#86 The evolving challenge of SEO measurement (with Alex Harvey @ ASOS)
![The Measure Pod The Measure Pod](https://www.measurelab.co.uk/wp-content/uploads/2023/07/The-Measure-Pod-artwork-Measurelab-and-CRAP-full-150x150.png)
In this episode Dan, Dara and Bhav had the pleasure of chatting with Alex, Head of SEO & Marketing Technology at ASOS. They delved into the world of SEO, measurement and analytics and discussed some fascinating topics. One thing that stood out was the evolution of SEO over the past decade, from the ‘wild west’ days of link spamming to the more strategic and data-driven approach of today. They also explored the importance of data literacy in SEO, the role of keyword-level data, and the challenges of forecasting in the ever-changing landscape of search.
Show note links:
- ASOS tech blog on Addressing LGBTQ+ bias in GPT-3.
- Bhav’s blog on The Region-Beta Paradox and its Impact on Product and Experimentation Thinking.
- CRAP Talks blog on Toys “R” Us vs. Smyths – A Modern Day Tale Of David And Goliath.
Follow Measurelab on LinkedIn and on Twitter/X, and join the CRAP Talks Slack community.
Find out when the next CRAP Talks event is happening on LinkedIn.
Music composed by Confidential – check out their lo-fi beats on Spotify.
Master Google Analytics 4 with Daniel Perry-Reed on the next GA4 Immersion 6-week cohort training course. Early bird, charity and group discounts availible!
Quotes of the Episode:
- “In terms of the analytics piece, everyone is kind of forced, I think now to become a, like a lot more involved, a lot more aware, a lot more capable with data, which I think is, is really exciting, but also a little bit dangerous at times.” – Alex
- “For a company like ASOS, you have to lean on the expertise that exists beyond just what the data that, that you have access to tells you, particularly for new and emerging trends like the retail teams and you know, the buyers and merchandisers that we work with, the insight that they have, the knowledge that they have of their, of their domain is really valuable and you can’t, you can’t dismiss it.” – Alex
- “I like you Alex. Like, I love hearing people doing these type of things properly” – Bhav
Let us know what you think and fill out the Feedback Form, or email podcast@measurelab.co.uk to drop Dan, Dara and Bhav a message directly.
Transcript
The full transcript is below, or you can view it in a Google Doc.
Intro
[00:00:15] Daniel: Even though we started this podcast talking about SEO and speaking to Alex was awesome, you hear about that in a moment. I have to say that I’m proud of myself because I don’t think I brought up GA4 once. So if you’re playing The Measure Pod bingo, I’m sorry, you can’t win that time. If you’re playing the drinking game, there’s no shots there. I think this is maybe the first episode I’ve never actively brought up GA4, so I’m going to give myself a little pat on the back for that. But I thoroughly enjoyed the conversation. I think it says a lot to the conversation we had with Alex with how good it was talking about ASOS’ approach to SEO and how that all works that I didn’t feel the need to bring up GA4 can you imagine?
[00:00:47] Dara: I think that’s true, actually. I think it is a sign of how good the conversation was because not only did you not bring up GA4, there may be tiny spoiler here, but Alex also mentions attribution towards the very end, and that’s also something that you like to mention quite a bit, Dan. So the fact that you didn’t bring up either of those two or somehow managed to shoehorn them into the conversation just shows how much you were, you were listening and we were enjoying. And I think from my perspective, what I find interesting about these kind of conversations is you’re hearing about somebody’s real experience. They’re not just kind of like talking theory to you or talking about how things could be. It’s actually talking about how this is working at a kind of big ecommerce brand that everybody knows about. So it’s kind of nice to just take that in and talk about the kind of challenges that they face within that kind of company.
[00:01:30] Daniel: With true expertise as well right. That’s the other thing is like we’re talking to someone that knows their craft and they’ve been doing it for a long time and it shows. And I think whenever you talk to anyone that is passionate and knows their, knows their subject, I mean they, you could be talking about folding napkins for all I care, but it kind of rubs off on you a little bit that. The intrigue that I have about someone that’s clearly passionate and knowledgeable about something is contagious.
[00:01:49] Bhav: And I think what’s really great when you hear Alex talking about SEO at ASOS, you realise that even though we didn’t cover some of the probably big topics around analytics and even attribution, how much information there is still to absorb and learn, you know, for SEO specifically and then I think Dan, you know, you mentioned it just now, when someone knows their craft really well, actually you can go down rabbit holes on topics within that subject that you never even knew existed. You know, by doing so you kind of avoid those very generic conversations of things that people probably expect to hear and you kind of go down these really interesting routes into like really unexplored territories.
[00:02:25] Dara: There’s a moment and I need to feel, I need to point this out for the benefit of our listeners as well because they won’t have seen it. But Bhav your eyes lit up when he mentioned p-values?
[00:02:34] Bhav: Yeah, no, they definitely did. As soon as he mentioned it, I was like, yes. I think it was p-values and confidence intervals. He mentioned those two things and suddenly I was just like on the edge of my seat.
[00:02:43] Daniel: So if you’re a fan of p-values and confidence intervals, be sure to stick around to the whole conversation with Alex. It’s a fascinating insight into SEO and like we said, very passionate and very knowledgeable about the subject at ASOS. So stick through to the end and you find how Bhav managed to take our rapid fire questions into less than rapid fire, adding some sub points into the mix. Who knows, we might make that a permanent fixture.
[00:03:04] Dara: Enjoy the show. So Alex, welcome to The Measure Pod. Firstly, thank you for agreeing to come on and talk with us today. It’s very good to have you.
[00:03:13] Alex: Thank you, Dara. Thanks for inviting me.
[00:03:14] Dara: To kick things off, we always get people to introduce themselves, so we like to call it people’s origin story. So you can go into as much or as little detail as you like. But just give us a little bit of your background leading up to what it is that you’re doing today in your, in your professional life.
[00:03:29] Alex: Yeah, of course. So I’ll do the short version. I, at a young age found myself in a sort of trainee web development role, mostly doing frontend, but it was at a small company, so you kind of got your hands dirty, doing a bit of everything. And yeah, everything that comes on with frontend dev in sort of 2007, so HTML and CSS and a little bit of JavaScript here and there. Flash was phasing out at that time, so I did sort of dabble at times, but never really committed to it, which was probably fortunate given the way it went. And then a lot of sort of like the design stuff as well on top. So like lots of Photoshop stuff as well at the time. And then part of the job was to track the websites and realised that I was doing a fairly inadequate job at doing that.
