Transcript
Sarah: Hello, Mike.
Mike: Hey Sarah.
Sarah: How are you?
Mike: I’m doing well. How are you?
Sarah: Well, thank you. Well, thank you. So this is a bit exciting. It’s a slightly different conversation than normal of what we have. I have been getting a lot of questions about server-side analytics lately. All about what is it? How does it work?
Uh, should I be using it? Should I not be using it? And what are the benefits, the risks? Lots of questions. Rather than me try to answer them a little vaguely as I would if I did that, I thought it would be a great opportunity to sit down with you, have a chat and answer some of these questions.
What do you think?
Mike: Sounds good.
Sarah: Awesome. So before we get into it, uh, I am mindful that you are in the Performance Crew. We’ve been team members for about eight and a half years, which is pretty cool.
Mike: Yeah, that’s right.
Sarah: And I’m interested to just explain a little bit about what do you do in the performance crew?
You’re a senior data analyst in the team. What does that mean?
Mike: Yeah, so, within the performance crew, anything that has, uh, a strong foundation of data. That’s the… the space that we’re working in. So there’s quite a lot around reporting and data visualisation. We also manage tracking configuration and implementation, which is sort of what we’ll talk more about today.
Yeah. Um, a bit of work around search engine optimisation, so that’s something that we do a fair amount of. We also do pay-per-click or PPC management, so Google Ads paid social. Uh, and a bit around CRO conversion rate optimisation as well. So as I say, anything where data’s involved, that’s, yeah. Where the Performance Crew comes in.
Sarah: Alright, I’m gonna kick off with the big question straight outta the gates as they say. What is server-side analytics?
Mike: Server-side analytics is, simply put, a tracking implementation method that relies on some server infrastructure that an organisation itself owns and it’s something that addresses a lot of the issues and shortcomings of a standard tracking configuration, many of which people wouldn’t even be aware of.
Sarah: Right.
Mike: But I suppose put practically, or in terms of the, the outcome that it produces, server-side management or server-side configuration is really a way of mitigating risk.
I’m gonna go down that path a little bit. We’ll come back to the tracking side a little bit later. Knowing that risk and decision making go hand in hand in business and in life, I suppose, can we look at maybe – specifically for not-for-profit, as well – can we look at what are, uh, some of the risks that exist when we’re talking about analytics and how server-side mitigates those risks?
Sarah: What risks are we thinking about?
Mike: Sure, sure. There’s probably, for most non-profits, three areas of risks. Or three types of risk. So one would be around the accuracy of their data. Everyone wants their data to be accurate. And once again, a lot of people won’t even realise that the level of veracity within their current tracking configuration is not as good as they think. It’s not complete. Um, another one is control, control of data, so that you can actually manage it in a way that is, let’s say, compliant with government regulations and. Secure. Uh, and then the last one is, um, having a configuration that enables you to be optimally competitive. Right. I see.
Yeah. And that particularly plays into digital advertising if you have a server-side configuration. You’re gonna have the competitive edge when it comes to advertising. And we can get into exactly why that is.
Sarah: Definitely. I think we’re coming back to that one. You said something really interesting just now around, uh, the inaccuracy of data. And I guess that creates a level of distortion in some ways. What, what type of just inaccuracy or distortion are we talking about?
Mike: Sure. There’s gonna be two types of distortion. Uh, one is quite self-evident or it might be the, the one that most people are familiar with, and that’s just incomplete data sets.
I suppose if you have a tracking configuration, and most people have a standard tracking configuration, That’s just not gonna record everything that happens on your website, right? You’ll have nine people visit the website, and your analytics software will only pick up eight. That’s just an incomplete dataset.
And while that’s a problem, it’s not really the big problem. The big problem is this other type of inaccuracy. Which is a lack of the connection points or the data that connects disparate sets of data. And so these would be the things, well, historically, like third-party cookies that we rely upon so that we can connect one person’s visit to the website across multiple sessions.
Sarah: I see. Okay. And so server-side analytics can help with those connection points?
Mike: Yep. Absolutely. Yeah. That’s one of the, the main areas where it’s providing value.
Sarah: Right. I see. And does that speak to the competitive advantage that you were just talking about as well?
Mike: Yeah, absolutely. It’s an issue of attribution.
Sarah: Right.
Mike: Particularly if we talk about things like Facebook advertising. It’s something that a lot of non-profits will be running: paid social campaigns through platforms like Facebook. And the real problem that you face is that the first time someone visits your website from an ad is probably not when they convert.
If we talk about things like fundraising campaigns, you’ll be running Facebook ads and you might get people to click on that ad and come to your page and read about your organisation and maybe read about a client story, let’s say. And understand, okay, this is how this organisation uses the donations that are given to it are gifted to it, and that’s the first visit.
And then there’s a phase of contemplation. And then that person, or some people, not all of them, are going to come back later and actually convert and donate. Yeah. Now with third-party cookies, or with a standard tracking configuration that relies on third party cookies, you’re going to lose a lot of those conversions that happen at a later date.
Those aren’t gonna be connected back to that initial visit. Because you don’t know. No, you don’t know the, the third party cookie gets cleaned out, gets deprecated. And then that platform, Facebook, for example, can’t actually see that that person who visited on this date has now converted. And that data is really important for advertising platforms.
