16 January 2023
This week, Ed and Tom dissect Shopify’s Commerce Trends 2023 report.
As usual, the guys take a behavioural science view, and they don’t pull any punches!
They talk about:
- How Shopify anticipates marketing will change over the next year, as CAC continue to skyrocket
- The impact of tougher privacy legislation (hello, £300+m fine for Meta from Ireland’s DPC)
- Shopify’s drive to help brands improve their marketing once they land on site – and why that’s too late
- Why Shopify’s “collaboration” idea probably won’t work
- What Brad Pitt and Angelina Jolie have to do with any of this
If you want to start thinking about how to replace third-party cookies at the top of the funnel – and how to look at data in a different way – have a listen!
Tom
Welcome to 2023 and the official launch of The Grapevine. For our regular listeners, you might be slightly confused and thinking where has Data for Bluffers gone? Well, Herdify who we work for in our day jobs is all about behavioural science and more specifically, consumer influence and how that impacts buyer behaviours. So we thought The Grapevine was a better representation of that. So we’re super excited to bring you that in 2023. And with that, there’s going to be a slight format change in that we’re going to be working with more and more guests, ecommerce, business owners, ecommerce professionals, people in the industry who can share inside stories and we can have a really good conversation about how behavioural science impacts their business and hopefully give you more to learn and more to take away. We’ve got some really solid guests starting out with Thomas Robson-Hanu who was a professional footballer he played for Wales and West Brom, but also he’s founded the Tumeric Co which as a personal user is a great product, if you like your tumeric and the health benefits from that. We’ve got Arthur from Mindful Chef, we’ve got Chloe from E-commerce Masterplan, we’ve got Karolina from Levitex, we’ve got Katherine from Copy House and Michael from Marketing Memeticist. think I’m pronouncing that correctly Michael apologies if not. So that gives us our first quarter of guests, we’ve got more and more coming so it’s going to be a super exciting time. But as always, let us know in the comments or email us Tom@herdify.com if you want to hear other guest ideas. We’re really welcome to that you know, to make sure you get the best out of the podcast. But as we kick off this year, it’s just gonna be me and Ed so as usual, let’s say hi.
Ed
Hi, Tom. Happy New Year. Yeah, I got a good break over Christmas and ready to get going for 2023. Looking forward to the new format.
Tom
I also notice we’re both wearing our branded Herdify hoodies.
Ed
Little Christmas present from the company.
Tom
We need to get our merch store up and running. I think we’ve entered a period where everyone is setting resolutions so I have to ask have you set any resolutions that you’re willing to publicly share at least?
Ed
Setting myself up for failure is not good for my normal mental health but like you know, my day to day life? So I haven’t I haven’t actually graded this year how about yourself?
Tom
Pretty much a staple one be be healthier, which normally lasts till about July and then it all gets unravelled. So
Ed
you know, good seven months isn’t to
Tom
try? Right? And I think that we’ve talked about it before, you know, I think we did we did an episode where you know, do your friends are horrible the type of deal friends make you fat, but really about how group behaviour is instrumental in behavioural change. So you know, very pertinent this time of year if you’re trying to change your behaviour, then it’s easier to change it with friends, I think is really interesting how a lot of gyms have caught on to that with January’s a year is the month for fitness, right. And they’re all like exercise with a friend because they know that if you exercise with your buddies, you’re more likely to stay. And I thought with that new year’s resolutions, group behaviour would be a good lens, which we’ll go into Wednesday’s to kind of pick out part of the Shopify 2023 reports. I’m sure most people listen to this will be familiar, but every year Shopify released a report. And we thought let’s add let’s take a group influence or behavioural lens to that and then see if we can shed some light and make it better by picking this that is the is the girl. And so yeah, with with Ed, when you’ve looked at the report, they’re kind of some core themes that you’ve pulled out.