[00:04:08] Alex: So sort of like I started to stick my nose into the world of online marketing and started to learn a little bit. And that led me onto a job at an agency called Fresh Egg, where I met Dara for the first time and I was agency side there I think for two and a half years. Moved up to London, joined an agency called Forward 3D, which is now Assembly, I think it’s called. It’s gone through a few, a few name changes, was there for a little bit and then eventually found my way to ASOS. So all of the agency experience up to ASOS was in SEO specifically and working on a range of client sites and started at ASOS as the technical SEO manager and have been there since 2013 and have sort of slowly been climbing way up the greasy ladder.
Topic
[00:04:53] Dara: Well, it’s interesting you see, you mentioned that we know each other, which we do. We worked together at Fresh Egg, which was 10 years ago, which is scary, but let’s just leave that there. We don’t need to expand anymore on that. But when we worked together, you were in the SEO team and I was in the analytics team. So I’m going to ask you probably a very big, wide question just to kind of get us into our conversation today. So back at that time, SEO was a very different game. It’s changed a lot in that 10 years. So can you give us your, kind of, in a nutshell, take on, maybe summarise some of the key changes over that time period and then give it a little bit of a kind of state of play of today, today’s SEO world.
[00:05:31] Alex: Yeah, sure. So agency side, when I started at Fresh Egg and Fresh Egg was just implementing best practice at the time, right? It is in hindsight what I would describe now as the wild west of the internet. And I went through a pretty tough learning curve at Fresh Egg because the work we were doing to boost rankings for clients’ websites, I didn’t really understand it or see how it was adding value. And a lot of the content that was produced was going onto junk websites that no one was ever going to see. And it was created simply to get a link onto a website. We did lots of good work as well, like lots of like long-term, like really good quality building. And I would say like my success in the SEO industry is completely down to the guys who set up the agency and sort of imparted all the skills and knowledge that I have today. So lots of really good technical work, lots of really good sort of long-term thinking around proper content for websites as well. And lots of like building really good relationships with clients as well.
[00:06:27] Alex: From when I was there, their retention, client retention was, was really, really strong. And I’ve no doubt it’s any different now. And a lot of that technical foundation I think still exists is still still true today, as true today as it ever was. But the big change certainly is the link piece. So in SEO you kind of think of three pillars. You think of technical SEO content, and the link piece, link citation piece, that last pillar is, is the area where it’s changed the most. You can no longer kind of outsource and spam your way to the top of Google search results. You now have to be really sort of methodical and critical in the way in which you, you go about gaining links to a website and at ASOS now we’re very lucky that as a big business, we have to do very little, almost none of that really because we have links coming to the website naturally every day. We have a very large affiliate program that is generating that kind of activity anyway. Lots of press, every time our technology team releases a big product that is accompanied by a press release and that generates link equity as well.
[00:07:20] Alex: And although a lot of that isn’t, isn’t optimised really critically for SEO, so it’s not like we’re not linking to primary pages that we want to rank. We do a fairly good job of sort of dispersing that equity around the website properly as an SEO team through internal linking. And then really, I think from 2010 to today where I am, a lot of the content theory and a lot of the technical theory still exists and it’s still true stay as it was back then. And the technical stuff is really just sort of augmented and grown as the different development techniques have changed over time and content now is really, it’s quite an exciting time for content because it’s the first time where people will start to think about using generative AI to properly start to create content for the web and on the surface at the moment, do a pretty good job of it. And there are some really good companies, small companies out there now who are doing this at scale and have a product that you can kind of plug into your website and produce fairly good content for, if not great content for a product page, for example to describe a product from an image. I think over the next sort of five years, that area will start to accelerate quite quickly.
[00:08:18] Dara: I’m sure we’re going to dig into that kind of generative content quite a bit more later on. But just before we, before we get too far into things, I wonder if the skillset has changed quite dramatically over that time as well, whereas at that point in time, it would’ve been the fundamentals of onsite SEO and maybe some of these, let’s call them maybe, although at the time they were legitimate, they turned out to be kind of then frowned upon by Google. So what’s the kind of skill set look like now for somebody who’s working in SEO and, and including more broadly. So obviously we’re going to come at this from a measurement angle on this podcast. So what are the kind of extra skills that you need to have outside of just the core SEO skills themselves these days?
[00:08:59] Alex: So I think the extra skill, I’ll start the extra skills piece first, in a big organisation like ASOS, you have to be quite self-sufficient and as a team, and my team are very data literate. So everyone is empowered to understand our analytics platform, how to access the data, how to scrutinise it and analyse it properly, how to retrieve it from the platform properly and all the rest of it.
[00:09:19] Alex: Doing things at scale is increasingly becoming challenging, more challenging. Like when you’re talking about a company that has billions of page views a year to kind of get to the things that you need to see over a long time period. Particularly if you’re doing any kind of forecasting work or regression work, you need a lot of data. So to get to that data, we’re having to upskill ourselves a lot more around languages like SQL and analytics tools, particularly Python, to be able to sort of, sort of look at that data properly and to be able to work with it properly. Because Excel now is just, it’s clunky for sort of the volume of data that we’re working with. That, I think in terms of the analytics piece, everyone is kind of forced, I think now to become a, like a lot more involved, a lot more aware, a lot more capable with data, which I think is, is really exciting, but also a little bit dangerous at times.
[00:10:06] Alex: In terms of the skillset you need. I’m not comparing apples and apples because I was agency side and now I’m in-house. I think in-house, you always need to be a good relationship builder and being able to win over people, being able to tell a story, being able to sell the thing that you’re trying to sell, the project that you’re trying to get pushed through, a development pipeline or a content pipeline or sort of forced into a content calendar. You know, you really need to be able to advocate for yourself really well, really strongly. I think from a technical standpoint as well, you need to be able to talk with the language of a development team now, more than ever I think.
[00:10:36] Alex: Especially as we find that people in products have a varying level of knowledge to SEO and if you get someone who’s good, great, but more often than not, we find that you have a product person you’re working with who has less than a base knowledge of SEO. So actually being able to serve as your own product person as an SEO now I think is, is more important than ever. And I think that’s true agency side because if you are at an agency working with a, maybe just like a generic digital marketing manager who looks after SEO as a number of other channels, you need to be able to go and have those conversations directly with the development team or a, you know, a head of tech somewhere who is responsible for a prioritisation queue and getting work done.
[00:11:15] Bhav: Alex, just on this, how do you, because I’ve worked quite extensively with product managers and engineering managers having come from the product analytics space. Do they get SEO, like, how hard do you have to work to get buy-in for SEO and making sure that all of their user stories and product stories include an SEO angle to, like, do you have to work hard to, you know, for these teams to buy into this whole concept of SEO?
[00:11:36] Alex: Yeah, you do. And you, you will find you’ll get people who are very passionate and who are unable to sort of create a story and sell it into the business for you. You’ll find that you get people who aren’t that particularly passionate about SEO, but are able to do it and sell the story. And so it is really like the skill in being a great product person I think is really clear. And being able to pull the right levers at the right time within the context of a technical dev queue and all the rest of it is, is really key. I think this is just a general point around those businesses. I think SEO for a lot of people is a bit of a black box. And it really shouldn’t be because you, it is really measurable and we have more data and more data sources, I think, than most traditional marketing channels.
