Guido X Jansen: [00:00:00] Welcome. And thanks for listening to another episode of cafe. And this episode, I talk with Luke Carthy and he will teach us how to use an, the Oklahoma to find onsite or version bottlenecks. And you'll find out why tea and biscuits are a golden combination. My name is Carrie Elsa, and welcome to okay.
What goes, where I show you the behind the scenes of optimization teams and talk with desk freshness about data and human driven optimization, and implementing a culture of experimentation and validation, I think is your missed it. In the previous English episode, I spoke with Morgan lags. Director of operations and certified partner outreach.
At comfort, we spoke about the festinating decent realized management molecules or leprosy. And we ended up talking about how electricity also led to convert, applying something that's called conscious business. By for example, eliminating bias in hiring and becoming carbon positive. You can also listen to it episodes, which is season two, episode 22 on this Roca fate or websites or in the podcast app you're listening with right now, this episode of Siracha is made possible by our partners online dialogue SiteSpect online influence Institute and gopher.com.
Welcome to season two, episode 25. Oh Luke. Thanks so much for joining us. Welcome to the show. And yeah. First question, of course. What makes those mis-cuts so special?
Luke Carthy: [00:01:35] I think the easy questions to ask is what is it that doesn't make biscuit special? they're amazing, right? They're delicious.
They're just go, we'd see, yeah, they're incredible. I'm just, I'm obsessed with them. They're a bit of a problem. They help me put on weight, especially right now. I don't think I've ever spent as much money as I have during quarantine, on biscuits they're on. but yeah, the biscuits. In the UK versus the U S there's always a big argument, whether I'm talking about British biscuits rather than actual American biscuit force.
Guido X Jansen: [00:02:03] since we're talking about UK biscuits, what kind of tea goes best with your biscuits? Does it depend on the biscuit? is it always
Luke Carthy: [00:02:10] not that much of a connoisseur I'll be real with you? so any kind of English breakfast or black teas is incredible to be honest, but a herbal tea and biscuits.
There's never really a good shot. That's all I'm doing.
Guido X Jansen: [00:02:20] That's good to know. when we meet in real life, that's a, I know how to prepare. That's good to know.
Luke Carthy: [00:02:25] Yeah.
Guido X Jansen: [00:02:26] Sure. onto a more a CRO, a related topic, what is your background and how do you, how are you involved with, zero in the browser?
Luke Carthy: [00:02:35] Yeah. So I think many people in the CRO industry or in digital marketing kind of fall upon it. I don't think there's many people who set from the outside of going to college and school and saying, I want to do CRO, It's not really a. One of those things. So for me, started on eBay, where I first set up my first year by business, joining schools, I was about 16 and so on phone lifting and it was quite successful, but there's a lot of competition.
Yeah. And I just did things using therapy, which is like eBay's, AA traps. If you like, what eBay's SEM rush, which helps you to identify key words, opportunities. how you should write your copy images, that sort of thing, to really help improve impressions and sales. And that CRO book was instilled in me from that point, that constant experimentation testing, free delivery, high price versus a lower price and separate delivery costs or.
listings that have a pack of one versus a pack of three and that sort of thing. So it really started there. and then it moved on into, along with me as part of my SEO career, really. So I loved it. I loved the idea of, finding ways to remove bottlenecks or make our process easier.
and it's the X just stuck with me ever since, but that's where it really started.
Guido X Jansen: [00:03:47] so you do a combination right. Of SEO and Sera.
Luke Carthy: [00:03:50] Yeah, definitely.
Guido X Jansen: [00:03:51] Yeah. And that's, that's good to hear at least
Luke Carthy: [00:03:55] interesting mix. It's an interesting mix. I think that the challenges, And I'm sure many people listening can relate whether you're in house agency or a bit of both, Is, Bringing traffic is great and banana well to grow organic performance. This is brilliant, the other side of it that really helps. Yeah. The business has been able to convert as much of that as possible. and I think both of those are hugely complimentary it to the point where normally when I onboard a new client or I'm working on a new project, it's normally CRO I start with, before we look at growing traffic, cause there's nothing worse to increase in traffic and then having a real.
