 and 2020. And I do think they’ll be replaced over time because they’re not going to perform as well as their competitors. 

[00:33:34] Dan: But is there, is there an inherent risk there that they won’t leave? And so the business will tank because I think a lot of the Stuff when it comes to the modern data stack and that’s modern marketing and you know, the kind of It makes a lot of sense.

[00:33:47] Dan: The three points you said about AI, sorry, generative AI changing stuff, privacy changing stuff and access or, or entry to port to cloud computing kind of reducing. there’s going to be people out there, businesses that won’t see that, or don’t have that, or they have staff that. are unaware or not wanting to address the cognitive dissonance of change, right?

[00:34:07] Dan: So is there, is it inevitable? Are there businesses that will continue to, to, to just keep cracking on going forward in what they’ve been doing? Is this an inevitable kind of success factor? I’m just trying to understand if, adapting to this change is something that has to be done or is it more so for the kind of hyper competitive market and the ones that have got monopolies, the only product in the market or you know, good brand recognition like Disneyland or Coca Cola, would they ever need to worry about this kind of stuff?

[00:34:36] Ken: Yeah. Well, you know, I think that there is a scale. Important thing about scale that we should mention, which is that if I’m a coffee shop, you know, I’m not that impacted by cloud and AI and these, and privacy and things we’re talking about, but there’s, when we talk about a certain scale where you can drive meaningful efficiencies out of it, Then I would say, you know, like businesses sort of come and go.

[00:35:01] Ken: And going back to the analogy I gave a moment ago, I’d look back to, you know, those big investment firms in the seventies that decided not to adopt computers, like how did they perform moving forward today? Computing is a huge component of it. And in fact, the reason I like this movie so much is because when I saw it, I was taking a corporate finance class and business school, and we were learning about all the algorithms for tracking like the performance of the market and like the, your ability to perform above the market, it’s called like generating alphas, but that was all brand new.

[00:35:36] Ken: Like all of the things that we were learning in those classes had been developed since that data became really available. And, you know, I think time will tell is a short answer, but I don’t think it’s going to, be a very easy road for people who are making decisions in going back to the specific descript, scenario we’re describing people who are making decisions exclusively on, What TikTok is telling them a smart decision is without thinking about how can I, like, as a business, if I want to brand awareness, that’s one input among many. And how can I better use all of the inputs that I have available? 

[00:36:20] Bhav: So if I’m a, a marketing director and I’m listening to this podcast, which I hope they are, what are the kind of like five, three, let’s say three to five things you would say to them? And let’s assume this marketing director is not. As tech savvies, you’re not in touch with sort of like modern marketing and analytics practices.

[00:36:37] Bhav: What are the three to five things you would say to them to go in and look at? I know we’ve touched on attribution modelling or mixed media modelling or regression, whatever it is. Is there anything else that you think they should research as a topic? We’re not going to be able to cover all of them, but it would be great if anyone is listening.

[00:36:52] Bhav: So, hey, look, you know, Ken’s been talking about the modern, modern marketing, modern marketing and I’ve been questioning, what should I do in research? 

[00:37:01] Ken: Yeah, I would say the most impactful thing that anyone can do at any size business is set targets, you know, and I think we’ve mentioned this briefly earlier, putting, some time into measuring what you want to accomplish or defining what it is, like, you know, maybe this is an OKR at a lot of companies, then putting the creative Exercise into answering that question of how will we know if we’re doing that and then setting a goal that alone is just, the most revolutionary thing that, you know, I bring to a lot of my companies and it seems silly, but it’s like, at the end of the day, we can be really confident that we are.

[00:37:48] Ken: moving the needle and when we’re not moving the needle, we can have honest conversations about why not, in a way that we don’t have to, it doesn’t have to be about anyone’s pride, you know, like everybody’s working together. We’re not trying to say like this person did a bad job. It’s like, no, we’re going to.

[00:38:04] Ken: Are we on track? Are we off? What could we do differently? Like that’s, it’s more of a cultural thing to me. That’s like the number one thing anyone can do. I think to your specific question though, the way that I think that a typical marketer. Can tiptoe into thinking in a more modern way about performance is to combine incrementality techniques with multi touch attribution techniques.