[00:12:14] Alex: So you can unpick it, you can scrutinise it really heavily. That’s despite not having any keyword, like any credible keyword level data, I say credible, you still get stuff out of Search Console, but it’s limited. Whereas, you know in GA (Google Analytics), in Adobe Analytics now you see no keyword information at all, it’s been like that for a long time. So it shouldn’t be, it shouldn’t be a black box and a business should be able to understand the value of it. It should be able to understand the ROI of it and it should be able to prioritise the effort accordingly. But it really varies, it really varies with who you’re working on, the part of the platform that product person is working on, the other things you’re competing against within the business as well as part of that platform.
[00:12:48] Bhav: I think with SEO and the, you know, calculating that concept of an ROI for an SEO, it’s in some ways it’s kind of like product. How do you determine the ROI of your product investment? Because it’s not like marketing where you have a direct channel spend where you can say, I spent X amount on PPC, X amount on Facebook or whatever. And I guess the bigger challenge is that, SEO is not a short-term quick turnaround type thing. So when you are doing, you know, when you are trying to get buy-in for SEO and you are trying to get product managers, engineering managers to prioritise the things you need, you know, do you get asked like, what is the ROI going to look like directly, or how quickly are they going to see this? Like, for me, SEO is such a long-term gain thing that it’s hard to get people to buy into this concept that what we do now may yield some value a bit further down the line.
[00:13:33] Alex: It’s really hard, it’s really hard. And quite often we’ll find ourselves having to do an MVP or proof of concept. So doing a small test to help people understand what it is we’re trying to deliver and the value that it will deliver. All of our projects will start with a business case and a user story and it will go into Confluence and all of that starts with data. For ASOS, Adobe Analytics is the platform of choice and it will, we’ll always start with Adobe going in and understanding the area of the site that we want to improve, its current performance and what it could look like. And then we’ll have to start augmenting that data with ranking with search volume, with click through rate data, joining it all together, and building what we think is a likely scenario that will come out the other end of it. Sometimes those business cases can be small and sometimes they can be huge and they can be really competitive. Like, as long as you’re doing the due diligence on your data and building the story correctly and building your model correctly, more often than not, it will yield a pretty good result for you.
[00:14:27] Daniel: So one of the things you mentioned just now, Alex, was around how things have changed right? In the last 10 years. Especially the big one, which is the removal of keyword-level data. And I was there during that period of time and there was lots of head scratching and confused people around, like, how are we going to change? In hindsight it’s changed an awful lot. You mentioned about joining your data to the analytics data, your SERP data, to your analytics data, and that’s now traditionally, or at least commonly done by a page, right? So what, what’s the role of the keyword level information you can get? What’s the role of that now within, I suppose, modern SEO but in a company such as ASOS, in terms of the scale, like what kind of role does keyword-level data do when you can’t directly tie it to sort of performance metrics and maybe other kind of directly measurable KPIs you might have?
[00:15:05] Alex: It kind of serves as the basis for your, for how you start to think about growth and calculate growth. So you still need to have that keyword-level data so Google Search Console is a great place to go to understand what’s happening on your site currently, but to understand potential for the future, there are lots of really great third party tools out there which will provide that. There are some really specific SEO industry ones like SEMrush or Searchmetrics. There are some sort of broader ones like SimilarWeb, which is a much, much broader sort of analytics tool now that has some really good data in it as well. And you need to understand the search volume, the demand for the keywords that you will want to target as part of the project that you’re working on.
[00:15:41] Alex: So if it’s something new, if you’re expanding your website, if you’re starting to create more viable landing pages to land new traffic via new keywords, you all need to go away and do that keyword research, understand the volume, understand the potential, and then start to work from there to try and forecast what you think you can get from that. And we’ll always do it through varying degrees of success. So 30% success rate, 50% success rate, 70% success rate, build three different models, three different forecasts, and start to present that to the business to try and let the business know, like the upper and lower bounds, for what we think the project will deliver. And usually if there is a decent level of scepticism from the engineers, which there always is ASOS, we will have to go through some kind of like MVP project and sort of like stand it up in a fairly manual way, so as to start to deliver results. It’s a long play, so you have to wait, you have to sometimes give it two or three months for that test to run, for it to build momentum, for it to start to deliver results before you can say critically whether it has had a success or not, whether it is, you know, it’s going to deliver on what you think it will.
[00:16:37] Alex: So it always plays a role from the context of like site expansion. And it always plays a role in terms of fashion in particular trends as well. So we will have a, you know, a set number of category pages that are described in a certain way, but that doesn’t mean that the product will be described in that way in 10 years time or in 5 years time, or even next year. So you have to be aware of the keywords and the intent and how that’s changing over time. We also find ASOS is a multinational site, and we also have to go and look at the keywords across different countries as well. So a good example at ASOS is we have a category called Mum Jeans, which in British English, American English, Australian English makes perfect sense, but in French, in German and Spanish and Italian means absolutely nothing.
[00:17:23] Alex: You have to sort of like go back and start to think about how to describe that product for that market, for that customer. And then sort of almost start your keyword research from scratch again. Colour is another really interesting one, which doesn’t translate particularly well across all markets. So I’ll use Russia as an example. Even though our Russian site is paused at the moment, in Russia, they don’t have one word for blue. It’s always given a shade. It’s either light blue or dark blue. So when we’re talking about like jeans as a really good example for the blue colour in particular. We have to think about, again, if for our product pages, if we are describing a product in a certain way for the UK market, does that make sense across every market? And how do we change our translation approach for each of those markets as well?
[00:18:02] Daniel: Localization a huge part of all this, I was actually going to ask about that if it didn’t organically come up, especially when it comes to the data collection, not just from an SEO marketing perspective, I can imagine each channel has its own nuance with individual profiles, accounts and all that, all that jazz. But when it comes to the analytics, how do you structure data when it’s so multifaceted, you’ve got different territories and different currencies, you’ve got different combinations of both, right? So I can be browsing in a different store with a different currency and I can have so many different combinations of that.
[00:18:29] Daniel: So what, what kind of process do you go about to kind of filter and segment by that? Is one of them dropped? Do you capture both of those? Are there kind of standardised ways of looking at that kind of data from your perspective? How does that work? Because I know that probably everyone’s, or everyone that’s got a multinational website has got their own unique spin on this and they’ve probably seen various different combinations of it.
