}
Jojo Furnival
Sitebulb's Patrick Hathaway was joined by an expert panel to discuss ecommerce SEO and growth in 2024:
- Jamie Indigo - Technical SEO Director, Cox Automotive
- Luke Carthy - Ecommerce Consultant
- Ash Young - Founder of Evoluted and carmats.co.uk
- Rachel Anderson - SEO Manager, Weedmaps
ICYMI you can now watch the webinar recording in full below.
Find out about Sitebulb Cloud, a revolutionary new cloud crawler loved by the webinar panel
Discover Sitebulb Cloud
Watch the webinar recording
Here's the webinar recording to watch at your leisure.
Don't forget to subscribe to our YouTube channel to get more videos like this! And find out how Sitebulb's ecommerce crawler can work harder for your team.
Webinar transcript
So it feels like there's been so much flux in the industry with certain volatility over the last few months, and I'd like to know how much you think this has impacted e-commerce specifically.
Jamie Indigo:
Oh, I love it when you ask because I have so many thoughts and opinions. Yes, Google doesn't want to be where you find the mall or the Black Friday deals, they want to be where you shop them. They've made a massive shift from showing category landing pages. Being an e-comm, usually your category name, Shop X, is your bread and butter. If you look at ranking shifts, you're seeing all those lovely category pages suddenly plummeting down, but the individual product pages themselves scoop it on up as Google puts those product graph results in there.
It's a big shift and you have to move from, I want to get you shopping for sleeping bags to really optimising the product detail pages for sleeping bags. What are you putting out in there? What kind of information's available? How much are you giving Google and how quickly and easily can they get it?
Luke Carthy:
Loads of volatility. Yeah, lots of pain and equal in the opposite direction, lots of pleasure for the clients. Like there has been an absolute yo-yo of algorithmic updates, that there is the kind of the classics sit and wait, fix it and wait for your praises. And then we've had... I've got a client that has a competitor that has a horrendous website, almost everything technically wrong with it, but it ranked beautifully for every competitive term and it makes absolutely zero sense, and it's just the two of them combat and kind of go against each other like this. And it's crazy.
But tapping into Jamie's point, yes, Google is greedy, it continually gets greedier, organic space gets smaller, sponsored space gets bigger, and the more competitive the marketplace, the less organic space there is. So it's kind of like a double whammy effect. But yeah, I think what's the latest victim? One of the latest ones, I think, there's an updates to flights to car hire is dominant, not so much anymore, but probably the early 2010s. Worked with a couple of car hire companies and some of these have actually gone out of business as a result of it. They just cannot compete. But yeah, I think the key word here really, no pun intended but it happens, is volatility. It's just crazy at the moment.
Ash Young:
Yeah, I mean we're in a strange niche where Google struggles to tell a difference between our product landing pages and our category pages because they tend to be very similar. But we've seen that our PDPs have done really, really well and we're benefiting from a lot of the changes recently, which is absolutely amazing obviously to be winning for once, it's good to see.
Rachel Anderson:
So the cannabis space is actually kind of interesting. So Google does not like cannabis products and search. And so what we're actually seeing is the category pages are performing better because Google does not want to be showing these PDPs because they're not really part of their shopping experience. The exception being where Google doesn't understand that something is a cannabis product and that's where you see the shopping results.
So that's always really fun to do a search and be like, "Okay, oh, Google clearly does not understand that this is weed," because it's showing a bunch of search queries and just a bunch of really low value sites that are not selling high quality product. So we're actually seeing, we're investing a lot in both PDPs as well as category pages because that seems to be what Google likes.
Jamie Indigo:
I love that Google is the search engine that they didn't know the cookies were medicated.
Luke Carthy:
Brilliant. I love that.
So as a result of any of these sorts of changes, have you just like shifted either your strategy or your recommendations, that your go-tos?
Luke Carthy:
I think it's a really good question. I think it's a horrible one to answer because it's kind of a yes and no. But to kind of sit on the yes, I think as always, and I'm sure it's a bit of a broken record, but your foundations are always going to be the same. So really solid taxonomy with clear distinctive categories, really good hierarchy structure of your parent and child categories, good filter structure, whatever is used, the basic's there. PDP, good descriptions, all that sort of good stuff.
Where it gets a lot more interesting is probably around brand and using that differentiation of things like UGC and reviews and social and social proof. And they're the kind of differentiators, because I think nowadays, and I say nowadays, probably not only the last five years, maybe longer, technical gap in e-commerce is closing. I think maybe 15 years ago, technical probably was quite a good competitive lever because the likes of Shopify, one as established as what they are. But now it seems that Shopify literally powers the backbone of every successful e-commerce brand in somewhere or another. So then the technical thing, the technical kind of component of SEO is taken off of the cards, off of the table. So your differentiating is even smaller, and this is where brand and content and tone of voice and loyalty and all this sort of stuff makes a big difference.
So I think for me personally, I have transitioned more from technical to more conversion, search, analytics and that good old giving customers the ability to completely and wholeheartedly trust the brand. Yeah.
Patrick Hathaway:
That's interesting, actually. So I have seen technical sort of feel like it's kind of done this. And with the rise of JavaScript, it's becoming more and more important again. Possibly that happened after your shift away to the CRO. I'd love to hear, I know both Jamie and Rachel have dealt with JavaScript issues in the past. I suppose, do you agree with Luke's sentiment about the playing field getting level from technical or not?
Rachel Anderson:
Not in the SEO space, it's not a as developed of an industry. We have a lot of really small businesses that are running very small websites that are usually using a pre fab solution that's not Shopify. It's something within the cannabis sector. And most of those are not set up very well for SEO. So there's a lot of opportunity for us to rank because they don't know how to do that. So it's a little bit less sophisticated of a space.
