#117: How To Use Server-side Tracking On Ecommerce Sites
In this episode, I talk with Edward Upton, Founder and Technical Lead at littledata.io, about how to use server-side tracking on ecommerce sites.
This episode is sponsored by Fluorescent, a Canadian-owned design agency who have just launched their newest, boldest Shopify theme ever. Learn more at fluorescent.co.
On the Show Today You’ll Learn:
- How to use server-side tracking on ecommerce sites
- What's the difference between first and third-party cookies
- The benefits of server-side tracking
- How to choose the right attribution model for your business
- And more
Links & Resources
About Our Podcast Guest: Edward Upton
Edward started Littledata after experiencing first-hand the pain of getting accurate ecommerce data as an advisor for MADE.com and JD Williams group. Before that, he founded Teachable.net, a marketplace for teachers to share digital education resources.
Top DTC brands around the world trust Littledata’s smart connections for accurate ecommerce analytics. Since launching the first Shopify app in 2017, we have empowered thousands of data-driven brands to make better decisions to accelerate growth.
He graduated in Psychology and Philosophy from Magdalen College, Oxford and after working for Accenture, the UK Government Digital Service and others he's always been fascinated by how technology can improve how we think and live.
Listen & Subscribe on your Favorite Podcast App:
Apple Podcasts | Spotify | Amazon/Audible | Stitcher | Google Podcast | YouTube
Please support the show if you liked today's episode:
- By rating and reviewing the show in the app that you are listening to, you will enable us to invite bigger and more impactful guests.
- Please subscribe and get notifications when new episodes are released. Sign up with your email below.
- Share the podcast with your family, friends, and co-workers.
- Tag the podcast on Instagram @clauslauter and let me know what you like about it.
- If you like the content and would like to support the podcast, you can buy me a coffee here.
Claus Lauter: Hello and welcome to another episode of the eCommerce Coffee Break. A lot of online merchants just wanna sell their product, and then one of a sudden they have to deal with a lot of technical things. So there's third party cookies, there's tracking, there's Google Analytics, and then one of a sudden the whole business becomes very technical and.
Most of them at some point will be overwhelmed. Now, in the last couple of months, a lot of things have happened, specifically when it comes to tracking and, , cookies that go away and all of these things. And we wanna dive a little bit deeper into this topic today. So therefore guest on the show today is Edward Upton.
He's the founder and technical lead at Little Data io, and we talk about how to use server site tracking on eCommerce sites. Edward founded Little Data to make it ridiculously easy to connect sales, marketing, and customer data. He started little data after experience firsthand the pain of getting accurate eCommerce data as an advisor for mate.com and GD williams.group.
Before that, he found a teachable.net, a marketplace for teachers to share education resources, top DTC brands around the world. Trust little data smart connection for accurate eCommerce analyzers. So we'll dive into that in a moment. Edward, also graduated in psychology and, of the Lene College, Oxford.
And after that he worked at Accenture and the UK Government Digital Service. So let's dive into it and say hello to Edward, how are you
Edward Upton: today? Hi there. Great. Good morning,
Claus Lauter: Edward. What got you from psychology into eCommerce tracking? I think that's a long way, .
Edward Upton: Yeah. Yeah. It's , quite a meandering river.
, yeah, I graduated in psychology. I think it was the brain science end of psychology, I didn't wanna be a scientist, but , I've always been fascinated by business , and I guess the intersection of, where business or meet psychology, how to, affect people's behavior, , influence people, et cetera.
I wanted to be an entrepreneur. Those who are founders, know that that's your natural home. You might do other things , to train up. To start a business myself and, started as, say my first business in my twenties, which was in ed tech.
grew that to a certain scale and then. Found that there was very selling to schools as a very tough market. , that first business was a lot about analytics. had a product that I realized as basically as the product manager, as well as the founder.
I needed to track what was happening, who was using it. And I got started getting in deep into Google Assets and tracking and Google Tag Manager. Then, as you say, did that, a consultant for a few bigger brands and realized there was an opportunity to take what I was doing as a consultant and really productize it for a lot of, , midsize and larger brands that, want to get all the best data about their customer journey.