[00:18:47] Alex: Yeah, so more or less everything is tracked to at ASOS. We have really good data granularity so you can sort of slice and dice the data to get at what you want to look at and what you want to see. So if you want to look at, at visits from a country, regardless of the store that they’ve entered the site on, regardless of the language that they’re looking at or the currency they’re selected, you can do that. If you want to go to a store level and look specifically at particular currency, particular language, you can do that as well. For SEO, it’s a lot more simple because Google will crawl and index a website, a URL structure, and as such, we’re always interested in that, that’s always sort of like our basis to start analysis.
[00:19:25] Alex: So we always want to go to the URL structure to start building out an analysis. And at ASOS, although we have 256 different, like what we call like store keys, we only present 12 distinct URL structures to Google. For myself, I can tell you how many people from Austria came to ASOS.com or to ASOS.de sorry, ASOS.com/de as it is now. But I can’t tell you specifically much more about their experience other than how they’re interfacing with those sites. So yeah, we always go to the URL structure and page type as well. So we’re quite often bundle by page type also. And then we’ll also build lots of segmentation to get to the sort of data that we need, that we really need.
[00:20:03] Alex: So quite often if you’re doing a fairly simplistic piece of analysis, you’ll just do it within the Adobe sort of visualisation tool workspace, and you’ll just build out whatever it is you need using the dimensions, metrics that are available. And if there’s anything more specific you need, then you’ll just start to build out segmentation for it properly. And at ASOS for myself and my team, that’s something we’re all empowered to do and something we are taught and trained how to do, because it’s really important that we have that literacy within the team to do that. But that’s not common across ASOS across all teams. More often than not, people will be reliant on someone in the analytics team there to do that job for them properly.
[00:20:35] Dara: And in terms of like interplay with other teams, you’re talking about like how you might take on a lot of the, at least some of the kind of data work yourselves, but then there is a separate analytics team, so it’s kind of fairly clear how those two would interplay. What about with other teams where, I mean, where does your responsibility stop within the SEO team in terms of you’re going to get traffic to the site, is it also your responsibility to then convert that traffic? And if not, then how does that work in terms of trying to get buy-in across other teams where maybe you’re working with UX team or merchandising teams?
[00:21:07] Alex: In SEO It’s not as pronounced as it is across the paid marketing teams. Obviously with a paid marketing team, you can create audiences, create cohorts. You can really try and laser target the person you are talking to and there is an element of flex there in terms of, in terms of trying to improve your conversion rate by targeting the right people at the right time. In SEO we don’t have that laser target ability. It’s a lot more blanket bombing, which is a bad analogy, you know, it is at scale. We are aware of the conversion rates for the channel and how they change. And we have, again, coming back to products, we have product people who are responsible for different areas of the site who are product owners for each individual area, product managers, they’re all tasked to try and improve the conversion rate for those areas of the site as well.
[00:21:55] Alex: If we are doing a piece of competitor analysis and we see something that looks to be quite a distinct difference between us and one of our competitors. That is a piece of insight knowledge that we might take to a product person and say, we’ve seen this. We think. Like our hypothesis is this around a particular page layout, or the way they’ve structured their data or on the page or whatever it is. More often than not for SEO it will be to do with UX, UI and the concepts that, you know, bounce rate is a metric that Google looks at to understand the quality of a page and the validity of that page within a search result, even though they don’t say yes that is something that we look at. There is a pretty well known belief that is something that is concerning.
[00:22:33] Alex: So we may look at our page, a competitor’s page and we may see something that is, that is really sticky, that we think the customer is going to, like, interact with and they’re going to go deeper into the site with, and that’s something that we may take to a product person and show them to that. But I’m not tasked specifically to drive traffic that is going to convert. There’s always an element of that you know, you always want to make sure that the content that you’re putting on the page is targeted at the right people at the right time, it’s the right intent. You’re trying to serve the right content based on the query that a person is using or that you’re optimising to as it is. So you are always mindful of that, but how you measure it, how you affect it, it’s a lot more grey, I think.
[00:23:09] Bhav: Does your ability to optimise the page impact SEO rankings? I used to work for a travel company and we used to want to do A/B tests on our landing pages. Now I think, from what I understand I don’t know if Google’s views on this have changed is that you can run an A/B test on your landing page provided you let Google know that it’s a test or, or let them know somehow, but you’re not supposed to run it for a long time. Therefore, the variation of that page doesn’t accidentally get indexed over like what the original page should be like. Can you still do that? I mean, I don’t know how, what Google’s views are on this.
[00:23:43] Alex: If you are testing a page or two different experiences on distinct URLs, then that might cause some issues, but probably not. What you want to be really careful with is, is showing, essentially showing bots a different experience to the user. Now Google is not just crawling and indexing pages, it’s also rendering a page and you want to make sure that what Google is seeing within the source and as part of its crawl is also the same as, or marries up and matches quite closely to what the user is seeing. There is flex, I think within that to do testing properly.
[00:24:12] Alex: You know, at ASOS we use Optimizely as our testing tool. And we don’t, as an SEO team, we don’t have any kind of hard and fast requirements on what the team can, can’t do, in terms of like testing elements. Unless they want to do something really big and crazy, in which case we might want to get involved. But in general, like day-to-day testing of elements and parts of the page we’re pretty okay with that if there is a larger overhaul of a complete template. So, for example, if we were going to overhaul our homepage completely, then that is completely something that we’d want to get involved with and make sure that, that number one it was optimised properly from the start, but number two, that it was being done in the right way as well. And you’d be surprised at these big businesses how key people are to start calling us to get stuff done quicker. You always need to make sure that they are doing it in the right way and that they are using those test ecosystems in the right way.
[00:24:56] Bhav: That makes sense. I’m curious, do the words homepage redesign cause you to lose sleep at night? And how often have you heard them over the course of your career?
[00:25:04] Alex: I tell you what, there was one big redesign of the ASOS website around 2017 and we did a lot of work for it from an SEO standpoint. I mean, we had sort of over, over 60 separate recommendations, documents, built out for each section of the website that we had really hard and fast, clear recommendations, requirements for to meet SEO best practice. And like any business, timing starts to get squeezed and people start to cut corners and you sort of end up in a place where you need to fix forward. And I think that’s quite common, right? In terms of a product mindset, shipping something early, learning fast, iterating fast. I think that is all generally accepted sort of practice from a product standpoint. It’s better to get something out the door and start learning, and start iterating and start improving than sitting on something for years and years and hoping that’ll be perfect at some point because it never will be, there’ll always be something you need to add on.
[00:25:59] Alex: And for big websites in particular, the rate of work can be quite slow and before you know it, something else has happened somewhere. So California might introduce new accessibility guidelines, which mean that you have to come back and rework a whole bunch of stuff. Like the EU might do something around data privacy again, and you have to like come back and revisit a whole bunch of stuff. So it’s better just to get, I think, just to get a, what is a close to sort of 80% good and then iterate forward on that.