I will say though, in general, the competition, the differentiator seems to be more around what unique content you have, whether that's UGC, whether that's utilising large data to, I don't know, Chewy and Amazon or writing review snippets with AI, that kind of stuff. Being able to provide unique content from other people is more important. But I wouldn't say that technical is less important in my space.
Jamie Indigo:
I think it's yes, and I think it's no because we have an algorithm now. "It depends," my friends. We have an algorithm now that really has updates every day. People are like, "Oh, how'd the last update affect X, Y, or Z?" And it's like, "Which one do you mean?" There's experiments, there's updates, every piece of them all the time. The bigger you are, the more technical matters. The smaller you are, the more content matters. There's a trade-off to it. And that's even different in every vertical. So if you are in the medical space, you're going to have different standards to hold up to than if you're in the drop shipping space.
I think the big thing that came away from... The big thing that equalised a lot of pieces and took a lot of players out of it is we all got trust issues shopping online during the pandemic because there were a lot of products out there that were not what they said on the tin. So Google is like, "We have to keep your trust. We're going to be a little bit stricter about this." And there's definitely babies in that bathwater.
Patrick Hathaway:
Ash, do you see... Obviously you kind of, to some extent, picked your own niche to work in. Was part of that when you chose that, was part of that because you saw a landscape that didn't have this sort of issue or was easy to get access?
Ash Young:
Yeah. So when we started, a lot of the players in our industry were on lots of different e-commerce platforms, and what's happened over the last three, four years is they've all moved to Shopify nearly. So a lot of the players are on Shopify.
So Luke's point has got a lot of credence for me that actually you're not competing so much on a lot of technical things, but actually what you're competing on is how well those businesses can set up their Shopify sites and set them up for best practise. And a lot of businesses just don't have the skillset or the resource or the knowledge to do that. So they go with what's out of the box, and actually they lose a lot as a result of that. And you can have really well-optimised Shopify sites, but obviously you've got to go through the effort, you've got to know what you're doing.
So for me, the big shift has been that, everyone's moved onto the platform that we're using because to a certain degree, we talked about it so much at the start about what we were doing, how we were doing it, and people have literally gone and just, "Oh, we'll use that exact same apps, exact same everything." So it then becomes a very level playing field.
Luke Carthy:
Yeah, that makes a lot of sense.
I don't know about anybody else, but when I read product reviews these days, my AI bullshit radar is on permanent red alert. Is this a problem?
Jamie Indigo:
It's so big of a problem that Mozilla has developed a tool to help you spot fake ones.
Patrick Hathaway:
Really?
Jamie Indigo:
Yes. I will have to find that link.
Luke Carthy:
Yeah, I think it goes back to the whole once you've exhausted something to the point where it's no longer a competitive advantage, it gets abused. So I think, yeah, to your point, reviews is a great one. So this is kind of where you lean more towards things like user generated content, so Q&A, actual pictures. And actually nobody's perfect, business, person, otherwise. So seeing some negative reviews can actually help from a constructive perspective.
So Ash, I've always had a laugh at some of your tweets about some of the absolutely outrageous nonsense that people fire out at you, but I think it's great for someone, when a customer's placed an order, they've kicked off and they've said, "Look, it's crap. Customer service is terrible." And then you've actually found out they've been calling the wrong company and they've left a review on your platform.
If you're responding to that, it kind of shows a lot more sincerity, it shows a lot more genuine. So I think going back to big players like Amazon and eBay, Amazon, I agree, I will not trust reviews whatsoever until I see a good quality picture that looks like it wasn't taken in a studio in UGC, right? So I can get an idea of whether it's true to size, depending on what it is I'm ordering or whether it's reliable or whatever. But yeah, it again goes back to that differentiation. If everybody's got great reviews or everybody's got review saturation, then there is no sense of differentiation anymore. So it just becomes a norm rather than a situation that you're choosing competitors between.
Ash Young:
Yeah, I agree. You've got to have that mix of good and bad reviews. I love bad reviews when they come through because it makes everything seem so much more real. And I read a lot of our reviews and they're not written with AI, but you look at them and go, "He writes that much about CarMax?" Right? "He writes a paragraph of text gushing about how good their CarMax are." And you do worry that it puts people off and people think, "Well, you've made it all up, you've bought all these reviews."
And it is a big thing. I look at sites, I look at reviews and I question whether they're real or not. I look at some of the competitors that have launched and they've launched with a thousand reviews and I'm like, "They're not real because you launched last week."
Yeah, you can see it straight away because you're living and breathing it and you know, well, you've just launched, you've literally just started trading. They're not real reviews and it is a big issue. And trust for us is one of the things that we've always tried to get across, is that we're a real business. We're actually, you can talk to us, you can email us, do whatever you want. And it's more about getting that content and getting that UGC content analyst people to see the product, see people using the product.
We just did a behind the scenes once, people can see how they're actually made, because a lot of people say that we make them in China and it's not what they're ordering that's going to get shipped in. It's going to get shipped in, it's going to be absolute garbage. But we showed how they're made in the UK in some videos and they're doing really, really well, which is good to see.
Patrick Hathaway:
I mean, is video one of the answers to that question? I suppose this is kind of a two-part question really. It's like integrating video, short form video into your strategy, but then also is TikTok and Instagram, are these either channels or threats to SEO? I've got about nine questions in one here. Talk about video, somebody.
Ash Young:
I think video is [inaudible 00:17:24], right? It's harder to fake video effectively at the moment. So I think if you get some really good UGC, it's great on TikTok, on Instagram and social channels, but we're just about to release an update on the site where we actually bring those same UGC videos into the PDPs in the same format so people can actually just have a browse, have a scroll, see the product, see how it fits, see how it works, see how it clips in.