But they don't have either the in-house expertise to track that manually. , or, the ability to pay for some of the very expensive, , let's say full suite, , virtually integrated analytics tools out. , so our basic premise is that the Google has provided and still provides an excellent, tool in Google, R six.
It's just that there's increasing difficulty of how you actually set that up, , as you say, respect on the one hand of user privacy, , regulations and concerns. And on the other hand, actually getting, , brand's perspective and a complete picture of how your marketing is working and driving.
Claus Lauter: Okay. So little data I owe, and I love the name because it sums up what we're talking about, tell me what's the difference between what has happened in the past with third party cookies and all the tracking that has happened in the past and what's going to happen in the future?
Edward Upton: Yeah, sure. So if we roll back, five years, , there was a very standard way of, all these big, , advertising platforms, how their ads were working on, a third party website. they'd drop a, third party cookie, which was neither authorized necessarily by.
Site, , but also actually Googles, you use first party cookies to just track, , so first party means set by the, brand or the brand's website, on behalf of these users. Say, Look, this person is the same person who's being paged the next time.
The way that web analytics has worked for actually, really for the last 15 years is that you visit the page, , there's a tracking script that loads on that page. the tracking script interacts with the cookie, and the cookie says, Yeah, this is anonymous person. it doesn't say that your name's Edward Upton.
It just says you are number 1, 2, 3, 4, and then every time you visit on that same browser, we know. You are the same person. We can link all those page views together, and then that opinion is sent to Google servers. , and aggregate it into a, customer journey, into, anonymous, , analytics.
but then what happened, , a couple years ago is, , , obviously it was a big cla down third party cookies. Like don't wanna be trapped by thumb, fur period. I don't, never consented to and, never, , understood. problem is that the baby's now getting thrown out with a bath water.
That what Apple particularly started doing and, Google's increasingly doing Google Chrome , is not blocking just third party cookies, but also first party cookies. So saying, Look, Google Attics, you can set this cookie but will wipe it. So basically, , it's much harder now to stitch together all those touchpoints and then if that's not bad enough, a whole load of kind ad blockers are even blocking the pings to Google server.
So , if you're relying on what was called client side tracking, your reporting is increasingly broken in two ways. First of all, you can't link all the task, touch, work together to understand what, , marketing or, emails or ads or whatever brought them to buy. But also you're just missing, , maybe 20% of your, user behavior, , because they've been blocked.
So little data kind of five years ago pioneered what we call service side tracking. And what this means is that rather than, , having to add any scripts, I mean we do have a script as track the landing pages obviously, , but from the add card onwards, , we can just infer what the user did by, listing into what's happening on the eCommerce platform.
, and that brings a few benefits. First of all, it's , totally. Robust, anything that's happening in the browser. Ad blocking, , cookie blocking, , users opting out. doesn't affect the tracking. , secondly, it's more secure, so you don't have to, , worry about anyone inception that user data.
, but thirdly, it's, just much richer and more reliable. , every time in order comes through, you'll get a hundred percent of those orders on gig. What a lot of brands come to us saying is, I dunno what to trust in my Shopify store. I made a thousand dollars this week, but in Googles it tells me there's only $800.
So which one is true? and the reality is generally the brands trust the eCommerce platform Shopify, because that is basically their source of treat. That's where the payment trust being made, and that's right. And so Little Data's product. This basically helps make sure that the analytics tool gigs or, another tool matches what you see in shop.
Claus Lauter: Yeah, very good explanation. I was actually about to ask you why are the numbers in Shopify different than in Google Analytics, and I think you just told us why that is. Now when it comes to the attribution and said a lot of things have been taken away from Facebook, from Apple, and so on, so forth. What's the best way to assign the perfect marketing attribution to your business so that you really know where is my customer my order coming from?
Edward Upton: First of all, one of the reasons we have always been big boosters for gigs is we think it allows you to compare apples to apples across your different channels. Each individual marketing platform, whether it's , Facebook ads or email marketing platform, or an affiliate marketing tool, they all are very greedy in what they claim.
The sales driven by and by greedy, I don't. Completely fake. I just mean that if the customer touched one of those channels, And then bought something, Facebook says, , hey, that Facebook ad for that sale. But I think everyone, , understands that most eCommerce journeys are actually a number of touchpoints.