Big problem or inefficiency. That's costing you sales it's much.
Guido X Jansen: [00:04:31] Yeah, it's a bit cliche. They go hand in hand, right? You can't do one without the other. I was working for a company, two, three years ago and they spent like 12 million in Edwards, but they had no idea. And that was mainly lead generation, the website.
And they had no idea how much ROI or revenue they got from that. it's mind blowing out some companies. just fully focused on one half of the equation and then
Luke Carthy: [00:04:53] yeah, we'll get the other,
Guido X Jansen: [00:04:56] so what did you use specialized in specifically in eCommerce? So a trend, a transactional, a website. So w what does that make you tick, on the eCommerce side?
Luke Carthy: [00:05:05] Yeah, I think it's a bit of a double edged sword because on one side there's nowhere to hide, e-commerce is fundamentally how many sales and average order value and products and everything else. But on the opposite side, it's very immediate. If you like an instance to see where you're making a difference, the rather than lead gen sites insurance and that sort of thing, or B2B inquiries, there's a whole funnel there's leads to nurturing to, a pipeline, what would be commerce?
even in a B to B environment, people are placing orders and immediately exchange of money. So yeah, it's great to see that journey and it doesn't matter how long I've been in eCommerce, right? The feeling never gets older of working on something and seeing your first fail or working on something and see him.
improve growth in sales or performance. So that is infectious. I love it. And I'm absolutely obsessed with it. Yeah. but yeah, it goes all the way back to the rebates where I started and it's what I absolutely adore doing.
Guido X Jansen: [00:05:54] That makes sense. A bit more fun, right? If you can directly tie your actions to results.
if you're working on lead gen rep side is usually a couple of steps further where the real revenue, sets. And it's more fun if you can directly, look at the orders or
Luke Carthy: [00:06:09] yeah, exactly.
Guido X Jansen: [00:06:10] Today, or we're talking about, this is how to find Shiro gold using a C SCO crawler. And you did a session on that last week at the digital elite live, right?
Luke Carthy: [00:06:21] Yeah, I did. It was a lot of fun, although there was this, it was really weird because you don't actually see anybody partaken in your event. So you just have to hope. Exactly. Yeah. We're enjoying what you're talking about, but the feedback after the event was great, there was a bunch of questions. but yeah, I, it was my first property.
Did he talk conference? I think. And, it was interesting.
Guido X Jansen: [00:06:38] so what is it about, how do we find Ciro gold using an SEO crawler?
Luke Carthy: [00:06:42] I guess the, to give you the top level as to how this topic even came about, I think. I wanted a topic that would make people think what, like really how, and really instill that.
I need to find out what the answer to they fits and that apparently worked, it was the reason why I ended up getting, getting through to mascot. It was a brilliant idea. So the way how it works is, CRO is not all about AB tests. It's not necessarily about, variances and controls and all that sort of stuff.
sometimes CRO is just things that needs to be done. you find a problem, you address it, you fix it. And so to give you a perfect example, one of the things I've talked about is crawling sites search. So as we know, statistically, a really cool reference and post that CXL made is that a third of all will be commerce users.
On average, we'll use a site search. So it's a fundamental part of a user journey and conversion. So one of the thing that I speak about is crawling your tongue. 10%, whether that's 10,000 or a hundred queries, or even all of them, if you wanted to find out which particular search queries return no results or poor results.
so a perfect example here. there's a retailer here in the UK called Holland and Barrett, and there are, healthcare brand, but if you go to their site or at least he did, when I, to this presentation and you search for Holland and Barrett, sorry, search coronavirus on Holland and Barrett.
There's no results. Fine. Yeah, which I find is massively ironic, considering it's a healthcare provider, They sell and sanitizer, they sell paracetamol and all that sort of stuff that helps alleviate the symptoms or reduce the spreading of it. Yeah. and that's a perfect poignant example of how important it is to make sure your site search is satisfying customer demands as well as possible.
another example there, I looked at best buy. And identify that one of their top searches, according to HRS was online account. So people clearly couldn't find where to find an online account or how to log in. So they searched for that. But on doing that, there was poor results, but no results found. clearly there's an experience opportunity there where someone's searching for online account should maybe be redirected.