You can answer so many technical questions as well with a 30-second video that actually in text or pictures, it's just really painful to consume. And people don't want those really polished product videos. They actually want to see a short, snappy, maybe a little bit rough around the edges video that's from a real person or could be from a real person.
Luke Carthy:
I like to... You ever heard of Keith Lee, the viral TikTok guy who does food reviews? If not, go and check him out. But some of his early stuff, he sits in a Paw Patrol chair and just reviews takeout spots. They are the most authentic, legit, scrappy bits of video you can ever come across in TikTok. The editing is terrible, but the engagement is brilliant.
And the more recent stuff, it's still incredible, but it's nowhere near as scrappy and as authentic. So I think it's kind of taps into Ash's point that sometimes someone kind of breathing into an iPhone and filming with the wrong camera and then eventually getting... The kind of thing we try to do a video call with your mum, that's the kind of level of content you want to see reviewing your product rather than a high quality studio back-lit...
Don't me wrong, there's nothing wrong with that. It's promotional material, but when it's UGC, the more authentic and real and low rent it is, the more kudos it gets. But I think this is kind of where you start to bridge the gap between SEO and UX and conversion and actually just appeasing the customer. It's less than about technical makeup and more about just giving the customer what they need to make a decision to buy from you or one of your channels or somebody else.
Patrick Hathaway:
And I suppose that's kind of like what the whole TikTok thing is as well. You've just got someone who's stood in front of a camera basically going, "Oh yeah, I've got this thing." So do you-
Jamie Indigo:
Never trust a review more than when they have a pimple patch on. I know that's a real human, I trust that.
Patrick Hathaway:
So how have any of those sort of new social channels, has anyone seen any impact from any of this stuff? I honestly, I don't really know anything about it other than I scroll straight past it.
Luke Carthy:
So there's a brand of... I'm a big fan of cars, right? Big petrolhead. So I'm probably on Ash's radar at some point, but I am a big fan of a brand called Stjärnagloss, which is, they're a car detailing brand of products.
I'm sure Ash had probably seen these guys before, right? But the guy who does the videos is absolutely unreal, but I am pretty sure he's riding a wave of absolute cash from TikTok because all the subsidised orders. So he sells a product for like 12 quid, TikTok Shop is subsidising it and they're selling his product at 3 pounds something for new customers. So TikTok are just giving a discount and they're basically losing money on everyone's first order to try and gain traction. So they're going against the likes of Facebook, they're tapping into live shopping, but some people are making some serious money off that platform. But yeah, I think it's huge.
And then that's got nothing to do with technical SEO whatsoever, and then probably is more towards the space of actually marketplace or social SEO or social optimization. So it's no longer about canonicals and 301s, it's actually about tagging an ad setup and all that sort of stuff. But the whole thing is an entire ecosystem of selling, still the same products you're selling, it's just for the watch out.
Ash Young:
It's almost going back to that face-to-face selling, right? Because he does really well because his personality is-
Luke Carthy:
Electric.
Ash Young:
... huge and it comes across in all his videos and you want to buy from him and you believe in his products. And that's what it goes back to, it goes less about the boring technical side of things and more about actually how you might have bought from someone 20 years ago.
Patrick Hathaway:
We can't be calling technical things boring.
Ash Young:
Sorry.
Patrick Hathaway:
Outrageous.
Luke Carthy:
Boring, mate. Come on.
Patrick Hathaway:
I've got some technical questions so I'm going to move on to one of them, to the boring stuff. We were talking a bit before about how some of the technical advancements have levelled the playing field a little bit.
I want to talk about Core Web Vitals because of all the clients I speak with, I would say that e-commerce businesses probably tend to be the ones most focused on Core Web Vitals... would you agree that that is a trend? Is that something that e-commerce care about more?
Jamie Indigo:
Yeah, no, there's 85 tracking marketing pixels in the head of every page and it shows.
Rachel Anderson:
You got to think about crawl budget. With these e-commerce sites, they're usually quite large, not all the time, but most of the time they're quite large and Google has a certain amount of resources that they're going to dedicate to crawling, rendering and indexing your website. And so if you have a faster response time, they're going to be able to get to more material on your website, which becomes very, very important. So that's where we've seen the biggest impact is when we improve Core Web Vitals on the surface, Google can crawl more things faster and then we see better traffic as a result.
Jamie Indigo:
Also, just humans can use your site. Please, let humans use your site. You don't need to load everything. Just let me see the thing and do the thing. Thank you.
Luke Carthy:
Yeah, big time. Thing I see time and time again in this space, I think the nail was hit on the head a second ago. I'm not too sure if it was yourself, Rachel or Jamie that mentioned it. Probably yourself, Jamie, about 80 tracking pixels, e-commerce is terrible for it, right? You've got your Hotjar thing, then you've probably got your TikTok pixel.
Jamie Indigo:
And they all want to be right after the opening head tag, you say no.
Luke Carthy:
And then you've got three Google Analytics properties, one's probably UA, which doesn't work anymore. Then you've got these ones that are native to Shopify. It just, it's an absolute shit show before you can get to anything in the dark or anything rendering in the page. So first thing, then you make a menu's massive and JavaScript heavy and then you've got massive images and then yada yada, lazy loading and animations and it is exhausting.
Jamie Indigo:
Horizontal carousels.
Luke Carthy:
Yes. All that stuff.
Jamie Indigo:
No one likes those.