Ed
So I think so the report covers everything in E commerce, as you would expect from a brand like Shopify, it covers, you know, the retail side of things. So physical retailers that also look at E commerce, it covers supply chain and all the issues that we’ve seen in the supply chain over the last couple of years, particularly with COVID. And it also covers sort of more interesting for us or more specific to us marketing, and how Mark how Shopify anticipate marketing to change over the next year and what they think will be important. I think it’s always worth bearing in mind in these situations that it asserts to a certain extent This is Shopify is, you know, pitch to you for the next year. So as an E commerce marketer, it’s it’s a focus on okay, how can they help you with what they think the problems will be? And what what the problems that are going to face marketers across the year. And it’s unsurprising that the, what they really hone in on is the loss of first party cookies, something we’ve spoken about, you know, time and time again, comes up in a lot of our discussions with our customers that as cookies have been turned off, or, you know, opt ins have become standard across Apple, for example. And also Mozilla. cost of customer acquisition has been driven upwards. That’s a trend that looks like it’s going to continue and kind of go into hyperdrive this year, if by I think Well, I think is now actually by next year, just about Google finally, ternal third party tricky. This is, I think this has been said before, but this is the year that people really need to get their house in order that Shopify is point of view, and have a plan for the post cookie world when using third party data is going to be more difficult and thus, trying to acquire customers through the traditional methods is going to be more ultimately more expensive, because you’re going to be serving a lot more ads to people who aren’t as targeted for your audience as they were before. Okay.
Tom
And to your point, though, there’s, they’ve obviously got a product play in all of this, right?
Ed
Yes, so if you think about it, if you consider Shopify versus Google versus Facebook, Shopify don’t have an advertising platform, but where they are for many people is the best store of first party data. Right. So they store the first party data of all your purchases, but also the visits to your store, and any other information related to that. And therefore, I think it’s fair to think that Shopify is establishing itself or wants to establish itself as the place you go to analyse your first party data, to look at it, and to use that in your marketing and also in your sales. So a lot of when Shopify talks about marketing, it is actually a lot of the time is actually talking about in store marketing, so personalization of landing pages, things like that. So doing more with your store, once people have arrived at you, which helps with new customer acquisition, but only from quite near the end of the funnel. Yeah. And that’s always something to bear in mind. And that’s where I think actually a lot of marketers are really struggling is that, you know, at the other end of the funnel, the wide end of the funnel of what you might call the sales funnel marketing funnel far away from the sales point, you know, once someone’s on site, you can convert them into a new customer, but what third party cookies were doing a lot of the time, and what a lot of people are using them for is to find the people that you want to get to your site. Yeah. And to try and get people to your site efficiently. Yeah. Okay.
Tom
And did they? Well, in the report, a, they sort of started talking about some of the customer acquisition costs, you know, the numbers, they’re really they’re painting a picture that, well, we’ve still got cookies and acquisition costs are going up. So it’s only going to get worse, you know, once we lose cookies and third party data, but what did you see in the, in the customer acquisition numbers? That was interesting, if anything?
Ed
was, I think, I think what was interesting is that the so they included some stats from statistics, where they had how much paid search, CPC was rising for different different categories of product. And what they found that was beauty had the largest rise and this is in q3 2031. So these are actually reasonably old stats, but these are trends that are continuing. They found that they had a rise of 41% And that compares to a rise just 14%. So less than half of that for sports or recreation. Now, what that suggests to me and this is purely hypothesis I haven’t tested this is that sports and recreation are product categories where your interest might be actually fairly well. revealed, particularly on social channels, whereas beauty is a more general product category, it serves a lot larger range of people. You know, rather than, you know, in sports or recreation, if you’re going to buy a bike, that might be something that is quite easy to tell from your activity on Facebook or on Google. So inside the advertising platforms, yeah. Don’t forget, that’s where third party cookies are being replaced, right? It’s targeting within the platforms themselves. Whereas beauty is a much broader category, a lot of products and beauty is harder to pin down, sort of who might be interested in those products. Yeah, I think that’s potentially why cost per click has gone up. Also, it’s worth saying that they are products that during the pandemic had a much wider up or uplift. In E commerce. Yeah. So during the pandemic, a lot of people switched to both ordering things online, you know, daily staples, they used to get from the supermarket, a lot of people switched to ordering that online. And what happened, post pandemic was people switched back to buying them in stores. And because that movement back to stores, your online advertising for those beauty products was very became very expensive, because people weren’t cutting their budgets. Yeah, and they were effectively targeting people buying online, but actually a much smaller, that’s actually a much smaller group of people, but they’re spending the same amount on it on that, whereas products such as in the sports or recreation category are more likely to be higher value purchases, that people are less likely to be ordering a bit to be requiring that in store experience.