So a login page or, maybe a message pops up to say, Hey, are you looking to log in to find your orders, whatever that is. And it's just finding things like that, whether they're immediate commerce queries, immediate, I want to buy queries or just information or a search queries that you can use to help improve experience.
And that's. Something that can be done fairly quickly at scale, using an SEO call. That was one of them.
Guido X Jansen: [00:09:11] Marketing mergers have some suffers and the sheriff for AB testing has been impacted too. If you want to keep that thing to enterprise standards, but safe, 80% on your annual contract, you can consider comfort with their summer release.
You can take advantage of full stack and hybrid features,
Luke Carthy: [00:09:29] strong
Guido X Jansen: [00:09:29] privacy compliance, no blink. And enterprise great security
Luke Carthy: [00:09:35] feel good about your smart
Guido X Jansen: [00:09:37] business decision infest. What you save back in your CRO program, check out www adult's
Luke Carthy: [00:09:43] overbook
Guido X Jansen: [00:09:44] slash 2020. How does a SEO crawl or fit into their process?
just, manually searching for a brand name plus Corona. doesn't sound scalable.
Luke Carthy: [00:09:55] One of the ways in which you could work in a real world situation is to use Google analytics. In this case, you find your top, let's say 10,000 or 5,000 search queries. but every site search will normally have some kind of URL pattern.
So example.com/search/search query. Yeah. So you just create a file where you have that static, part of the URL Ord slash the keyword. And throw those 10, 20,000 your ELLs into, such screaming frog and do it by list mode. then what you can use is something called, as we've said, custom extraction, where you, tag something that's on the page that you want to extract.
So it could be the number of results found. It could be a message that pops up to say no results found when you have no results. And what will happen is you'll get a running list. And this is so difficult to explain without visuals, right? But you'll get a running list of the crawl alongside the data you want to extract.
So to use a very pointed example, you could have example.com/search/um, for an of ours. And you may have that as like your top 10 search in that same row, you would have say no results found in column a. And then column B, you might have something else in column C and so on. So that way at scale, you can very quickly identify what particular search queries are returning, what results and what items and at what prices and that sort of thing.
so it's really quite scalable.
Guido X Jansen: [00:11:19] And, you've already done this. What are the most. yeah. Memorable findings so far.
Luke Carthy: [00:11:24] Okay. Yes. I wish I could give you specifics, but I'm pretty sure I'd get shot by one of my clients,
Guido X Jansen: [00:11:31] but more in general, the types of,
Luke Carthy: [00:11:32] there is one absolute goldmine one, which I was amazed that they couldn't find before I got involved in it.
there's a manufacturer or dessert. There's a, B to B eCommerce client that I have. and they sell one of their most popular brands is something I sell. Now, if you search for that most popular brand, you had really bad results, like really irrelevant results. And the problem was the name of the brand was quite a generic name.
and I'll try my hardest, not to say, but there, it was a super generic name. So by searching for that brand, you've got super generic results. Yeah. and we're talking in the tens of thousands of searches a month. It was a huge volume keyword. So by identifying that this was a problem, we overnight fixed, the relevance and the importance of that particular key word to really improve the performance of that search results.
And almost within sort of 72 hours, we saw an improvement in conversion. We saw a sale through of products that hadn't sold for years, that the business just assumed there wasn't demand for, but actually there was, it was just customers. Couldn't find them online. and then it got buy in if you like to go and invest and spend more money and resources on CRO.
and on that customer experience journey, but it was through finding that real quick, low hanging fruit. And it's something maybe like an hour, like we're not talking a huge project here. That's at month, an hour of research, that really helped to get buy in and to fix one of their core search problems.
Guido X Jansen: [00:12:54] Yeah. So what kind of issues do you usually encounter? Is it. issues in the customer journey that like the big differences between maybe the expectations that are, created during the search phase, then only not Matt's, on the websites or is more like a, just like a Mo more product wise, the actual services or products that the company.