, depending on the, value of the product's, be multiple touch points where people decided to buy. So really you need to understand not just the last click that brought them there, but the journey, , the multichannel attribution of that. We recommend that people can try and trap not just, in one isolated channel, but trap the channel side by side.
but the second part of that thing is that there's different attribution models you can use, typically in gig addicts and in most of the tools, , it's a last click attribution. So whatever was the last, I would say last non-direct click to be accurate, there was the last no one sort of source of that visit, whether it be a click on ad or a click on email link or, that is attributed to the sale.
Now, one of the reasons I'm really excited about Gig Lasiks four is, , Google's opened. A feature which they've had for their enterprise customers, their 360 customers for a few years, but they've opened up to everyone what they call data driven attribution. And what this means is basically using Google's clever algorithms, clever machine learning to sift to all of your customer interactions and figure out on average, which them are value adding.
by value adding, I mean, if the user hasn't. Interacted with that platform where they're still aboard because a huge amount of brands, in my opinion, waste money advertising to people who would already have purchased. , and this is the dirty secret with sort of retargeting and Facebook and stuff.
It's like, okay, great. Yeah, some of the time is getting 'em over the line , and advertising is useful. It's reminds a customer that maybe had abandoned to come back, but some of the time. They were already gonna purchase. your click is sort of wasted in a way. And I think that's what data driven attribution tries to solve is that very old problem of whether this is, , actually, if you hadn't done that thing, the user still would've bought.
Claus Lauter: Okay. That will definitely help you in assigning your marketing budgets and then also getting the feeling what's, the user journey at the end of a day, And not only focus on one touch point. Now, how can you imagine to work with little data? Do I do not use Google Analytics or Shopify reports anymore, or how does that look like?
Edward Upton: Little Data is connector for, , Shopify and BigCommerce with, GoogleX. We also, , have connectors for Facebook ads, because there is a reason to build audience and stuff that you want that send data points in there.
, and also for Segment people send it to , other analytics tools. , what we're recommending is, , Shopify Analytics, if you just wanna look at overall like sales volumes and product solds good, but it really doesn't give you any particularly go marketing analytics point of view.
It So we feel like for any, , ambitious growth brand, , Google should be a. Better choice, , for making those sort of decisions from. And more importantly, it's tightly integrated with Google Ads.
So ability to run clever, , campaigns in Google Ads, particularly ones that are driven based on conversions. So, , trying to target cost per position other than cost per. those are much better powered via Google assets. So you send the data to Google assets, and then you can share those audiences or share those, , , conversion goals, those s with Google ads.
Previously Google Ads had their own sort of reporting, conversion tracking thing, but increasingly Google strategy is just look, send it to Google and then share that data with other Google marketing.
Claus Lauter: Understood. Now, a lot of merchants might have concerns when they say, Okay, now the tracking is on my side, server side tracking.
And that brings them into the question is like, how secure is the data? , data security policies and whatsoever, specifically as, Apple said, we're taking them away because we wanna make sure that the customer data is. Secure, which I think Apple is coming up with their own ad platform soon, and that's the main reason, not what they were selling.
how does it work for Little Data? , does it comply to GDPR and all the other things? And how do you make sure that the data , is safe?
Edward Upton: Yeah, absolutely. Little Data is app that is hosting all that data pipeline for the brand. So they don't need to worry about maintaining that kind of cloud data security.
That's our job, , as data experts and, we've got our schools about how we, secure that with best practice. And I think that is actually really important cuz service side tracking sounds great, it obviously involves a lot more engineering effort to make it work.
I know very few brands that've tried to do that on my own and most of them have eventually come back to us cuz , they've found it to be technically a lot more complex thought. So yeah, we're offering a fully hosted solution, you know, a SaaS product where people just pay a, monthly annual price just for us to manage all of this for them.
, and part of that management is, as you say, respecting the customer consent , on the front end. So, in Europe, the regulation says that you need to , opt in way, not rather opt out. They consent to be tracked. On Shopify they have a, consent api.