[00:38:34] Ken: So what I mean by that is if I am say, I’ve got, Google Ads and I’m running some search campaigns, I’m running display campaigns, I’ve got YouTube, running, and maybe I’ve got. Other things happening and that’s not that are not in Google Ads or in other places. I get instant feedback on when people are clicking on those ads, because I can see they’re landing on my website.

[00:38:59] Ken: I can see this campaign is performing better than this one, and I can get a good relative view of what’s working. But what I don’t know is. Does this actually create value for the business or not? I think the concept that you would say, you know, we spent 100, 000, we made 200, 000 is silly, right? I mean, because there’s so many variables in there.

[00:39:22] Ken: There’s things like accounting for profit. There’s things like accounting for the customers that would have bought from you anyway had you not advertised to them. So it’s not helpful for answering that question. It is helpful for optimising relative performance of the different tactics you have in the market.

[00:39:38] Ken: Though, the, the other thing that you can add on then is this incrementality, like, okay, I’m running in one, I’ve got a couple of geos where I’ve just turned everything off. Do I see a difference in the areas where I’m advertising versus those where everything is shut off? That’s a pretty easy way that you can say like, yes, now I can see I got a measurably boost in revenue.

[00:40:02] Ken: Above what I see when I shut everything off, like doing A/B test like that, a media mix model is just an A B test. That’s kind of thoughtless, right? It’s like, it’s just going to look at your historic data and see when you crank things up and down. Another version is a randomised control trial, which is where you’re like just planning. 

[00:40:22] Ken: I’m planning to shut off everything in Texas for a week and then shut it back on and crank it up really high. There’s all different ways to do this. and then the other thing that is kind of lazy but nice is like a pre post test. So you just look at the data and say like before and after we started doing something.

[00:40:39] Ken: Did we create a forecast? We look at our data up to a point. We made a change and we forecast the future. Then we look at the actuals and we see if there’s a gap between our forecasts and the actuals. So all of those are great techniques, but the idea is that those answer that bigger question of is this working, you know?

[00:40:57] Ken: and I think that you need that. I think that’s very important. It takes some effort to get. And I think you also need to be able to look more quickly at that attribution level data to say like, okay, this campaign’s performing better than that one. 

[00:41:11] Dan: So you, you’ve mentioned attribution a couple of times and I’m going to have to jump in because I feel like I’m about to explode.

[00:41:17] Dan: And I have to ask, my background is in attribution. That’s how I cut my teeth in the analytics space before moving into a kind of Google analytics specialty, I suppose. Attribution can tell you what you want it to say. And we keep talking about things like conversion rates and return on investment or return on ad spend.

[00:41:34] Dan: But all of these can be wildly different depending on the model you pick. so when it comes to attribution, you’re saying about combining that, that there’s still an element of the human. Kind of deciding them and picking the one that makes their numbers look bigger and often things like last click attribution, upweights things like search marketing, especially brand marketing.

[00:41:54] Dan: And then kind of makes the things like social and prospecting campaigns rubbish. So there’s still an idea there that we’re all going to be reading different things and overlapping each other. And how do you address that? Like, do you have a go to model that you just say, use this and move on? If you’re building these dashboards for your clients that are using and combining attribution techniques with, incrementality studies and things like that, like, how do you address that conversation of attribution, not only because there’s infinite variety, but also the observability, things like cookies and things like that, that it’s dependent on that they’re inherently flawed and playing a bit of devil’s advocate here, cause I’m a fan and I agree with you about the attribution modelling, but I think this is what a Kind of agency will come in and be like you absolutely cannot use attribution modelling because of all these reasons I’m just wondering what your perspective your I suppose answer to that would be.

[00:42:42] Ken: Yeah I mean my answer is attribution modelling and I would loosely define Attribution as anything that’s based on a cookie identifier that’s tracking my behaviour, right? So it could be first touch last touch but anything in that category Like you mentioned, we know it’s, we know it’s flawed, but not only that, it’s got these other challenges with things like it, it only sees its own impact.