Tom
And I don’t know if it’s related, but also because of the rise in, in recreation or outside recreational activities. During the pandemic, there was massive supply chain issues, you know, I bike and windsurf to various different sports. Both you couldn’t get parts for love nor money, right? And I’ve heard that from people to do different sports. So I guess there’s there was less need to require people because actually, no, there were there were waiting lists, you know, a friend owns a bike shop. And you could sell anything to anyone because those people just wanted to buy a bike. So I guess things like that are probably not covered. It did did just made me think I wonder if I would have Strava a bit off topic. But what does Strava sell their data because they they have a great a great fee, they can tell when you’ve run 1000 miles and your train is and you need a new pair or you need that x miles and your bike and service
Ed
drivers in each interesting one. For people that aren’t aware Strava is an app that tracks your your activity. And it is a very interesting sales or potential advertising platform. I think is not used for that. But it is but equally Strava has a history of increasing the both the the cost and the amount of the service that requires a subscription. Yeah. Suggesting that they are struggling. They do struggle to monetize the product, right? Yeah, there’s a free version. And they’re constantly I mean, I have the free version of Strava. And they are constantly trying to give you reasons to join the paid for. And then maybe that is that is the road, they’ll go down they’ll start trying to sell advertising space within the app, which currently there is none. I think it’s while we’re on the subject, I think it’s gap in America has launched a sort of fitness app that wants to compete as part of AI as part of a think I think it was gap as as part of a kind of drive effectively to collect more first party data around people using their to their athletics where
Tom
they got to be a bit a little bit careful. I think Shopify got dinged not once but twice for exposing secret US military bases because soldiers were doing laps in the desert on a relatively small thing and when when you’ve got I don’t know how many people are on a base but when you’ve got you know people doing laps all day you actually get a heat map. So there are these spots in the in the desert where I think they were supposing Special Forces bases where they said we’re not going to we’re going to turn off default data sharing and will happen again but random aside that I took us down there with with that, but I guess when we go back to where that started in terms of using first party data and the CAC numbers you know, so yes, yeah, they they Paul CACs going up, and you know, we can hypothesise why the gaps there, but you know, I think they’re seeing it, you know, smaller scale. But you know, we see that from pretty much every one of our customer conversations. And you know, first party is first party data, certainly seen as the, as the Saviour, or the light at the end of the tunnel. But they made some interesting comments around how they feel first party data to be used, which I think would be interesting to discuss. They’re like, firstly, what, what are they supposing people are going to do with their first party data?
Ed
Well, I think firstly, there’s a, there’s a big assumption that you can get very detailed, first by data. Right, this is something that people talk about first, by data as a saviour. You know, what can you learn about your customers? By talking to them? You know, this is what I mean, some we’ve discussed this on the podcast before some people in the US like to call zero party data. So
Tom
seeing that rubbish. And
Ed
it did it did by asking your customers questions, you know, asking questions of your customers, which are not, they’re not giving up the data that is necessary for them to buy your stuff. Right, but having that what people like to call a relationship with your customers. And I think that’s, there’s a lot of presumptions, in the kind of ease at which people think that they can get people to give up their data, you know, Google, give away a lot for free, and yet, struggle to get people as we found with what’s happened on Apple and iOS devices, struggle to get people to agree to give up any real like data, they don’t even have to answer questions for. So there’s a massive assumption that people are going to be willing to tell you things just because there are brands you like to wear. In particular, I think this is a problem for bigger brands, you know, a small a smaller brand might be able to convince people to give them data to help them out, I think a bigger brand is going to find that harder. And secondly, that the big, the big kind of proposal, I would say in in the Shopify report for how to deal with this, or what to do with the first party paper data, and how to use it to find new customers explicitly is for brands to combine their data. So for, you know, for one clothing brand to go to another and effectively what it sounds like is they call it a collaboration, but really is a it’s a it’s a excuse to share each other’s first party data. And then use that as a as a way of acquiring new customers. Now, it’s very unclear how this would work, you know, is it is it that you collaborate with people who have different data to you? To try and sell your brand’s products? So they become your customers? And then you get data on them? Or is it explicitly? And because it does suggest it does, you know, mentioned in the report about date, people give me consent for this to happen, you know, is this that if I bought, you know, a herd of fire hoodie, and when I heard of fire gonna email me and say, we’ve got, we’ve got this other brand of trainers, you might like, are you happy for us? To tell them ever know everything about you, but tell them your email address? And then we will send you, you know, emails about their trainers as well?