Luke Carthy:
And Shopify is horrendous for it. So actually I appreciate now I'm about to give you a complete U-turn on what I said earlier that technical isn't that much of a differentiation, but I think in some cases, more isn't more, right? So yes, apps are important. Yes, functionality is incredibly important in any e-comm in any environment, but actually more isn't always more. Yeah, just be in a situation where you go to a poor area of signal, try and use your website on a mobile phone and see how frustrating it is. And if it infuriates you in the middle of a countryside where your signal is just about 3G, imagine how that's going to impact people. Or in London going around in the tube, signal's a problem.
Again, it goes back to simulating real environment, doesn't it? Like be your user, be your customer. Fast websites are never a bad thing, unless you are in the world of insurance where actually loading results fast makes it seems as if it's dodgy, but that's a whole nother topic. Let's not go to that.
Jamie Indigo:
I mean, for e-commerce, it is constantly a battle between features and functionality. We want new shiny things. We forget that when we add all of the shiny things, we move much slower.
Patrick Hathaway:
Yeah. All right, I've got one question, only one I promise, on crawling. So I guess it really comes into play when you're dealing with really, really big websites. So like millions and millions of pages. And if there's 50,000 pages, yeah fine, you can crawl the whole thing and you can make all your decisions based on that...
But if you've got millions and millions of pages, how do you deal with that? What's the tactic? Are you crawling it all or are you crawling bits of it or a mix and match?
Jamie Indigo:
I feel deeply qualified to take this question. I'm so excited. So say 50,000 pages, I'm like, "Oh, that's so cute." Oh, I missed the days when it was like, "I work in e-com with 70,000 products." When you're trying to crawl a site that large, devil's details matter more than ever. You have to be precise in what you crawl. That crawl config is everything. Love Sitebulb, we got those segments in place, they are my friend. We use our rules right. You're not going to get it all. And if you do get it all, you're going to crash your machine. If you don't crash your machine, it's going to take three hours to load it and then you saw something shiny or a fire happened and you never got back to it.
So what is the minimum that you can crawl to get an accurate sample? Work out some math. You have 6 million products, what type of distribution do you need to have an equal view across them? Really, it works best when we live in a world of having good URL structures, but we adapt. Get crafty, get clever.
Luke Carthy:
It's a really interesting one for me because I do everything I can. My lazy brain kind of works in my favour here, which is if there's a massive crawl... So I've worked with a large automotive brand in the past, which I won't mention. It's coming from my NDA or not, I don't want to get into trouble. But huge website, millions of pages and I didn't want to crawl it. So the way that I handled it was I actually asked some questions, how many categories, how many listings? Kind of give me a number. And if we are in a situation where the total number of URLs that I can see in search console roughly matches the inventory and category count, then we're probably going to be okay and then I can start to crawl. But if we're in a situation where I'm told we've got 100,000 listings, we've got 5,000 categories and there's 82 million URLs, in hell, something's wrong here.
So this is where you kind of dive in. But the crawl for me is a secondary thing that I like to jump into after I've done the basic kind of 101 of what should we expect? And then to jump into Jamie's point, templates are massive. You make one change, you have maybe 2 million URLs using that same template. That's a substantial impact potentially. So you've really got to be careful, but a small sample from each template or from each subfolder or however you're structured is great.
And then to really help bring things to life, and Patrick and I have spoken about this I think during lockdown, I love the cycle content extraction, customer extraction tool where you can just grab stuff on the fly, which can make your life easier. And the templating thing as well, which kind of gives each template a name. But yeah, I think the short tail of it is if it's a massive website, don't just hit start and hope for the best because you'll end up upset, irritable, yeah, you will be in a fire, your laptop's going to crash and it's inefficient. It's a waste of time, resources and everything else and there's a chance to see the ends there to stop all that from happening.
Patrick Hathaway:
There is by the way, another solution if you don't want the laptop to go on fire and that's to use Sitebulb Cloud.
Rachel Anderson:
I think I managed to just crash that again, actually.
Patrick Hathaway:
Oh, god.
Jamie Indigo:
I'm so proud of you. I love that.
Rachel Anderson:
Well, so here's the deal. We're working on internal links. For that, I do need a bigger crawl because I need to know what's going screwy. And so that's what I was trying to do and I tried to do it twice in a row and it didn't like that. But if we're going back to the 50,000, I think 50,000 is a great lazy way to get a feel for where the problems are. So then you can config a real crawl so that then you can do a larger crawl.
That's kind of been my technique and definitely what I did back at Deepcrawl back in the day. So I think that's a really helpful way to get started. If you're like, "Oh, there is a lot of mess here. How do I even identify where those are?" I think that's a nice way to get started. But yeah, in general, you don't need to crawl everything. It's all about what percentage of things you can crawl because you can even crash your cloud crawler.
Jamie Indigo:
Make sure you put exclusions in for the 80 tracking pixels in the head. That'll save time.
Luke Carthy:
Absolutely.
Are there any particular growth strategies you're working on at the moment in e-commerce, stuff you're seeing work particularly well right now?
Ash Young:
I mean not SEO but particularly social side of things is working really, really well for us, getting that content right, letting Facebook do its thing with big campaigns where we're not being too specific and letting the algorithm figure stuff out. Same with PMax in Google Ads, that nearly runs our entire Google Ads account now. We've managed to move it all over and it's really, really good.
Big issue I see coming with e-commerce is the reliance on giving all that information to Facebook and Google is that we're giving all this conversion data. And by today, Google's not tracking a lot of our conversions. We know they're coming through because we can see them, but Google doesn't know they're coming through. So what happens with machine learning algorithm tomorrow when it's missing half the conversions? Does it go, "Well, that's not work. We should stop spending money?" Or does it figure it out and keep going? And that's where I think and certainly evil as we see-
Jamie Indigo:
Google always want you to just spend more money, come on.