So they have a lot of the cookie banners can integrate with Shopify. The problem is though, that, with a lot of other setups, that consent is not passed on to the tracking script, ? So people click. No, I don't wanna be tracked, but they're still tracked, which is obviously not GDP compliant.
You're not complying with customer's Right. To be, ignored or to be, , not tracked. of the later offices integration with those sort of cookie banner, that cookie consent banners that says, Look, if you opt out, then we track you and at the moment that you do opt in, because typically, , one of the problems is that these has a landing page, cookie banner comes up.
, you click allow, but you need to then send that page for you because if you don't send it to the next page, you've lost all the, DTM tags. You don't know , what the use of that. So, Yeah, GDP compliance is important, , for brands in Europe. We help 'em do that, But I think going back to the service side, tracking from the checkout onwards, the user is effectively opted into being tracked.
From the point of which they identify themselves in the checkout and then complete the order, then whatever consent they do on the front end, we can still share that data with, the brand and the brand.
Claus Lauter: Okay. So in other words, as with server side tracking, you actually can follow the customer journey much deeper into the process as you would have been able to do that in the past.
Is that right?
Edward Upton: Correct. , and where the users consented, tracked on the front end link , the link that to checkout and order back to the browsing behavior.
Claus Lauter: So how does the implementation of Little Data work? For instance, you have a Shopify store or a BigCommerce store.
How does that work? Yeah,
Edward Upton: it's very simple. It's , a snap that you install on the store. , there's a few setup steps just authorized, , permission , to add the tracking scripts, to akin to those events coming on, Shopify, BigCommerce as servers. , and then just some configuration steps to, sales, you know, where do you wanna send the data?
And obviously , a few more settings around, you know exactly how you want to send the data in different formats and so on. with our installation, there's a very simple automatic route we pick all intelligent defaults for you, and then there's a manual where you can do more of the configuration.
, And takes about five to 10 minutes to set up, , and the data starts flowing instantly. What we're not doing is, pushing a sort of tag management solution where there's work to plan, implement tests on, because as said, it works as a, out the box product.
We know that for any brand installing it, they're gonna get exactly the accurate touch points because all of the, data layer behind the scenes are standardized in that platform. And , that's the beauty, I guess, of offering what we do. And, really, as I said, where we could take all those learnings and consultants and productize them, we can offer it for a much lower price than someone who is having to manually implement all this.
Because as I said, we programmatically know it'll just install and.
Claus Lauter: What kind of expert level do you need to have to then see the results in Google Analytics or coming from Shopify?
Edward Upton: the knowledge you'd need is, analytical rather than technical implementation.
So I think obviously, To make use the best you would need to be able to, , navigate yourself around Googles, find out how to, build reports, build segments, , share audiences , with Google adss and other tools. , we provide some good, so how to videos and how to do that.
We've got some good content on help center. Googles can be a little bit daunting for beginners, but I do recommend that people, you invest the time in that. We're saying, Look, you don't brand, you definitely don't need to do data engineering , to use this. , you don't need to be an expert in, , GAG or Google Tag Manager, but you do need to upscale , or find people that can help you.
Actually do the higher level task, which is analyzing the resulting data and reading the reports that Google
Claus Lauter: producers as Google is moving from the old Analytics to GA four obviously, what's your take on there? Should people then, if they're already in the process anyway, do the switch over to little data or what would be
Edward Upton: best way?
Yeah, I highly recommend it. at the moment we're recommending that people kind. Tag in parallel. So what that means is that they'll track in parallel. So that means we send the same data to universal assets and the same data to Google for. And the reason for that is there are lots of, , changes with the way the reports work built and the metrics are driven.
And so we want BrandScript again, compare like for like with the old new, not have to worry about maybe the changes because We didn't implement GA four properly. So I think that's the first advantage of using us in that sort of switch over scenario is you can be sure, you can really focus on just one of the reporting differences while only worry about what the setup's different.
there is not many other good solutions for gigs forward Shopify. Shopify doesn't have their own, out the box. What's , right connector , for Googles explore. , I don't think they're gonna launch that and come soon.
, the history of Shopify, Googles is something I would say, reluctantly supported. So I don't expect when it launches next year to be, anywhere in fully fledged. you can quite easily in there some other apps on the apps. So that will help you. , track the page views in Google on 64.