Tom
Yeah, well, just as a quick aside there, if you want to buy her to buy her do let us know, we might be selling them. So we’d like to serious point. I think I had a similar Well, yeah, obviously, then they’re not. It’s not clear in there, what what they’re proposing, but yeah, I had a similar sort of red flag if you like, you know, let’s take the June repo, right, who I’ve already mentioned as one of our guests, and I said I’m a subscriber to this, this guy, Tom has set it up because he had injuries from being a professional footballer, and, you know, tumeric is, is well, well publicised to help with those sorts of things. Now as a subscriber who also has sports injuries and things cheering my health, I’d be deeply uncomfortable then going well, he’s obviously got a sport injury. So let’s sell this data to lots of other people who have sports injury related products that might be able to help him you know, I can think of nothing worse of being, you know, all these other Bo Fred’s reaction from other people going, Oh, you’ve got a sports injury, you know, let’s sell you something. So I was Yeah, I was really interested to read that and how they think that that’s going to have again, square that circle with personalization, but also on the other side of the report, they already talked about how people are nervous and don’t trust how you know how much data or sorry, how responsible brands are with data. So it’s an interesting, interesting juxtaposition.
Ed
Exactly. And it’s I don’t know where it stands from a GDPR perspective, I think that’s something, too, to investigate. And if anyone listening knows, please send us a message and tell us but as I understand it, like a standard data agreement would be for with us with a standard great agreement with a store is that you agree that they can use your data to sell you their products. No, they can send you they get, you can sell, they can’t sell that data to a third party to sell you the third party’s products, but it’s a bit of a grey area, or I’m not sure how it would be covered if they if they send the email effectively, if the store you deal with sends the email, which advertises a third party store. Now, that to me, feels like it’s basically a third party advert that’s being sold by the first party.
Tom
You know, it’s not quite, but it’s almost a slight variation on retail media, you know, someone’s already got your customers, it’s actually then you’re, you’re using it. And the other thing to me that jumps out, and you know, we’re always biassed with with this approach. But it still doesn’t deal with the base, I guess, psychology, if you like, or behavioural science behind how we buy, you know, there’s, there’s been a lot of well, over the last 30 years, it’s been demographic variations on demographics and interests, which there’s plenty of publications out there is not how people buy. But just because someone has purchased a product in a similar or, you know, competing, sorry, a similar or overlapping area, or complementary is the word I’m scrambling to find doesn’t mean that they’re going to trust, trust you to make that recommendation, you know, people learn from people they know or trust. And just because they buy from a brand, that doesn’t mean they trust from a brand, I think there’s a really subtle step. But they might be interested to see how it plays out. But they might be overstating that. But just because you buy from a brand is doesn’t mean you trust them to make further recommendations from you. It just means you buy you buy their product.
Ed
Yeah, definitely. I think and we mentioned, as we mentioned earlier, like that, you know, for smaller brands, they might have, you know, almost cultish following, and that might work. Yeah. But for larger brands, I think it’s really difficult. And also for Yeah, for brands we started established that kind of sense of trust. With their, with their purchases. Now I would I would guess that, as you know, I am always of the opinion that brands think that people care about them more than they actually do. And that’s because, you know, a lot of surveys have a very, I would say unclear narratives of whether people are loyal to brands. Yeah. You know, most, most surveys you see suggest that I think the most recent numbers I saw from last year, I think this was the Dhobi survey. And they were they were sort of around 70% of people claimed to be loyal to brands. And this very similar number would change for a low price. So loyalty only goes so far as you wanted to add I think I think as you say, like trusting someone to make recommendations for you, as a brand is not how humans traditionally or Yeah, culturally have developed, right? We trust recommendations from people we trust. We spoken about this a lot. We trust recommendations for people we trust. And in general we don’t trust brands and supply there’s
Tom
there was a there was one day I forget I forget who the quote comes from. But brand, a brand isn’t something you tell the customer you know, a brand isn’t something that you tell the customer what it is a brand is what their friends tell them it is you know, it comes back to them. If you trust that the brand, the brand message would be the one that resonates but it’s not it’s what it’s what it, how it how it gets moulded in the community is is how it works.