Ash Young:
Yeah, or just spend more money. We've seen a lot of issues where clients break their tracking and then Google loses its mind basically because it's now reliant on that data to actually perform. And that's a big issue. But I think for growth, if you get it right, you can just eat all your competitors.
Luke Carthy:
Yeah, it's that whole classic, the 1%. So the kind of north star metric of finding out exactly you're messaging your customer and then scale it, but finding that kind of catalyst, could be life's work for some people depending on the industry you're in.
But yeah, I think this is less of an SEO question irritatingly and it is more, to Ash's point, of a brand exposure, things like email, CRM, lead magnets, promotions, bundles, comparisons, UGC, all of that stuff, which is all what marketing is all about. Like SEO is a consequence of marketing, it's not a marketing challenge in itself. So people still buy from peoples. People still need the desire to buy a product, whether that's SaaS or whatever it might be, a new mattress. Whatever it is, there is still an audience that wants to buy something. Technical kind of brings the barrier down and makes it more accessible for people who are in the market or researching that. But there is a broader sense of what's going on.
There is, to come up with a really tactful example and to your point, Ash, about data, I think this is incredibly unloved as an area. In almost all clients I work with when I do things like GA4 audits and CRO and so on, we have to have a solid baseline for what's performing maybe 90% of the time. And that's probably been quite a conservative number. Analytics is broken somewhere.
So I think probably the catalyst for me when it comes to growth is making sure your data is completely watertight before you scale. And that goes for SEO as well because attribution's a problem, especially with this consent mode V2 nonsense, which I don't want to spend so much time on because it makes people cry. It's horrible.
But there is a solution for that. I'd love to plug a tool if I can called that Littledata and that kind of just basically ticks all the boxes for your e-comm tracking and just kind of removes a lot of the stuff. It's all server side, the whole about compliance and so on. But that normally solves a lot of problems for particularly people on Shopify. But yeah, data analytics, attribution, all of this stuff is incredibly painful to get right, but when you get it right, it's much easier to scale and rely on the information that you have. So there's nothing worse than spending a 100K on ads and then everything comes as direct. You're like, "Shit. What the hell was that for?"
Ash Young:
Yeah, I agree. Yeah.
Jamie Indigo:
Tying back to Ash, SEOs for a long time, we've had our lane, we've had our GSC, we've stayed there, it's very nice, our home, it's occasionally broken. We have to venture out now, we need to be in the merchant centre, we need to be looking at our feeds. They're incredibly important. If you want to be in that product knowledge panel, if you want to be in the shopping graph results, those are populated by feed. They're a blend of ad feed and on page. But Google likes being lazy and they prefer people who open be lazy so they could expend all this energy and effort to crawl your 6 million products or you could just send them the information parsed out in a nice little bot friendly way.
Luke Carthy:
I like that. I like that.
Patrick Hathaway:
So we've been talking a little bit about Shopify today and I have certainly found myself, I think I would guess that I convert more easily on Shopify sites or other sites that use those one-click checkout things, like the link pay or whatever it's called. Has anyone actually got an experience of that or any of these things having a measurable impact on conversion, repeat buying, anything like that?
Ash Young:
I mean we get a huge number of people emailing us because they've checked out accidentally. I have one now when I was doing checking yesterday and a lady emailed in saying her son had pressed the button while she was on the website and managed to check out, because it's that easy now because everything's got your details saved. You've got Shop Pay and Shopify, we do Apple Pay, Google Pay, it's just so easy to check out, isn't it? You literally can just press a few buttons and you're in. It makes a massive difference.
Luke Carthy:
I would second that. When Apple drops was a thing before I had inventory problems, we had a new piece of tech which didn't last very long. It's called Fast Checkout. You may have seen it running around. They burnt through money at a ridiculous rate, but they were effectively trying to be one click strike. And the exact same problem as yourself, Ash, just like you literally click once on the thing and you've got an order to the point where you have a five-minute countdown to cancel it.
So you click buy now and your natural assumption is that there's another popup somewhere or another screen and no, you click that thing, it's done, man, the order's done. So it's faster than Amazon's. So as a consequence of that, our conversion rate was pretty solid, but we probably have 10 or 15 emails every couple of days or maybe a week of just like, "Hey, really sorry, I forgot to put this in or I've got the wrong address in there."
So it's a double-edged sword. I think yes, it can help, but the worst thing about situations like that is you still have to pay the fees. So even they've clicked buy fast, they've got their money, I still have to reimburse the customer minus my fees now. So I'm at a loss. So they can be great, but I think as always it's sometimes easier isn't always better. There's that fine line between one click checkout and actually as you've said, Ash, Apple Pay, Google Pay and those consistent experiences across every single site. Contactless for e-comm is Google Pay, Apple Pay, that's all really.
Patrick Hathaway:
Yeah.
Rachel Anderson:
So do you think that having a checkout experience where they're recommending other products to the consumer is actually a better checkout experience than one click?
Luke Carthy:
That's a good question. I mean, why the hell not? Everyone else... I mean it's one of those things. Maybe try and experiment. There's very briefly, I guess one thing that worked really well for a client of mine, they're in the pet industry and actually they only sell dog food. But of course when people get puppies or kittens or whatever that may be, there's so many other things you need to think about from harnesses to crates and beds and that kind of stuff. So they just ended up going, "No, we don't sell this stuff," but they've just said, "Get 10% off all this stuff," and just set up affiliate deals through email and they cream a commission back from everybody else. And they did that for the checkout and that was wildly successful for them. So they lost sales as a result of it, but actually the commissions that they got by proxy was brilliant.