[00:43:08] Ken: Doesn’t take into account things like would somebody have purchased a product had you not advertised, that kind I do think that when you’re just focused on Advertising. As long as you are aware that what I am doing with this data is trying to optimise the relative performance of my different tactics, then I think attribution is really important.

[00:43:32] Ken: Media mix modelling is an incrementality technique that has its, that has its place, but it’s also important for you to be able to see, I’ve just pushed out a campaign. Maybe it’s been running for a week. Maybe I’m getting the number of click throughs on this one that I’m getting on that one? If not, maybe my creativity needs to change those kinds of things.

[00:43:50] Ken: I think the attribution model you choose to answer your question should be informed by the channel that you’re looking at. So, you know, if I am looking at an acquisition channel, then I probably want to look at a first touch attribution model because that’s the one that really answers the question of Like are people initially clicking who have never known my brand before clicking through this ad.

[00:44:17] Ken: It’s not, again, it’s not really perfect because they could be the same person on another device. But if you see over a large number of people that when you slice the data, that way. If this campaign performs better than that one, then you can make an informed decision to say, I’m going to shift money into that one, or I’m going to figure out why that one’s working and not this one.

[00:44:37] Ken: I don’t think attribution is what I would use to answer the same question as a MediaMix model. In my mind, Media Mix Model answers this question of like, what are you getting in return for it? That attribution isn’t really trying to answer. 

[00:44:51] Bhav: I, yeah, I love that answer. I think it’s so true. And I think, Dan, do you remember, I tried to bait Tom Goodwin into joining this podcast for a discussion around this? so Tom Goodwin’s the person I was referring to. His was a blog post that for lack of a better word, triggered me. He was talking about, oh, you know, click through rates, this, and, you know, if all, you know, if all you’re going to, we need to move away from this click through rate mindset. And I was like, no, well, we have moved away from it.

[00:45:16] Bhav: and he was talking about branding and how he’s like, never purchased any, he looked around his room. He’s like, I’ve never purchased anything that I’ve, there’s nothing in my room that I’ve purchased that’s come after I’ve clicked on an ad. And by the same logic, I would say, well, there’s, I look around here, there’s nothing that’s come I’ve purchased has come from a pretty billboard or a, you know, of a, some branding campaign.

[00:45:35] Bhav: pain. And actually it’s not true. I think in reality, it’s a mix of both. And you said it there yourself. I think what he was doing was confusing two different things. He’s confusing mixed media modelling, where we have to look at the performance of these very quantifiable channels that we, you know, we never used to have like radio, press TV.

[00:45:53] Bhav: These are really hard to quantify channels in the past. And we have to use methods like and what have you. And then we couple that now with our ability to be, to be able to track customers and what they click on. And then we bring in things like branding elements, like actually, does this ad look nicer?

[00:46:08] Bhav: And does it drive more clicks? Yes. If it does, that’s great. But is it converting into more value at the end of the day? And if it’s not, then we need to go back and reframe. And actually it needs to be this harmonious or symbiotic relationship between, you know, good quality ads that do what they’re supposed to do, which is, you know, drive sales.

[00:46:27] Bhav: but also send the right brand message and build that awareness and, you know, be, you know, stand out from everything else. So I completely agree. And then going back to your points around, you know, increment how combining incrementality techniques with multi touch attribution and trying to understand the value.

[00:46:45] Bhav: Man, you’re preaching to the choir here. You basically, you call it modern marketing. I call it measurement strategy and we’re all working towards the same goal. And as soon, when you were talking about it, I was like, you know, like, where have you been all my life, dude? So again, just some Ken loving there. 

[00:47:04] Dan: So, so what you’re saying, Bhav, is that your approach isn’t modern. Is that what you’re saying? 

[00:47:07] Bhav: I’m saying that I never would have called it modern. I just would have called it logical. And I came into this, into this podcast thinking that we, you know, we were going to talk about modern marketing from a And actually what you’re talking about is like so on point with what I, what I preach about, which is that modern marketing, modern product analytics, modern, you know, experimentation is really about quantifying things and having the right measurement in place to track the value that you really want to, that you really care about and trying to find out how a connects to be, you know, and that’s for me is what measurement strategy is.