Luke Carthy: [00:13:17] offers. Yeah, it could absolutely be both. It could absolutely be both, but I think more it's towards the latter. to give you a, I guess a, another example, so a lot of companies will have conversations of why is my, why is this particular product or category, dropping in rankings or inversely?
Why is this particular category or set of products now, all of a sudden, really popular. The, what might do a, custom instruction across every product on their site. If I can, I don't have the best sightings and extract things like price, information, stock, status, product description, product title, and then compare it to SEO metrics.
The number of keywords it ranks for and how much organic traffic is getting. And as a result of that, I can start to see if there's a pattern. So normally you can find that if things have dropped is because they've been out of stock. Or if things have increased organically because they're in stock or the prices change, but you can start to satisfy those questions because a lot of people will assume, especially in my experience, working in house in the past before I went into consultancy is, why are we getting this traffic here or why we send less of this online, but no one ever seems to think and look at the.
the core fundamentals for us, which actually we don't have that much stock or we have no stock at all. Or even the prices increased by 25% while other people have it on promotion right now. And just the smooth mechanics of things like that. And the whole kind of, Argument of incremental growth, really comes into play here.
and even though businesses have all the tools to go and find this information themselves, they know their product pricing, they know their descriptions. Normally they're siloed into different seams. So be able to do custom extraction, bring that all together and bring the evidence can really help to, merge and solidify those filings.
Guido X Jansen: [00:14:58] And do you usually use the SCO crawler? is that where you start and then. Figure out where the optimization lies or do you, is it usually that maybe you see something weird in Google analytics and then use your SEO crawl to more or less find the answers or potential answers, for the way of things that you see elsewhere?
Luke Carthy: [00:15:21] Yeah, so it certainly normally starts in GA yeah. that's normally like a good starting point, but equally, sometimes even if I don't find anything in GA that's particularly. cause for concern or even a real good positive that I want to try and bottle up and an echo elsewhere.
sometimes it's really good. Should it run some custom extraction anyway, to see if he can find any patents? So one, real world situation is there was a client of mine that wanted to invest in product descriptions, which comes up quite a lot in the world of eCommerce. Do we go away and spend. money time resource on rewriting, a bunch of product descriptions to see if we can improve sales.
So what normally would happen is I'd go away and again, crawl a whole category or a number of categories and extract their product trips. The, see if I can find a relationship between the length and quality of a product description versus sales performance. so we can try and get an idea whether it's actually going to be a worthwhile investment.
To spend time, money and resource on rewriting, prescriptions for 15,000 products, or not. And normally the answer horribly is it depends, depends on what you sell. Depends on, whether you sell, manufactured products from brands or whether you're responsible for selling your own branded products from the airport.
but it certainly helps to either shape. Or bring the cans argument to our proposal that's been brought forward.
Guido X Jansen: [00:16:40] Yeah, it feels to me like it's a similar, way of working as you, as I do personally with doing user research, you often see some weird things in Google analytics and then I use user research.
To drill down on the issue and find out where their real, problems with the users lies. And it's sometimes also it goes the other way around. So you do a user study, then you find out, Hey, that's a weird let's, look in Google analytics. If we can find some data there that's. Also shows this weird behavior or might explain this weird behavior.
Luke Carthy: [00:17:13] Yeah.
Guido X Jansen: [00:17:13] So yeah, it's a combination of tools usually that, that's yeah. And that gets, it gets you closer to an answer.
Luke Carthy: [00:17:20] Yeah, absolutely. But
Guido X Jansen: [00:17:21] what kind of tools do you use? we sell it as yoke crawler. In general. Are there any specific tools that you'd like to use here?
Luke Carthy: [00:17:27] Frog? I think I've mentioned a few times is a really good GoTo and it's dirt cheap like to use.
but I'm also a big fan of site bald. I think it helps just to add some more context around the data. So scrubbing from is great. If you want to just drill into the specifics, the raw data, but sometimes it can help you out. especially if you're working towards a relatively tight schedule or, people aren't asking for a full, comprehensive audit and it can help me just define, issues that screaming, frog don't won't necessarily identify.
in the CRO space, then I think Hotjar, and heatmaps, and any of the kind of heat map console, there's always a brilliant one, but heat maps, I feel are only as good as the data that you can add to them. maybe when heatmaps first introduced in the early two thousands, Crazy egg hit the scene and everyone was going absolutely insane for this sort of stuff. And it can help paint a picture to see, what things are clicked on. And, we know the story, but where it really gets powerful is if you can leverage that with data and do the analytics with things like custom dimensions.