[00:26:23] Bhav: You mentioned something earlier, sorry guys, I want to go back to it. And I just want to call out, like, I rarely hear this so it’s music to my ears. And you mentioned confidence intervals when you’re making projections and giving a forecast moving forward. It’s one number, and I love the fact that you are using confidence intervals. So I just wanted to give you like a, like a kudos on that one just because I made a note of it and I really wanted to just say like, it’s amazing to hear people talking about confidence intervals as opposed to the, like, just the average or something like that.
[00:26:48] Alex: SEO forecasting is really hard and I have to do this for, so broadly speaking, non-paid channels as well at ASOS from time to time. And if you’re doing that within, if you’re forecasting for a paid channel, it can be quite black and white. You know your cost per click, you know, a whole bunch of data points where you can sort of forecast forward quite accurately. But for non-paid channels, it can be a lot, a lot harder. For me, my go-to approach is always get as much historical data as I can. Try and make sure it’s as clean as it is or it can be. And sort of flatten out any anomalies that might exist within that data and then sort of like start to try and predict and forecast forward through that.
[00:27:21] Alex: And sometimes that is something as basic as using the Excel function within Excel and building forwards, which is exponential smoothing, is the methodology within that builtin function. Sometimes for the bigger, heavier stuff, we’ll use Python to do that as well. Because it’s an inexact science, because you can’t guarantee results because it takes time and Google is liable to change. I always want to give a range to say how confident am I in these results? And, you know, we’ll use r-squared values and p-values as well to try and make sure that people are happy or they like to give them a level of confidence in our numbers as well. And how accurate our data set is in the forecast as well, to try and at least give a little bit of confidence around that. With SEO I think you always have to give a range, yeah.
[00:28:00] Bhav: I like you Alex. Like, I love hearing people doing these type of things properly because often people just like finger in the air type it, but actually there is an opportunity to put a little bit of science behind it. You are right, there will be a tonne of things that are outside of your control. Google releases a change, which impacts the algorithm, which, you know, will affect your sites. But generally if you use your historical data, understand what the seasonality is, take out the anomalies, what you’re left with is the trend and your, and you know, the overall seasonal pattern, which you can use to forecast forward. And it’s usually good enough, like you may want to go down like very heavy Amazon type models for forecasting, but actually like the Excel one coming out of the box is probably going to be good enough to help you make a forecast of the next 12 to 18 months within some level of confidence that you, you know, you could probably be quite happy with. What do you forecast? Is it revenue or traffic?
[00:28:49] Alex: Traffic, because with a brand like ASOS, we have a lot of brand traffic which converts at a much higher rate than what we would describe as our non-brand generic traffic. It’s always traffic that we forecast on and, within the context of SEO, you can’t, with brands, like you can’t optimise towards brand. Like you can’t increase brand demand. So it’s always just sort of sat there as a residual and you can capture it. Whereas, and you can’t, obviously, you can’t optimise towards or you can’t increase seasonality trends or you can’t, you know, increase demand in anything. That’s very much the job of marketing. If they want to try and like, make a product or a moment a thing, they can, you know, spend money and try to do that. But yeah, it’s always traffic and we are aware as much as anyone else that, and to come back to the conversion rate question, we are, as a team, we are not challenged to improve conversion rate for the site. So there will be things that the technology teams are testing, the product teams are testing that will change how traffic converts on our site. And again, that’s kind of outside of our realms of control. So it’s always, for us, it’s always traffic.
[00:29:50] Bhav: And do you have to split between brand and generic? Because you’re right, there’s going to be an element of this, which is largely outside of control, which is how much is the ASOS brand dominating the, you know, your brand keywords. Do you ever get asked like, we understand that brand is what it is. What are you going to do about generics?
[00:30:06] Alex: You have to split them. It’s so important to split them, particularly for a big brand like ASOS. Because if your traffic is going really well and it’s climbing really well, but a lot of that is driven by brand because you’ve been spending a lot of money on like, out-of-home advertising, then you need to be able to tell that story. Not only to explain why traffic for SEO is improving, but also to try and help support the wider marketing team who are delivering that activity, and to try and correlate back the timings and the effects and the spend and what that looks like on SEO as well. And that’s really valuable because as a marketing function that isn’t digital, the teams that are running the big branded stuff, they lack data. So if you can tie it back to the stuff they’re doing, that’s really helpful and really beneficial. It benefits the digital marketing teams more broadly because it’s cheaper to buy branded traffic than it is to buy non-brand traffic. In terms of a defensible moat, like our brand is our own, you know, and like we can mop up as much brand demand as we possibly can, it’s not a problem. So it’s like, it’s hugely defensible. You have to split and I always split, I always give a view on what I call branded traffic, which at ASOS is simply just three pages.
[00:31:07] Alex: We have our generic homepage and we have a men’s and women’s floor landing page, and then everything else. But generally, I’m looking at the pages we optimise, which are category pages, product pages. Maybe some help stuff, maybe some size guide stuff, but it’s the category and the product that drives the volume.
[00:31:20] Dara: What do you do, Alex? And actually maybe Bhav, this could be a question to you as well. I’d be keen to get your thoughts on this. If you don’t have historical data, if it’s a brand new product category and it’s quite different from anything else that exists on the site, how do you, how do you handle that? Do you forecast, and if so, how might you approach that?
[00:31:35] Alex: So I have a really good example where a chap called John Mooney came to me in 2014, 2015. He was our head of menswear design at the time. And he was like, Alex, I need a page for drop crotch trousers. And I said to John, no, you don’t. You do not need a page for drop crotch trousers because that is not a thing and Google tells me that’s not a thing, so you do not need that thing. But John was right, we did need a page for drop crop trousers because Kanye West started wearing them and everyone else saw that and then we needed a page for drop crotch trousers pretty quickly. So for a company like ASOS, you have to lean on the expertise that exists beyond just what the data that, that you have access to tells you, particularly for new and emerging trends like the retail teams and you know, the buyers and merchandisers that we work with, the insight that they have, the knowledge that they have of their, of their domain is really valuable and you can’t, you can’t dismiss it.
[00:32:27] Alex: So, you know, I have people in my team who spend a meaningful chunk of their time going to trend presentations and trying to understand what’s going to be big for the next season, and trying to figure out if there are things there that we need to be able to target because we think it’s going to be a, a big bet or not. So sometimes it is a little bit of sticking your head out of your own world and trying to understand what’s going on and trying to see the opportunity and talking to these people and understanding, sometimes it’s about understanding, you know, how much product are they going to buy. And it’s as simple as saying, okay, if they’re going to be buying like X amount of products styles across and the depth is going to be Y like sometimes that just means that we need to have a category to support it and you start to target those terms properly. Sometimes it is around looking at some of the emerging stuff that Google sees, but less so it, it’s Google’s a lot slower to see that stuff than what you would see now through your own buying teams or even through a social team and what they’re picking up through social listening, not that that is an avenue that, that we are, I think, tapping into as much. I think we’re very much more sort of keen to talk to our retail and buyers and merchandising teams to make sure we understand what they’re doing and what they’re seeing. But certainly something you could pick up through social listening as well I’m sure.