[00:47:46] Ken: I would call you an early adopter. You know, and I would say. 

[00:47:49] Bhav: I’m a pioneer, Ken. I’m a pioneer. 

[00:47:52] Ken: You’re a pioneer. What I would say I’m trying to do with this modern marketing thing is just give that a name. Right? I, I, I’m saying that there are people who are really smart who are doing this today. I’m not trying to invent something new that I’m teaching the world.

[00:48:06] Ken: It’s more like. There are people like Bob out there who have figured out a smarter way to make decisions. And I’m trying to think about like, where does that come from? And how do we communicate what it is to people? And that that’s really what I’m trying to do with modern. 

[00:48:20] Bhav: I mean, I love it for what it’s worth. I love that you’ve called it modern marketing.

[00:48:27] Ken: Well, I’ve trade, I’m a trademark that, so, so don’t use it again. I just bought the domain. You have to pay a dollar every time. I just bought modernmarketing.com.

[00:48:39] Dan: Ken, I’ve got one, one last thing that I want to kind of tee you up for before we start wrapping up and going into some of the fun stuff. So, your background is very much around Google Analytics. You were the manager for Google Analytics 360 and a lot of stuff. Now, I have to give you 60 seconds about Google Analytics 4.

[00:48:55] Dan: What are your thoughts? How do you feel about it? Is it game changing? Is it crap? Is it awesome? What do you think? Go. 

[00:49:02] Ken: My answer is that I’ve been avoiding using this language because it’s a great data collector, but that’s really the way that I’m using it now. I love the way that the data is collected and I love the integration with big query.

[00:49:17] Ken: And that’s all I really do these days. So if someone feels. Comfortable, taking advantage of those tools, which are really, really powerful. I think it’s the best choice out there for the price. but even if you avoid the fact that it can be totally free, if your volume is low enough, I think it’s a great, great tool.

[00:49:40] Ken: I am very, very frustrated with some of the user interface challenges that, you know, everybody. I Struggle with it, but I still use it. I still prefer it over lots of alternatives. I will say when clients come to me and they ask for an evaluation, there are times where I might suggest something else just based on what they’re trying to do or their skillset, or particularly their comfort using. a tool like BigQuery, things like that. It’s not a perfect fit for everyone, but that’s my answer. 

[00:50:13] Dan: Okay. And, and one last point, just because you’re not a search discovery anymore is GA 360 worth it? 

[00:50:20] Ken: Oh, well, yes. 

[00:50:22] Dan: I don’t need this. I don’t need the sales pitch. It’s a yes 

[00:50:25] Ken: Or no. 

[00:50:25] Ken: I’m not trying to sell you anything. But because I love the integration of BigQuery so much, if you are generating more than a million events in a day. You really need to pay for GA 360 so that you can get the BigQuery export. Otherwise, it cuts you off and your data doesn’t flow into BigQuery. I have one client who I get an alert like one or two days a month because they had a spike and they went over that limit.

[00:50:48] Ken: And it’s frustrating because it’s like, well, now the BigQuery data is going to look weird. And it’s, you know, so once you start bumping up a million events a day, 50, 000 is the starting price in the US. It scales up from there. It’s worth it. and it used to be 150, 000 starting price. So I, I still, I still am a fan.

Outro

[00:51:13] Dan: So Ken, we’ve got, we’re just going to wind down now. I’ve just got a couple of questions. So, we’re trying something new as we’ve kind of brought back the podcast. I’ve got three AI generated questions for you based on the impact of your blogs around modern marketing. So I don’t know how well this is coming up. I’m doing this live on the podcast. Right. Question number one, if you were to create a marketing campaign for any fictional character or universe, who would it be for and why? 

[00:51:45] Ken: A marketing campaign. I mean. 

[00:51:47] Dan: If you had full creative control. 

[00:51:49] Ken: Full creative control. I, oh, that’s tough. I’m a big fan of the show Bob’s Burgers. So because it’s where my passion is, I would love to do a marketing campaign for that show.

[00:52:07] Dan: Do you already have an idea at this point? Or is it just like, kind of like you want to start there and see where it goes? 