Ed
And within the Shopify report, the collaborations they suggest as examples are between brands in similar industries or even actually the same industry often so like luxury fashion, for example, that a going after already quite a similar group of people, but this is not a way of advertising your brand’s to a new sort of body of people. It’s a way of effectively selling, I think, an slightly augmented version of your product. Because a lot of these collaborations involve collaborative products. It’s an augmented version of your product, which gives something new to existing customers a lot of the time.
Tom
Yeah. And it’s a bit I guess, it probably worked. I would guess it works better when the brands that brands are known, you know, so one, they told me there is Gucci and North Face, right. And that’s there, too. Yeah. All those iconic brands in their own urban areas. And it’s really interesting. And I guess I liken it to the celebrity marriage, right. You know, everyone knows when everyone knew of Brad Pitt and Angelina Jolie got together, right? Because two big celebrities in their own mind. But I couldn’t tell you who Brad Pitt or Angela’s next partners were right, because if they marry someone of a lesser statue on the fame, I’m not making comments about them on a personal level is less interesting. And I wonder from a branded flower point of view, it would be they’ll be the same, you know, the big brands? Or if you know about it, there’ll be really interesting. Otherwise, it might be a a don’t share my data play.
Ed
Yeah, I think I think that’s a good example. Because that, you know, that sort of, there are people who, previously a Gucci customers who might buy a north face Gucci product, and there are people who are North Face customers who might buy Gucci North Face products because of that collaboration. But a how many people who are regular Gucci customers become regular North Face customers because of that? Yeah. Collaboration? That’s the challenge. And secondly, they’re both massive brands, right? Yeah. Yeah, exactly. It’s not, it’s not Gucci and the corner shop. Yeah, Tao, the end of your row that’s doing something that adds up with, you know,
Tom
half price or different, you know, they’re
Ed
both, they both got big. They’re both big brands to begin with. They’re not smaller ecommerce brands, which is the situation Leo the situation and most marketers like ecommerce marketers are in, right? They’re working for smaller brands like Gucci and Northface. Yeah. And
Tom
that’s the interest. That’s the example they side, which is not maybe representative. But I think that I think what’s really interesting there, I think it’s a really interesting conversation. You know, are they we would I don’t think a lot of people would disagree that, you know, cookies out first party data, it is the trend we’re seeing? I think the big question, you know, well, the big question for me, is, and using their words, in the report, they say, there’s still one big question, first party data can’t answer alone? How do brands find and attract these new customers? Now, as we’ve discussed, their angle for this is collaboration. I don’t think I’m necessarily convinced, you know, we will see how the year the year plays out, you know, it’ll be an interesting one to see. I think for for me, it still comes back to you’ve got to look at the underlying behavioural psychology behind how people buy, you know, maybe just sharing your data with another business who’s got first party data is maybe not going to be the way that yields the best results. But you know, they’re a big company. Let’s I guess, let’s see how that plays out. I guess.
Ed
I think. Yeah, I think I think fundamentally, the the issue here and this, this is, in a lot of the attempts, I guess, to replace third party cookies, is precisely that people are trying to do what the web, people are trying to directly replace the third party cookie bit of the of the process of targeting, people are trying to say, Okay, how do we get the data in a different way? How can we get permission to use the data? Or in the case of Google and Facebook? How can we ring fence the data, and then serve it So advertisers can use the benefits of the data without having to have the data themselves? Right. And suddenly, we we haven’t mentioned yet is the ruling from the European it was the No, it was the Dublin Core, wasn’t it? I think last week, Oh, yeah. about Facebook. Yeah. Basically, by serving meta and telling them they can’t, they’re not allowed to use the data you put in because it’s social network to target you adverts even within the Facebook platform. That has to be an explicit decision. So you on Facebook, you have to say I want to see targeted ads and I assume that is Gotta be have to be more than just a button that says I want to see ads. I’m interested in wishes. One, I think you can click it on, I think Google has that option, right? Is it the, I want to see relevant ads or rubbish is effectively the other stuff, you’ll allow me to see stuff? I don’t want to say, yeah, you’ll still see the same number of ads, they just won’t be as relevancy. To exactly.