Sarah:
Mike: It allows the advertising platform to optimise. And target people who are likely to donate.
Sarah: Absolutely. I can see why the competitive advantage would be there now by having server-side analytics in play.
Mike: Yeah, I mean, one of the reasons people are, uh, contemplating maybe Facebook conversions APIs, another thing you might’ve heard of.
Sarah:
Mike: Yeah. That’s, uh, a platform-specific solution that’s solving the same problem.
Sarah: This is becoming clearer, which is awesome. Very briefly, uh, going back to risk. How do we mitigate the risk created by inaccurate data?
What do we do?
Mike: Hmm. Well, first thing that’s good to do is to understand what’s contributing to that inaccurate data.
So some of the factors that feed into that. And then I suppose there’s different ways to handle those different, um, influencing factors. So one of the things that’s going to be a factor there is just the fact that there’s increasing regulation around the way the data is collected and used. So you’ve got things like GDPR coming in and a lot of governments are upping the level of data compliance and, I suppose, the expectation on organisations to manage data in a secure way, but also, um, not collect certain types of data.
Sarah: Sure.
Mike: Yeah. So that’s one of the things, one of the other factors that’s going to feed into that and has always been, I suppose, a bugbear or something for analysts and data experts to deal with is things like ad blockers, browser settings, VPNs… there’s just a lot of user choice around things that limit what you can track and what you can see. And so there’s, well, with some exceptions, there’s really no getting around that and that’s just something you have to deal with. And that’s okay. And then one of the other things, and maybe this is the big thing around server-side configuration, is the, as I mentioned, deprecation of third party cookies. So, cookies installed by things like Google Analytics or Facebook that allow you to track a single user across multiple sessions. The shelf life of these is just reducing down, down, down, down. So we can actually rely on that less and less to give us a complete and accurate picture.
Sarah: I see. That’s really interesting.
Moving along then a little bit. Can we go to the practical side of the tracking configuration that you talked about towards the beginning of our chat. Can you expand on that a little bit more? What does that… what does that mean?
Mike: Sure. So it’s probably easiest to start with the standard method of tracking.
Sarah: Yep.
Mike: Often referred to as client-side tracking, uh, in contrast to server-side tracking. Now server-side tracking is really no different from client-side tracking. It’s just with an extra piece in the mix. Okay. So we’ll start with client-side tracking. So what typically happens, or what happens when someone visits your website is they enter in the domain or they click the domain that sends off a request to a web server. The web server sends back a response, and in that response is all of the code that’s required for the browser to build the website. And that response also includes all of the tracking code.
Sarah: I see.
Mike: So when the website’s loaded, the tracking code is executed and that establishes a connection between that visitor’s browser and the third-party platform.
Mm. So in this case, I suppose we’re talking mainly about Google Analytics. Yep. So there’s a direct connection between the client-side environment and the third-party platform and server.
<span”>Mike: Yeah. Yeah. So all it is is really just plonking down a server, that you own, that’s tied to your organisation’s domain between the client visiting your website or the the site visitor and the third-party marketing platform.
<span”>Sarah: So server-side then really gives you, as you say, that competitive advantage because you have far more control in the middle. Awesome. Last question for you then: This sounds fantastic, but what are the, who are and what are the types of organisations that would be best suited to, uh, think about a server-side analytics configuration?
I would like to say almost everyone, but I don’t think it’s everyone. I think there’s probably a select type of non-profit or, or different types of organisations that this is most suited for? Who would you say is best?
Mike: Yeah, I’d agree with you. Not everyone needs to immediately go server-side. Not everyone’s gonna benefit from it in the same way.
I would say, in my experience, all mid-size to large non-profits should really consider this. They’re gonna get all of the benefit from it. And that’s for three reasons primarily. So, one, mostly we see not-for-profits are investing quite a lot in digital advertising, especially around the the fundraising functions.
So they’re investing in these campaigns to drive donations. A server-side configuration is gonna help you optimise that ad spend and get a better return. So you will see that. Just, that’s immediate benefit. Uh, one of the other things that it will provide, and you can leverage to a, a lesser or greater extent, is more control around the data. You can be as secure as you need to be, as secure as you want to be. And it does give you that extra, let’s say, layer of, uh, moderation right around what you’re collecting. How you’re modifying and processing that data and where it’s going to.
Sarah: Awesome.
Mike: You can use a server-side configuration to capture and store all of your own data if you want to.
Sarah:
Mike: And then the last one is, once again, from our experiences, uh, non-profits usually have customer journeys that span a number of weeks, if not months. And as we’ve talked about, a server-side configuration is really gonna help you with this multi-session, multi-channel. Tracking.
It’s gonna give you an accurate picture of what happens over a much longer period of time.
Sarah: I see. That’s incredibly helpful. So, Mike, this has been awesome. So insightful. You’ve given me so much knowledge and understanding of what server-side analytics is, why it’s useful, who would be using it, as well as who would benefit from it the most.
I’m sure there’s going to be lots more questions. Maybe we can sit down and have another chat and see where we land.
Mike: It works for me.
Sarah: Cool. Thanks,
Mike: Mike. Thanks Sarah.