But it doesn't work for everyone, right? I imagine kind of bit of distraction in the checkout, you could find it doesn't work. It's one of those test it and see rather than a resounding, "Hell yeah," across the whole industry.
Ash Young:
I mean, you can go too easily. That's the Shopify, there's a lot of apps now where you can just one click add to your checkout on the checkout or even after the checkout. And we get a lot of people take those upsells on the thank you page and like Luke was saying, you've already paid and you go, you just press the button and it adds that to your order and charges you automatically, no need to do anything. I think some of those can be too easy and actually can be a negative experience because you end up having to refund people and deal with customer service issues for a marginal amount of additional revenue.
Luke Carthy:
Not just think of the free lunch, as they say.
So have you guys got any thoughts about the rise of subscription models in e-commerce, which kind of feels a bit like the holy grail, like guaranteed repeat business.
Jamie Indigo:
No, but I have to trust you to click subscribe. So just keep that in mind.
Rachel Anderson:
I mean, Weedmaps has had people in the industry that have had subscription boxes and they have gone out of business, so I'm not sure it's a great selling point in the cannabis industry. I think mainly because people want the ability to choose their product. There's so many strains, they make people feel different ways and so they want more ability to choose and customise. And most of the subscription box services were not able to provide that.
Jamie Indigo:
Wait, was it like a wine box service, you're like, "It has notes of..."
Rachel Anderson:
Yeah. So they partnered with a lot of cannabis brands. They were only in California. They had a set type of products. You could say what your favourite style edibles, flour, et cetera was and then you would just get a box once a month, and it would be like $200 worth of product for $100 or something, but you don't know what you're going to get. Is it going to be something you actually like? Is it going to be something terrible? You really have no control over it. And so I think consumers didn't really like that. So in our space, I think the only kind of subscription model that would work is if you had a lot more control over what was going into the box.
Luke Carthy:
Yeah, I like to think of this as like Android Auto and Apple CarPlay to the world of satnav, in that satnav was such a thing that you'd pay it extra for, probably late '90s, early '90s, where you pay a grand to get satnav built into a car. Now it's absolutely everywhere. Any car you buy now comes with one or both of those options.
My point is, again, it goes back to the old Shopify thing is as soon as the Shopify apps took off with subscription, everyone just bolted it on and thought, "Why not? Subscribe and save 5% off." With some things, it worked really well, like things that you consistently have. I'll go back to the pet food industry, dog food. Your dog eats that meal every week and how much they're going to eat, you order it. It saves you the ag of having to worry about replacing it.
It also works incredibly well where there's a lot of variety. So there's some industries, I've worked in in the world of fragrance where people love fragrances and there's so many to get around. So every month, you get a selection of six in the sample size. You can kind of play around and see what you like. Same with skincare, haircare, but there's just some industries where it really doesn't make sense. And I can imagine actually cannabis and that kind of area or even things like chocolate and snacks and so on, you want the variety. You don't want it to be forced upon you. Maybe even the purchase experience is part of why you do it.
So I guess it's like buying a designer bag and there's probably lots of people and their opinions with this, but you wouldn't want to have an automatic, even if they were cheap, you wouldn't want to have an automatic subscription of a bag because actually going into maybe a department store and having that sit down with a bottle of a glass of Prosecco, whatever and having that whole experience is part of it. I think it can be helpful. I think it can be massively destructive for businesses as well because like expired cards, inventory planning, returns, people kind of forgetting to cancel them, they're on holiday and it's a nightmare. But I guess for some, it works. For others, maybe it's just a bit of a flash in the pan of fad.
Ash Young:
Yeah, I think a lot of people have jumped into it because they can, but I think a lot of people don't think about how it's sold. So I mean we've got cats, we buy a lot of cat food, we put on a subscription, but actually how long does a big bag last us? We've got no idea. So we suddenly end up with like we're swimming in cat foods, we have to start subscription for a bit and we just worked with the clients, you alluded to this, sells supplements and they had our subscription so you could repeat the box every month was their default option.
And so actually you need to work out how long each pack's going to last somebody based on how long they're taking it, and then make sensible suggestions to them so that this one might be every month, this might be every two months. Depends what it's going to be and how long that'll last them and not just go, "Boom, yeah, everything's the same."
Luke Carthy:
Does anyone remember those Amazon buttons? You remember the-
Jamie Indigo:
The Dash, yeah.
Luke Carthy:
Right? I don't know what happened to them. They need to come back. There needs to be some kind of open source platform where you just hit the cat's face on you... not literally the cat's face, but you hit the cat face button, right, and-
Jamie Indigo:
Don't punch the cat.
Luke Carthy:
Don't punch cats, guys. Number one takeaway from this webinar. But you've got a little button with a cat face on it, you press it and then 24, 48 hours later, you get your big bag of cat food. Again, having that automation but also just the one click away.
Jamie Indigo:
There are tutorials on how to build those.
Luke Carthy:
Yeah, and then you've got the 6-year-old kid who's like... a.k.a my daughter.
Jamie Indigo:
That may be why they stopped.
Patrick Hathaway:
Lovely stuff. So we're going to move on to attendee questions in a moment. So I'd say this is pretty much your last chance to get them in. And don't forget to upvote any that you like the look of. I'd have one more thing I'd like to ask because we've been going on about so much about video content and third party reviews being so valuable, and this is the part of the webinar where Jojo tells me that I'm supposed to say how awesome Sitebulb is, but I'm not going to do that because we've got a few Sitebulb customers in with us today. So if anyone wants to have a go instead, please feel free. No one says anything now.
Rachel Anderson:
No, no, no. I love Sitebulb. So when I joined Weedmaps, we were using another cloud crawler and I did not like it. It was very expensive. It wasn't doing the things I needed it to do. I had to get customer support to do random things that I had always had the ability to configure myself.