So I'll give you an example. logged in versus logged out heat maps and seeing if there's a different user journey that people will take on average, across the whole set of users between a logged in experience versus a log, that experience, or even like a new customer versus a repeat customer, or even been able to drill down to someone who.
has been looking around for a good one number of times, what hasn't purchased versus someone who is someone who's constantly purchased. And every time they come to your site, because you want to understand the challenges of people who don't buy separately to the challenges or the winning combinations of things that are happening on the site of customers at Dubai, just looking at heat maps collectively and saying, we get 80% of clicks on this page here.
It is not, Not necessarily going to be helpful, so to segment and really shut down, that heat mapping site can be really quite powerful. Yeah. I think other tools, you mentioned user surveys, Google surveys for all its faults is a great thing to try and use it's relatively low cost.
there's been situations where I've worked in jobs before tried to get, say Hotjar, in the budget and the answer just been well, not yet. And I've just literally put five pounds of my own personal money just to go and run a survey to almost build a case for why we need Hotjar in the first instance.
I really liked that, personally, and then I've had, although not recently, but experienced with tools like Monetate, which I've really enjoyed and the personalization front. And that sort of thing, but they're big scale, enterprise level tools, but I personally really enjoy the small, cheap and cheerful tools.
I just feel to do CRO you don't necessarily need enterprise level staff, right? You don't necessarily need big budgets. And I think that's really important, to make clear
Guido X Jansen: [00:20:07] he has usually, unfortunately it does. There's no one silver bullet tool out there. It's usually a combination of tools as you need to get the insight that you need, right?
Luke Carthy: [00:20:15] Yeah. For over 10
Guido X Jansen: [00:20:17] years now, all mine died. Look at Pfizer's about evidence based conversion optimization with a focus on data and psychology. We see that analyzing data and recognizing customer behavior results in a better online dialogue with your clients and a higher ROI that team of strategists, analysts, psychologists, and UX specialists gathered valuable insights in your online behavior of your visitors.
And together with you optimize the different elements of your zero program through redesigned expert reviews, AB test and behavioral analysis. For more information about their services, go to Owen dialogue. Thanks so much for sharing that with us. and, I think for people that are interested, we can place the link to the, to the recording and, and, and the slides.
Yeah. From the event, if people want to go over them, the final thing I wanted to talk to you about you'll start the webinar share, right? Yeah. Let's talk shop.
Luke Carthy: [00:21:09] Yes. Yeah. So what are you doing there? Okay. So this was a project that was born out of locked down. and I think, earlier you were bored, basically.
Yeah. Yeah. I was bored very early on in the lockdown, I'm on the
Guido X Jansen: [00:21:20] board and you wanted more human content,
Luke Carthy: [00:21:22] literally. that's precisely what it is. And I've always quite a. I can't relax very well. So I always need something to keep me busy. Yeah. And in the beginnings of the pandemic and lockdown, I found it really difficult to switch off and try and imagine that things are going to wind down a little bit.
Yeah. So I thought, New project. but the whole point of this is I'm an extrovert. So it helps me to get some of that out in a webinar series. But also, so there was so many FAQ and so many people that were reaching out and asking me. questions, especially joining, not Danny commerce was screaming up knowledge of eCommerce wasn't necessarily at the same levels.
So thought, Hey, here's an opportunity to try and put some, insightful, informative stuff around e-commerce side that may or may not help people out. and that's what it was born from. So there's two episodes out there so far. The first episode was a question that I must get asked at least once a week.
Like probably more so even now, which is. How do I manage my out of stock products? Do I turn them off? Do I do this? Do I do that All of those questions. so that's the first episode. And then the second and one was into a real world case study of paper checks. they migrated to it. Yeah, the new website back in 2018 and even two years on, they still haven't recovered, their organic losses.