The important thing for any brand is to actually reliably track the, cart checkouts, orders and refunds, so you can get that complete picture. I think just the page views alone are, fairly useless. , And then, the only other option you'd have is, to sort of, , set this up and Google tag manager, , figure out all the data layer, , write some custom scripts for the bank page.
It's hard work , and I would say for most brands, it's still gonna miss a lot of the accuracy we can offer.
Claus Lauter: Yeah, very well. So I see that with merchants exactly, that , they're struggling to move over. And as I said, , in the past, the integration of Google Analytics into Shopify was like soso. , so little data definitely helps you there to make it probably faster and you get more data and more accurate data.
So , how surprising of little data, what can people expect there?
Edward Upton: We're actually just about to, , launch this week of free term. , so that's a new thing for us, but we want, , smaller brands to be able to try this out. We think this particularly ties in with the push to get brands on Google Line 64.
So we're offering product, , for free for brands, doing up to 200 orders a month, and then from 200 orders a month, , , up to 12 thousands a month. It's our standard plan, , which is. , $99 a month for gigs four. And then, , beyond that, we offer plus plans that come with more, , account management support for, onboarding to help you set up some reports in gigs and make sure it's all working.
and obviously coping with brands who have multi store setups internationally. Standard prices $99 a month, , with a discount annual. And, , for that they'll get up to all of the, , events they could possibly need , in Google.
Claus Lauter: Okay, before our coffee break comes to an end, you're doing, , data for a very long time, and also with the factor from psychology behind there, which I really like.
So what would be gold nugget? What would you recommend someone running a store? What should they look into to make it really work?
Edward Upton: The one thing I'm always pushing is, tracking sort of customer lifetime value. a lot of brands, , focus too much on, fairly short term as metrics around just trying to, , drive people to buy a one time thing.
But obviously, , what they need to take into account when they're thinking about what's real return investment is, , what are the kind of challenges and campaigns that are bringing me, , users that stick with me, whether I'm selling on subscription or just, repeat one of s. , , but also one of the kind of users that buy more and, don't ask for refunds.
These are things that will really affect the profitability of your average brand. Just being one of the things that tries to do is allow you to tie back all of those lifetime purchases and in fact, some lifetime values. As a dimension to gig use and reporting. If you can focus on lifetime value, you can really find the tactics that are bringing valuable users.
And you can tell Facebook to finding more people who like them. So you can say, Facebook, look here are my top, a hundred customers, whatever they're doing, whatever their interests are globally. Go and advertise to more people like them.
And I think that sort of tactic is usually powerful for growing, not just the one time purchase, but growing a customer base, which is truly valuable.
Claus Lauter: That's a great tip. Where can people find more about little data?
Edward Upton: So we have uh, active blog, blog dot, little data dot owe we have a YouTube channel, quite a lot of hits and hints and tips up on YouTube and obviously on all their usual, , social channel.
Claus Lauter: I will put them in the show notes and you just won't click away and people find you immediately. Thanks so much for your time. I think that was very insightful in getting an overview of happening in the world of tracking and talk to you soon. Thanks.
Today's Episode Sponsor
Today's podcast sponsors are Fluorescent, a Canadian-owned design agency who have just launched their newest, boldest Shopify theme ever. Infinitely flexible and loaded with powerful features, Stiletto offers endless possibilities to take your luxury brand to the next level. It comes packed with highly-customizable sections and built-in app functionality to create a compelling product discovery experience that eliminates friction and boosts conversions. Learn more at fluorescent.co today.
Sponsor Options: We offer a range of sponsorship options for the show. To find out how you can reach your target audience through the Ecommerce Coffee Break Podcast click here.
Get notifications when new episodes are released. You can opt out at any time.
Continue the Conversation
In our ECOM MERCHANT PRO community, you can connect with our podcast guests and continue the conversation.
Our community is also a great place to get advice from other Shopify merchants who have achieved what you are aiming for.
This is your safe place to actively grow your online retail business with the support of the most amazing and helpful group of ecommerce entrepreneurs behind you.
Running a Shopify business is tough, don't go at it alone.