And yeah, I talked with Patrick and he was like, "Yeah, we're creating a cloud crawler version," and it's been incredible for us. No more of my machine catching on fire, a reasonable price point, no restrictions on the numbers of URLs. So that's how I end up being lazy and managing to crash it. Although, crashing it just means you press reset and it's all fine. So it's not a big deal. I think the hardest part was literally just getting our engineering team to whitelist the IP. I think that was the hardest part.
Jamie Indigo:
I mean I fell in love with Sitebulb for many reasons. The biggest reason is your release notes, because they make me so happy. There is a certain level of honesty that comes with tech SEOs where like, "I'm fed up with this," and you capture it so beautifully.
But this tool has also adapted constantly. I've definitely sent you and requested, and you're like, "Yeah, okay, let's work with that." Constant movement, really working through improvement, really good communication. The other week, I broke a crawl so good and you were just right there to help. So I think it's the trust thing going, "Hey, I can trust these people if something goes wrong." We got support there. That's pretty lovely.
I also had an industrial crawler at another site and I had questions and the account manager was like, "Well go get certified in the tool." I'm like, "Really? You want me to get a fake certificate to go ahead and ask you the same question? Okay."
Rachel Anderson:
Oh yeah, and Luke, you mentioned earlier the scraper element, the-
Luke Carthy:
The content extraction bit, right?
Rachel Anderson:
Yeah, extraction. Yes, the easiest extraction tool that exists. You're just like, "Oh, I want this element," you can point and click. And then it's like, "Oh this one?" And I'm like, "Yeah, please tell me all of the pages that we have 200ing that should be 404ing because there's no products available," and it was very helpful.
Luke Carthy:
Yeah. It's for me being Mr. Lazy, right? So I love the work that I do, but I hate being inefficient to the point where I'm the kind of person that, the irony of this is I would spend 10 hours trying to automate something where I could spend two hours just doing the damn thing.
But the point is I love Sitebulb because you set it off in all kinds of terms and purposes. You set your crawl and then all the kind of insights and stuff are kind of there. Yes, as with always, your tool is only as good as the person using it, but it really sets you up for success and pulls things that won't even think about.
One of those things is automatically setting the crawl rates when you're crawling, when it identifies Cloudflare or when it identifies again a Shopify site, right, or another CDM or there's a situation where it finds something that it shouldn't, like obviously before a through on redirect move for certain user agents and all sorts of stuff. It just pulls them in rather than you having to crawl first, walk off, go for a call or go for a cup of tea, you burnt on half an hour and then you go, "Oh shit."
You just get all of that before you start to crawl. So I'm sure probably over the years I've been using it, it saved me a good amount of tears and hours. Yeah, big fan.
Ash Young:
I think from our side of things, I personally use it a little bit differently. Our site is incredibly template-driven. So like Luke, I mean everything's templated, everything comes from data. The issue with the data is it's entered by humans and that breaks things and I actually [inaudible 00:49:33]-
Jamie Indigo:
Humans do ruin everything.
Ash Young:
... everything's correct and no one's broken anything, no one's broken any of the links, any of the scheme that we automatically generate from loads of different bits of the page that people set up and it's good to be able to crawl and make sure that everything's doing what it should be doing and then go and fix the data that someone's entered incorrectly.
Patrick Hathaway:
Well guys, thanks for all of those. We can now put them on our product pages.
Attendee questions
I know we talk a lot about Google, what about Bing? Have you noticed increase in traffic moving conversion rates in recent months?
Luke Carthy:
Bing conversion is always higher most of the time. That'll be my quick 10-second answer. Older people with more money is a very broad gross overstatement. Internet Explorer, kind of, people who know exactly what they want. But yeah, long story short, I think Bing's brilliant. It's a smaller audience, for sure, but I think if you've got someone using Bing or what's it bloody well called now, Copilot, normally conversion is greater. Don't know why, but yeah.
Jamie Indigo:
Yeah, Bing loves us and it's great. I love Bing for that. I also love Bing because they're like, "Oh it's your data, did you want it? Here's Webmaster Tools." Don't count them out. Even if they are small as a search engine, they power a lot of other tools. They provide the answers for a lot of functionality you may not think of immediately. So thanks, Bing. If you're watching, we love you.
Ash, how did you overcome the initial challenge of getting traffic and orders in the early stage of CarMax, I'm operating on limited budget so I can't just throw money at ads or Meta, at least not at scale.
Ash Young:
Otherwise, we just threw money at Google. But we did it in a very small way. So we weren't ranking anywhere and we had a great domain name but we didn't rank for a long time so we just rolled out bit by bit and we put the money behind products that we knew should sell to figure out why they weren't selling, to make them sell. And then once they were selling, we just kept scaling from there. Obviously that can cause cashflow issues but it's the best way of doing it, is actually start small, iterate and then grow from there rather than going in day one spending a load of money.
What does the panel feel about using AI to generate short descriptions on PLPs? Could this be considered spam AI content generation, or since it's only short form content, not AI generation of longer form editorial content, is this safe to do?
Luke Carthy:
I have really strong opinions on this, to be honest. I want to go off in a little bit of a mini rant, which is, that's like equivalent of asking what do you think about getting a writer to write? Like using AI is such a broad thing to say. It's like you can't just lazily use AI to write a description, right? You've got to prompt it. You have to give it some examples of what you're looking for. You have to give it a tone of voice. You have to maybe give it other documents to ingest and understand, and there is a lot of setup to get AI to a point, in my opinion, unless you've already got a custom GPT to use or whatever system you've got to create genuinely acceptable, I wouldn't even say good, genuinely acceptable content worthy of putting on your site. So I think if you are happy to spend the time and invest the resource and the setup to get it done, absolutely.