[00:52:11] Ken: You know what I, you know, what I would do in that situation is I would get the cast in a room and I would not bring any, this is very modern. I wouldn’t bring any preconceived expectations. I’d be like you guys are some of the funniest people I know. Let’s just enjoy talking about what we could do together. And that’s how I would approach that. But I would set targets after we got through that. 

[00:52:33] Dan: I like it, bringing it back. Okay, question two. If you could have any superpower to help you with marketing, what would it be? And I’m going to extend that to marketing analytics and stuff. If you could have the superpower. 

[00:52:45] Ken: if I could have a superpower, it’d have to do with time. I think that if I could just add a couple of extra hours to the day, I could get so much more done there. There’s so many, there’s so many things that race through my mind all the time about, this low barrier to cloud computing conversation we had earlier that I actually, I teach this class on using.

[00:53:13] Ken: the Google Analytics data and big query and specifically in this tool called data form and a couple of days ago we were in this class and we were talking about how it was only. Two years ago that Google cloud added Dataform and I was laying out, here’s what people used to do before that. Isn’t this ridiculous?

[00:53:32] Ken: You know, like it’s like, it’s like the tools are making life so much easier and I’m constantly faced with choices about how to invest my time in kind of building out new things. And I would love to not have that as a barrier anymore and just have enough time. To get in, think thoroughly through all the new functionality that’s available, all the ideas that I’ve got, like just explore it. So there’s your answer, add a few more hours to the day. 

[00:54:00] Dan: I like it. and last one. what is one quirky, I’m reading this verbatim by the way, so don’t hold me again, hold this against me. What is one quirky or unexpected talent that you have that your audience might not know about? 

[00:54:14] Ken: Well, I have a guitar, but I’m not really a guitar player. I was a drummer for a lot of years, and I hung out with a lot of guitar players at the time, so I’ve got these relics of that. So, yeah, if there was something I could do, you know, to have fun with some friends, I’d love to play drums again. Maybe that’s the answer. Amazing. 

[00:54:36] Dan: I think from doing this, these podcasts for a little while, there’s a number of people that play a lot of instruments. I’m wondering if we can get the next super week or something like that, like a, like a one time only band formed or something like that. 

[00:54:47] Ken: Hey, I’m game. In fact, a couple of years ago, a lot of people brought guitars to super week. So nobody brought a drum set though. I don’t think I’m going to be the one.

[00:54:55] Dan: Just pots and pans. That’s it. Round the campfire. amazing. I promise you that’s the end of our AI segment. So, thank you for humouring me and our chatbot for doing that. And, Ken, before we go, are there any last plugs or anything you want to share or anything that you want to tell the people listening and watching before we start wrapping up?

[00:55:15] Ken: You know, the only thing I would say is I am Really excited about a lot of the cloud work that we’re doing. It ties into all this. The reason that I decided this is a good time to quit my stable job and start a business is because I do think that there’s such a need for companies to have some expertise.

[00:55:37] Ken: And like how to use cloud, particularly in my world, Google cloud to do in useful things with your data, like running a regression analysis, like we discussed earlier, but even just like joining data across sources so that you can see, how different things correlate with each other. So, I’m very excited about a bunch of training that we’re launching in this area, a DiveTeam.

[00:56:05] Ken: And so that’s my, my thing. If I could plug one thing, I’d plug that. I think it’s super, super valuable. So, I hope that even if someone’s not taking our training, that people are really staying on top of what’s happening in that space. 

[00:56:19] Dan: Amazing. Thank you so much, Ken. It’s been a pleasure to speak to you properly and record it for our audience. And, yeah, just for our audience, all of the links that we talked about include links to the DiveTeam and the blog posts and everything about modern marketing will be in the show notes. So whatever app you’re in, click the link and it will all be there. Thank you so much again, Ken. Thanks for joining.

[00:56:39] 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.

[00:56:53] Dan: If you want to see our lovely faces. is, and our lovely guest faces while we do this as well. Make sure to subscribe to the Measurelab channel to make sure we 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 Bhav. Thank you for listening and we’ll see you on the next one.

• Spotify
• YouTube
• RSS Feed