Tom
I was just adding, it’s really, as a really good lens, I hadn’t really thought of it like that, that actually, they’re looking at the problem. And they’re not looking at the problem of how can we use data differently? They’re always just looking at is Google’s taking our toys away? How can we get our get the same toys back? It’s not really maybe looking at it through a customer lens of, you know, because there’s, there’s two sides to the first, the data argument, right. And probably the leading one is, is regulation and privacy concerns from consumers? You know, we’ve we’ve talked about this many times, but consumers are much more aware, you know, things like films like The Social Network, that was, what was it 2019. Now 2018 Really opened people’s eyes to the some of the predatory nature of data sharing, you know, not all data sharing is predatory. But it can be certainly used that way. And that’s caused a lot of this regulatory noise and pressure that’s coming, which is, which has led to things like Google, originally trying to get ahead of it, maybe by the time they actually sought solution, and they won’t be ahead of it. But yeah, just trying to replicate how you get the data doesn’t really acknowledge too much about the consumers initial fears, which Yeah, I think probably adds weight. So again, be interesting to see how that plays out.
Ed
There’s sort of the two sides of this, which is the regulatory side, and then also the consumer behaviour side. And the fact that people became wary, and you say, like, people became wary of, I can’t remember the exact words you use, but you know, people using it, for using their data for slightly nefarious things. And it wasn’t just a I think it wasn’t just the people who came weary of that. It’s like people that became people’s default. Understanding was if someone was my data, they’re going to want to do something bad with it, or something that I don’t necessarily want them to do with it. And the thing is, no one, no one wants to sit or read data use agreements, to flick through it and go, Oh, no, they already want to do the good stuff. Right? Yeah. So that means that everyone’s going to like the trend is very much that everyone opts out is opting out of everything, because they assume something, it might be something they don’t want, and they get to sit there and read to make sure that they actually are happy with everything. So the only options left to companies are to add, you know, provide real incentives for people to pay them to share their data. And, you know, there are talks of companies, you know, paying people to share their data, for example, or paying people to collect their data, I believe, hi, can’t remember the name of it, someone was telling me that there is a web browser, which essentially runs an ad blocker to replace the ads that are placed on websites, or heard about that. And then you get a cut line, as a user of that blocker, you agree to share a lot of data, but you get effectively a cut of the money that comes back from that.
Tom
Yeah, well, there’s also that that plug in, isn’t there, the chap that was on Dragon’s Den a couple of years ago? I think Sam has his name. I can’t remember. I don’t know his full name. But that’s it. I think he’s eventually got a Chrome as an ad blocker. All right, I guess that does exactly that. You can either block your ads or opt in to get paid into their wallet. You know, a few pennies. Yeah. You pence an
Ed
ad. That bad must have been must have been wire it was someone was talking to me about Yeah, over Christmas. This was but but it’s yeah, it’s things like that, like, even if it’s not a solution, it is, is something that people are talking about. And if it’s on the table, people might start to use it. Yeah. And anything that disaggregates the sort of the, the holder of Google, in particular, in the in the browser space is going to, you know, make marketers struggle more than they already are, unless they look to solutions, which work outside of the sort of a system replacement
Tom
point, the walled gardens.
Ed
Yeah, yeah, I think I think either you end up in a situation where you’re trying to do a lot of things directed by different people. Or, you know, you’re trying to deal with a lot of organisations with different different ways of replacing third party cookies and different levels of control within that. There or you look to try and do To implement solutions, which help you find customers without having to deal explicitly in the weeds of how each of those programmes or those systems is working,
Tom