If you're thinking of doing it by just giving it a list of URLs or a list of category names and then blindly asking it to create content for you, you're wasting your damn time. The quality issue is still a problem regardless of who's writing that content, whether it's a person, you need a good brief, whether it's AI, you need a good brief, whether it's your nan, you need a good brief. The point is AI doesn't replace quality. It doesn't excuse quality being a problem.
But that being said, on the flip side, you haven't got to go and write an essay. It is a PLP at the end day. But yeah, just be mindful that AI doesn't solve it from start to finish. You still have to put the work in.
The big thing for me is that AI doesn't question the data you give it.
Luke Carthy:
Yeah.
Ash Young:
So a human who's looking at our data will go, "Well, that's wrong." So a lot of our listings, we get information and actually a lot of information we get for our listings is entirely in accurate, particularly around years and models that the products will fit. So we go out there, we research it, and we actually make sure it fits the models we're giving it or we're listing it for. And if you gave that to AI, you're just going to end up with a worse version of what you started with in a lot of cases, because it won't go out there and do that research or spot when something's obviously wrong.
Jamie Indigo:
I mean, that ties back to trust. Yeah, you have the information right, then I can trust you. If you don't, I'm not going to trust you. Also, just throwing it out there, Google could spot when it says SEO text. So if you're doing it to try and game anything, that's going to be detected, it's not useful, particularly if you're moving down why I'm there, so if that's going above the thing that I came to see, don't do it.
Rachel Anderson:
Yeah, I think AI stuff is one of the bigger threats to SEO, not because it's doing anything good, but because I think many businesses are distracted because leadership has been sold on this idea that AI can replace humans. And when it comes down to it, the models are not good enough yet. We've spent countless hours testing things, and the quality of work we get back is so, so bad. It doubles or triples the editing time, which is more expensive than hiring a content writer. And so I think a tonne and tonne, like all the friends I talked to this industry are doing the same thing, they're having to prove to leadership that AI cannot solve the problems right now. So if we're talking about what is the biggest threat right now, it's being distracted by AI instead of the things that can actually cause growth.
Have any of you guys seen quality signals passed from PDPs to PLPs when optimising for the review system criteria? Do you think optimising for this just benefits those product pages or can pass quality signals through the domain as a whole?
Luke Carthy:
This is why I don't work for an agency anymore. With all due respect, I think that's the wrong question because do your customers-
Jamie Indigo:
Fight.
Luke Carthy:
I just think, do your customers give a shit? The answer is probably not. So if you build an experience that's better for your customers, but technically slightly inapt, it's probably going to perform better than if it was technically brilliant. But from a UX perspective, it was a serious pain in the ass.
So by all means, we have to talk about these things. But if you are doing that as a way to try and sell more by proxy through SEO, I think that's the wrong question. I think there's a question behind that, which is why. But to answer that question directly, I mean maybe, but even if the answer was yes or even if the answer was no, how's that going to impact what it should do going forward?
And I think that's normally the kind of questions that when I get asked these things, will X equal Y in SEO? It's kind of like, "Well, even if the answer was yes, and even if the answer was no, would you do anything different? Is there any point in me going down that rabbit hole and finding the answer for you?" And if the answer's no, then it was the wrong question.
But not to disrespect the question, because I think it's an important one. But I just wanted to steer the narrative. And I think when it comes to e-commerce, we can sometimes get too caught up in the technicalities of what it is rather than who it is we're trying to sell for. No one cares whether your market stall is made from stainless steel, carbon fibre, MDF or plastic. All everyone cares about is what your market stall is selling and the person that's selling the thing and whether they're good enough, right? And I think, yeah, sometimes we over complicate it.
Alongside full hreflang implementation and internal linking, what is the best way to ensure good indexation across same language e-commerce domains. All right, so en-GB, en-IB, en-US, et cetera, to avoid any duplicate content issues.
Jamie Indigo:
Are they localised? Because if you're just making English content for multiple sites, but they're not localised, that's where things get real muddy. But if there's different meaningful reasons why the UK page is the UK page and the .com page is the .com page, you guys have something to work with. Good sub folder structure, praise be. Verify those individually in Google Search Console's URL properties. That way, you can get that great high level look at the index coverage for each of them.
Luke Carthy:
Yeah. Love that. It goes back to the why, isn't it? Like why are you doing that? Is it just pricing? There's probably another way to do that. If it's a completely from product offering because of legal warranty, cool, go for it. But in my very personal experience in hreflang, in may be half of the cases, which is probably like 10 over the years, we haven't gone ahead with hreflang because it wasn't the right thing to do. There wasn't enough to warrant having individual subfolders or individual sites or whatever that might be. It normally is just language in terms of spellings of Z instead of S in US and so on. Pricing and maybe lifestyle images could be slightly different, right? American home for a sofa versus a European or British home for a sofa. But yeah, as always hreflang's a beast at the best of times. It's simple in theory, but in practise amongst everything else, it can be a real headache and then you wonder if the rewards are worth the effort.
Ash Young:
We just launched the CarMax.e. Unfortunately, I couldn't get CarMax.ie, which has really annoyed me, but that's gone really well. We've used Shopify's built-in markets, which does all the hreflang setup for you out of the box. That's worked really well. We rank almost immediately really strongly in Ireland, which has done the job, but we've also tried to look to get local links at the Irish domain to try and speed up that process. So it's not just 99% of our links are to the .co.uk, but we've also tried to start building them to the .e domain.
Further reading on ecommerce SEO