So it was just a look under the hood of what's happening. so first episode kind of a blend of CRO and SEO. And then the second episode is more weighted towards the SEO side. And I might switch it up for episode three. And whenever that man,
Guido X Jansen: [00:22:46] and any topics lined up already,
Luke Carthy: [00:22:48] I've got a bunch in the QR.
I'm not too sure which one I'm going to do next. it could be like a, an interview with somebody just to pick some questions. but I think what I'm probably going to work on next. This is faceted navigation. what specifics about it? I don't know. Cause it's one of those things fascinated.
Now you can talk about it for hours. but yeah, that's probably one that I may a major up until next time.
Guido X Jansen: [00:23:08] The link to that in the show notes, I'm cured. interesting, topics to begin with. So are looking forward to one more of those,
Luke Carthy: [00:23:14] and I've just thought of a really pointed example of where. I'd like a CRO thing, CRO slash book fit,
Guido X Jansen: [00:23:21] throw it in there.
Luke Carthy: [00:23:21] Another example, wanting to throw in the world of CRM. And it's I don't know if it sits more towards CRO or book fix, it sits smack bang between but Aaron messages, no error messages typically built by developers, right? That they are designed to showcase an error when something happens.
It's very rare. at least at a small to medium sized business where you'd have someone dedicated to writing, proper informative copy for error messages. So to give you one example, there's a client that had phone calls, maybe three or four times a week. according to say, Hey, look, I'm at the checkout.
I've gotten a car. I know there's money on this card. I've used it in the supermarket like an hour ago. and it's not working. I can't pay on your site. So the usual argument comes in where you, the customer service raise a ticket. He goes into the it team, the it take a look, they can't replicate it. And it's just this whole back and forth.
yeah. Now with this particular issue, what we identified is let's go away and yeah. and call the error messages that happen at a checkout for any given event. And what we identified is when you, enter invalid card number, The error message that was popping up on the screen was something like error, one, one, but contact sales on phone number, which means nothing.
So no one era wonder below seven. so what I do, David, is I Googled that error message, that output code. And he actually meant when I looked at the Sage pay, API docs are a wonderful Oh seven one one seven, men being fairly common, but so we just changed the copy to say, number, please check your card number and retry.
And the literally within 24 hours, there was like a 30,000 pound uplift in sales, like on a daily basis. Now you could argue that is a book. that is a, the core thing that just needed fixing, but in my opinion, I think it sits more in the world of CRO because it's messaging it's copy it's experience.
and I think the reason why I wanted to bring this particular point up is because sometimes CRO is hidden in the most unusual of places. It doesn't have to be an AB test. It doesn't have to be, user testing. It could just be something that's needling the house, that issue that you can't quite put a finger on, but when you get it can make a world of difference.
Amen. Yeah, absolutely. great addition. No problem. No problem.
Guido X Jansen: [00:25:47] Thank you so much for joining. like I said, more information as you're listening to more information on this topic and, in the webinars series, Lucas started. Can be found in the show notes.
Luke Carthy: [00:25:56] Thanks, Luke. Yeah. Thank you very much, man.
I appreciate it. Good luck.
Guido X Jansen: [00:25:59] And, talk to you soon.
Luke Carthy: [00:26:00] Alright. Thank you.
Guido X Jansen: [00:26:02] Bye bye.
This concludes season two, episode 25 of Ciroc Fe with eCommerce growth consultant, Luke Carthy. Although we started out as a dish podcast, we are putting out more and more English. Content. And if you want to skip all the Dutch gobbledygook through zero.cafe/english, to see an overview of our English episodes and to subscribe, to get notified about new English episodes.
If you're interested in promoting your products or services to them, best Euro specialist in the world, please take a look at zero.cafe/partner to see how we can
Luke Carthy: [00:26:36] collaborate.
Guido X Jansen: [00:26:39] Next week, we'll have another Dutch episode with her alone, all the four, and talking about the master data-driven business that he created at the whole hall.
And the week after that, we'll switch back to English to talk about how you can find the broken areas or most broken areas of your website. It might talk with Connie Salah, CEO of investment, talk to you next week and always be optimizing.