[00:33:36] Bhav: I don’t have an answer for that one, Dara. I think what I’ll do is I’ll just share a small anecdote on something that happened recently. I was on TikTok or the TikTok guys, as I say, just to wind up my nephew and nieces. So I was on the TikTok and I saw this video, which I thought was super interesting. And it was some of these, some podcasts like those American, Joe Rogan type podcast hosts. And he had a guest on, he was talking about this phenomenon called the Region Beta Paradox. And I listened to it, it was only a short video on the TikTok, it was like two minutes. So I quickly watched it, found it really interesting, went and read the scientific paper, fell in love with it. And I thought about how this applies to the product space, and wrote a blog post. I quickly checked on Google’s search trends and saw that it was slowly starting to peak in terms of its popularity. And I assume a lot of people were listening to this podcast and, you know, were doing something similar. So I quickly whacked out a blog post off the back of it, and that ranks number three on this term, which I hope is growing in terms of popularity. So it’s a silly anecdote, sorry, but I just wanted to say it.
[00:34:32] Daniel: And a sly plug to your blog.
[00:34:35] Bhav: A sly plug to my blog.
[00:34:37] Dara: And you’re flexing your own SEO muscles.
[00:34:39] Bhav: No, this is a one-off. I only have two stories of SEO success. That one, and the time that Toys“R”Us were closing. I just happened to be coincidentally doing an analysis of how terrible my experience was with Toys“R”Us and versus how good my experience was with the other toy store, I forgot the name of it, Smiths, yeah. And Smiths was great and I just did a quick Toys“R”Us versus Smiths analysis put onto my blog post and I get like hundreds of page views on that per week, even now.
[00:35:04] Alex: The verses content types are really strong. Like you can, if you have a tool that you are working on really heavily or two that you’re maybe split between, like maybe you’re working with a client, two different clients and one works with one tool and one works with another, doing a comparison post like that is a really good way of like gaining traffic. A versus, or a comparison is a good content type.
[00:35:23] Daniel: With a brand such as ASOS, with a brand that’s kind of predominantly, I can only imagine mobile focused with apps that have got tens of millions of downloads, there’s probably a, a larger percentage of the traffic or the volume of purchases at very least going on mobile or on App, and things like social media being a heavily dominating force within the kind of fashion space as well. So I suppose, what is the role of SEO and how much of that is now shared or focused or has the attention been split and shared between, you know, the social media platforms in terms of the organic reach over there and the app store reach and the app store optimization, which is a whole other concept entirely, but how does that live within your world? Within ASOS or within your career? How do you see kind of SEO kind of playing its part within those kind of component or complimentary pieces?
[00:36:05] Alex: It’s really interesting, I think, and it’s evolving and I think it will continue to change into the future as well. So, at the moment, I would say SEO at ASOS still plays a very, sort of much traditional role that, like the same sort of role that it has played for the last 10 years, 20 years within any ecommerce business. But we do start to see some, particularly in the younger generations, some shifts in the way they search, the way they use the internet, the way they’re, you know, they’re discovering products. And we also see shifts in the way platforms approach, the way in which they want to retain attention in their own ecosystem, within their own walled garden, TikTok being a really good example of that. On TikTok, there is an influencer, she just does like videos of her strutting in clothes basically, she’ll walk through a door and approaches the camera, she’s got a smile on her face. And we see in our, in our keyword-level data, in Google Search Console, we will see people copy a comment on her TikTok video that includes an ASOS reference and a product code, and enter our site through Google search so, because there’s at the moment, there’s no way that I’m aware of to get directly from TikTok to a page outside of their ecosystem.
[00:37:09] Alex: We see people like copying and pasting content out of TikTok, putting it into Google search, and then coming to our website and when we see this, it’s, it’s not small numbers, it’s pretty substantial, like enough for it to make an impact, and for us to spot it and to be able to see it. So like we see at the moment people are using TikTok for discovery. They’re using Instagram for discovery and Pinterest and wherever else. Each of these platforms has different ways to allow a user to navigate in and out of their platform depending on their own rules and what they want, but we still see people where they need to, reverting to Google to search for things that they’ve seen on a platform.
[00:37:43] Alex: But that I think will change over time, particularly as I think there’s going to be a preference for platforms to keep the retail experience within their own ecosystem. I’m not a TikTok expert, but I think we’re seeing, I think that is a thing within TikTok where they want to keep the retail experience, the checkout experience within TikTok, so you never leave their ecosystem. I’m sure we’ll start to see that across a number of apps as well as we go into the future. So when you are able to discover and check out within a social platform or within an app or whatever it is, I think we’ll start to see that change in how people interact with the web in particular, and there’s a little bit of a, a middle ground at the moment I think we see with Shein where they try and as a company, they try and sort of combine the two things together.
[00:38:23] Alex: So their app is very much around a social experience but you can buy all the stuff that’s on there. So there’s like, the Shein experience now I think is a little bit of a mashup of the two where they’re trying to bring, and doing it very successfully, bring the two wells together. I don’t think it’ll be like a permanent shift, but we’ll certainly start to see a bit of leaching of trade and commerce as it starts to shift over into the platforms a little bit more.
[00:38:43] Dara: I mean, I’m only asking you to speculate here because obviously none of us will know for sure, but how do you think Google might respond to that threat? Because they’re obviously going to be thinking about that and aware it’s, you know, it’s not just the AI threat, but they’ve also got these huge platforms like TikTok, which are effectively stealing their traffic.
[00:39:00] Alex: In my opinion, my opinion only. If there are two tech companies who have already been there and done that with a lot of this stuff, it’s Google and Apple. Like they probably looked at this stuff 10 years ago and made a decision on it and have a view for the future which they want to, which they’re building towards and which they are building towards. When we think about generative AI content. ChatGPT is huge, but it’s database is only updated to, I think it’s 2021. Whereas Google’s Bard will be, they will lean on like their historic technology and for me as an SEO I’m thinking like the Google caffeine update, which was all around like speed of indexation and getting content into the SERP. They’ll already have that kind of stuff, I’m sure plugged in. So Google’s generative AI experience will be, I think a lot more up to date than anyone else’s. And they’ve probably been forced, you know, their hands probably been forced to kind of get going with this a little bit quicker than anyone else and you might see that with shopping.
[00:39:52] Alex: So Google’s like PLA shopping enterprise, their world, that whole world. It’s the biggest product catalogue in the world, and it is completely crowdsourced by everyone who wants to advertise on Google. It is not beyond the realms of possibility, if they haven’t already started doing it, if they haven’t already announced it somewhere, they will start to, to offer a checkout experience within their own ecosystem. And I think that will be, for a big company and probably for lots of small companies, a lot more palatable to sign up to that than it will be to a TikTok or an Instagram or whoever it is simply because of the scale and because of the trust that you place in a company like Google.
[00:40:28] Bhav: I mean, this might seem like at silly point but isn’t the fact that ChatGPT partially or partially owned by Bing going to play a role in itself for ChatGTP to have an up to date record of historicals? Bing, I have been trying to capture Google’s space and like to go into that space for a long time now. And actually acquiring ChatGPT was a really good move because right now Bard isn’t very good, but you’re right, Google aren’t, you know, they’re probably going to continue developing on it in the background and leverage all of their historics, but that’s not to say that ChatGPT doesn’t have that same access level that Bard does and arguably like ChatGPT is a hundred times better right now, in my opinion from what I’ve seen when I’ve compared the two. But they do have all of that historical rankings and all historical data coming from Bing that might make it a serious competitor.
[00:41:17] Alex: Yeah, completely. If you’re asking me to choose one between the other now I’m on ChatGPT. If I’m being lazy or I don’t know how to do something properly in Python, I’m on ChatGPT asking it to great code for me because of the power that it has and the accuracy of it as well. And it’s surprising, it’s surprising how accurate it is. I think when you look at the freshness of content it becomes really important when you are thinking of like politics or money or medicine and healthcare and that whole world. And you know, they tread really carefully around a lot of these things now already. Google who are starting to introduce in the US starting to test a generative AI experience within the SERP have already said they won’t do it for healthcare and for money based queries because they don’t want that system to get it wrong because it’s so sensitive. But yeah, I think at the moment you’re right, ChatGPT is a hundred percent stronger, it will only take Google time to catch up and surpass and then yeah, I’m sure of it. And I say this thinking also that Apple probably did this 10 years ago and were like, no, let’s not do that it’s a bit of a flash in the pan, let’s just focus on something else over here for a bit.
[00:42:18] Daniel: Let’s not forget that Google doesn’t actually have to be the best at something, just the fact that it’s named Google means people will use it. Well, there’s plenty of examples where Google has sort of second rate products or tools and people use them because it’s Google, or they think that they’re better because of, I like to refer to as the Google halo effect, where people just assume Google’s right. I mean, you must have this Alex, when using things like Adobe Analytics alongside Google Analytics. Like people often just assume GA’s, right? And then Adobe is the one you are always validating against, right? And if for no other reason than people have this assumptive approach to Google being correct, or the best or the most up to date or the, the more advanced version of something where quite often you peel back the curtain and it’s always just playing catch up but a lot of people won’t realise that. So yeah, I mean, Bard’s not there yet, but it will get there.
[00:42:59] Bhav: What’s that saying, the first one to market or something like that. There’s an advantage of being the first to market and even though ChatGTP doesn’t have the most freshest content, a lot of the stuff that I personally use it for doesn’t require up to date because I’m, you know, I’m not going to ChatGPT to stay up to date on politics or what’s happening in the world. I’m using it for knowledge development and I think compare that to Google, right? I no longer want to scan through like 10 different websites to find exactly what I need. I’m going straight to ChatGPT, so you know, it might be a case that actually they end up competing in a very different way and ChatGTP takes in, takes in its entirety one element of Google, which was like knowledge and fastness to answers that don’t require the freshest of data. And then Google then maintains and owns everything that is related to shopping and ecommerce and the more commercial space.
[00:43:47] Daniel: Yeah, for sure. Who knows? We’ll see. At the time of recording, we don’t know, but give it a week, couple of weeks and we might.
[00:43:52] Dara: A question on the use of AI, Alex, because my mind, you know, initially obviously jumped to generating content, but I assume there’s use cases especially now with things like ChatGTP-4 and all of the, the kind of, I don’t know what they call them, plugins, or add-ons or whatever the correct term is. But I imagine for, even for keyword research or competitor research, it could be really useful for that, where you feed it in a bunch of competitors and tell it to do a kind of site analysis and come back to you and tell you where it thinks they might be performing well versus badly and maybe create lists of keywords, things like that.
[00:44:25] Alex: I haven’t actually tried that, but maybe I will go away and do that as a job this week. I would say it’s pretty strong for creating content at the moment. You can ask it to create category page contents in the ASOS tone of voice for a particular category and give it a word count, a limit, and it will do a pretty good job of it. The only thing I would say is that, and there’s been lots of studies and research around this, and even our own data science team we have a tech blog at ASOS and they published recently talking about biases within the language. So there’s been lots of study over, like ChatGPT-2 and ChatGPT-3 around like racial bias or antisemitism that creeps into the content that it produces. Our own data science team recently published a piece looking at like LGBTQ+ language and the biases there that exist within the content that it, you know, that it’s creating. So you have to be really careful about it. So I think for content generation, it’s never going to be, for us in particular, it’s never going to be a, like a one and done solution.
[00:45:19] Alex: I’m never going to give at the moment a generative AI tool, like a complete rein to publish content on the site without it having some level of editor review, editorial review and sort of check to make sure that the language is right. I mean, it’s really powerful and in the future it could speed up an awful lot of stuff. There’s some really great companies out there starting up, some of them, like in particular coming out of Y Combinator, who are producing AI based content for product pages where you just give it images and it will produce all the content for you. That sort of implication or that sort of rollout as well, I think is really powerful for ecommerce, particularly at scale, where it is time consuming and costly to produce.
[00:45:55] Alex: Like for ASOS, I think it’s around 80,000 products we have go live a week. To produce product descriptions for 80,000 products every week is a pretty herculean effort. So yeah, so to be able to automate that to a degree could be quite powerful. But certainly in the near future, I can’t see it going live without some kind of human check.
[00:46:14] Dara: Which I guess would be good practice anyway. If anybody writes a piece of content, there should be a, you know, an editorial layer. Yeah, it just made me think, I wonder, and I don’t know if any of you know this, I wonder how good these like ChatGPT and Bard are at translating compared to maybe just the standard like Google translate. Because I’m thinking of you here, Alex, with the localization, and I wonder if some of these large language models will be better at picking up some of the nuance and understanding the intent with the message. Because you know, you get that when you use Google Translate sometimes it just doesn’t quite pick up, it’s literal translation.
[00:46:46] Alex: Having managed the localization team at ASOS for 18 months, I would hazard a guess to say they would be strongly against it in quite a big way. And it is one of the things that I find quite funny is that if you give 10 French people a piece of English content to translate you’ll get 10 different responses. And then if you put them all in a room, they’ll spend two days discussing it. Translation isn’t a perfect, you know, you’re never going to get to a hundred percent accuracy with translation. There is always an element of preference in it. Certainly for people working in that industry, I feel pretty confident in saying that, that there would be a healthy dose of scepticism in letting a machine do all the work for them at the moment.
[00:47:24] Alex: And certainly like for us, with our international SEO efforts and content efforts, we want to do two things, right? We want to have a bunch of keywords that we want to target, which we’ve identified because of search volume and the importance of that. But we also want to have a piece of content that is localised correctly for the audience and have those two things combined. And again, perfectly within the realms of, you know, a machine learning model to do for you. But I think to get it right, to get the tone of voice right, to get the, you know, it right for the audience and for the market and the rest of it I think we’d still want a person to do it.
Rapid fire
[00:47:56] Dara: Okay, this has been fascinating, Alex. We’re going to let you go really soon, but just before we do, we’ve got some rapid fire questions for you. Just to wrap up the conversation and if you think of these in terms of within your kind of within your role or within your sphere, or you could think of them more broadly in terms of kind of SEO and the broader kind of digital marketing industry. Question number one, what is the biggest challenge today that you think will be gone in five years?
[00:48:19] Alex: I guess we kind of touched on it during the conversation, and I think marketers are having to come around to the idea of being able to self-serve with data and combining data sets and I don’t think this is widespread in the industry yet. It’s certainly improving. I certainly think within SEO it’s improving. There’s a lot, a lot more chat within the industry about people using some of the tools and techniques that we’ve described to improve their ability to analyse data. But I think it needs to go further and I think it needs to be like a journey that everyone goes on. Like, certainly in the context of ASOS, I think the SEO team is probably a little bit further ahead than most others in its ability to do that. So yeah, I think that’ll be the biggest challenge is like everyone sort of going on that, on that journey and being able to upskill, take the time to do it as well. And yeah, sort of everyone being able to be their own advocate, be their own source of truth and do it properly as well, I guess.
[00:49:05] Dara: Okay, so assuming that that all happens, then what will be the biggest challenge in five years time?
[00:49:10] Alex: Again, I think coming back to something we already said is we’ll be around like attribution I think, and we haven’t really talked about attribution in this chat, but in the context of checkout existing beyond just your core platform, like what does that look like in the future? And how do you come at that as a measurement company, as an analytics team, as a channel owner? Like how do you start to think about like the different sources of information and the different entry points into what a site could be now for you, whether that is a store experience on a social platform, whether it continues to be your own website, whether it continues to be a store experience within a search engine, and how do you start to sort of tie all that together and then understand the complete journey, right?
[00:49:49] Alex: Because then things like conversion rate changes. Or you have like four or five different understandings of what conversion rate means based on the channel and then where you put your effort or where you put your spend or yeah the areas you want to back based on your audience. I think that’ll become quite a big challenge in the future. You guys have talked about it before, attribution measurements is I think a challenge for a lot of companies. It’s one of these things where there’s no right approach to it, it’s just what is best for you and best for your company. But I think it will become even more complex than it already is.
[00:50:17] Dara: I think you’re right and that actually sounds like a future deep dive podcast, and I’m amazed given you said it didn’t come up in this podcast. It’s one of Dan’s favourite subjects, so I’m really surprised it didn’t. So I think we might have to, we might have to come back to that one and maybe ask you on again, Alex, to get your, get your thoughts on it. Okay rapid fire question number three, what’s one myth that you’d really like to bust?
[00:50:39] Alex: Really selfishly as a channel owner for SEO and I talked about it before it, like, a lot of people have this black box mentality to SEO and there really shouldn’t be this black box, black box mentality to it. It’s hugely measurable. For most companies, it has an insane ROI. If you can unbundle that and show the success and educate as well as you can. Like, it doesn’t have to be snakeskin, snake oil salesman, like peddling their wares at every corner, which is kind of what it feels like I think when you’re on the outside of it, there isn’t anything too difficult in understanding it. I mean, it’s a complex and wide reaching subject, but it isn’t anything that’s beyond the ways of man to understand. So yeah, I certainly think that that would be like the myth that it isn’t a bunch of like snake oil salesmen going around and selling their wares. It is something that you can understand quite granularly and you can track quite granularly and that you can measure quite infinitely and that you can, that you can build upon as well.
[00:51:31] Dara: If you could wave a magic wand and make everybody know one thing, what would it be? It’s similar actually, maybe to the last question. It will, in terms of your answer, so you’ve got to think on the spot here.
[00:51:40] Alex: For the next pound of marketing investment spent, you’re best to put it in SEO. It will deliver far and away beyond most other marketing channels if you do it properly. If you’ve got a good team, a good agency behind you and you are willing to invest in it for the long run as well, because it’s not something that pays overnight. But this is one of the challenges we have, right, is that businesses like marketing spend through Google and Facebook and Instagram, because, you know, it’s money in and money out. They can track it, they can see it pretty much daily what they’re getting out of it, and finance teams love that.
[00:52:09] Dara: Okay, last one. What’s your favourite way to wind down, Alex, when you’re not working?
[00:52:13] Alex: So we have a six month old daughter, so at the moment, anytime she’s asleep, I’m happy if I can just sit down and relax and not have to think about stuff then yeah that’s pretty good. And it’s brilliant like having a kid, this is our first, so having a, having a child is brilliant and spending time with her is brilliant and like life as we know it has changed and has become a lot more family based and a lot more family oriented, which is fantastic. But yeah, to properly unwind and switch my brain off, you know, an hour or two in the evening where I can just sit quietly would be really nice.
[00:52:41] Dara: Alright, well listen, thanks again, Alex. Well I’ll speak for everyone and say this has been a really good chat and we appreciate you coming on The Measure Pod and chatting to us.
[00:52:48] Alex: Thanks for having me, it’s been a real pleasure.
Daniel Perry-Reed
Subscribe to our newsletter:
Further reading
![Meaurelab company values](https://www.measurelab.co.uk/wp-content/uploads/2021/09/Measurelab-Blog-Co-Values_1200x400-Blog.png)
Firing up some remote feedback with My Three Things
![An AI generated image of modern marketing - Measurelab](https://www.measurelab.co.uk/wp-content/uploads/2024/07/DALL·E-2024-07-19-11.26.58-A-humorous-cartoon-style-scene-featuring-a-diverse-group-of-people-looking-confused-in-front-of-a-whiteboard.-The-whiteboard-has-icons-of-clouds-shie.png)
The modernisation of marketing and the role of marketing analytics
![Elevate your data and analytics game using BigQuery - Measurelab](https://www.measurelab.co.uk/wp-content/uploads/2024/07/Analytical_Training_banner_.png)