The complicated world of reporting

Available to listen to here.

Transcription

Hello and welcome to unmuddled Kite Factories podcast series, where we take complex media matters and unmuddle them. Gabby Krait, head of Digital Operations. And I’m joined here today by Nora. Hey. and guy. Hello. Hi. And they are some of our most embedded people in reporting at the Kite Factory, and they’re joining me today to chat about the complicated world of reporting, in particular in digital, and how we can get the best out of our data in this complex world. So I think let’s open by sort of chatting between us on what’s so difficult about reporting nowadays. Why isn’t it just like a neat can you just tell me how many sales I’ve got? What’s the complexity in the modern world? Well, obviously the big thing at the moment is cookies are going away. There’s so much discussion around data privacy. You’ve got GDPR, you got changes that Apple has made, you got changes that different browsers have made. Whereas in the past you were able to track user all the way across the web and different websites, you were able to see exactly where they’ve been, what actions they’ve taken, and then they’ve converted. You were able to do that. Whereas nowadays, as we’re kind of moving away from that, kind of granting uses, more privacy, being more mindful about how we track them, essentially we can’t track them down anymore in that way, and we can’t really measure our performance in the same way. We have to kind of look at it in a different way than we used to. So that’s like the big challenge at the moment. And it’s kind of trying to find the ways of how do we get there, how do we look at performance, how do we shift our thinking? More so away from looking at all our campaigns, all our performance uses being kind of a transaction and more looking at engagement and moving that relationship into what you actually are giving to users rather than kind of trying to get something in them in return. Less transactional. And I think multitouch multichannel campaigns have always presented a problem as well, which is how do we track, particularly if it’s above the line and below the line? How do you measure the impact of television when you’ve got when you’ve got things that like Search, which are both self generating to a certain extent and also will uptick or should uptick if you’ve run a successful TV campaign? And it’s how to unpick those channels that have obvious metrics and reporting tools based on like Search and those that don’t, which are obviously predominantly the above the line channels. And I think Attribution has been how do you properly attribute has been that question has been around for at least 15 years that I’ve been involved in. Sometimes you feel like we’re near to cracking it and then something comes along, like the cookie banner. We go, oh no, now it’s moved further away. Again, technology has arrived. That helps a lot. Things like the subtle tool that we use in flight deck allows you to pull a whole load of data together. And I think the other factor pulling away from an attributed view quite deliberately are those big silos like Facebook and other players who don’t really want anybody to really look outside of their little ecosystem. There’s always a push pull going on in that attribution space. There’s no single silver bulletin. But there are lots of things that can be done to try to allow a broader, more holistic view, as you say, not just across that, what is your conversion point, what is your success? KPI but also how you measure it across 1012 whatever channels and your previous experience, like on the advertiser side, like what you just described, does that reflect the tensions and the difficulties that you have from working as an advertiser in terms of how you receive reporting from an agency? Is that reflective? We’ve tried various different mechanisms to resolve that and it very much depends on how much internal resource you’ve got to throw at that kind of thing. If you have a data science team, then great, you can give them multiple data sets and let them have at it. On the other hand, I’ve also worked at clients where actually it’s quite important and there is probably not much hope of doing anything more than a quick sort of simplistic take on it which sometimes shows you just confirms what you suspected, namely that your offline is affecting your online, for example, but hasn’t allowed you to do anything kind of real time or optimization. It’s more post campaign analysis snapshots. But yeah, we’ve never used it I’ve never used it to drive an agency kind of relationship or KPI that’s normally been done from far more kind of in a last click. Let’s just look at a single platform by platform basis which everybody knows is simplistic, but at least it works. And that’s as far as I’ve got. And just actually out of interest personally from an organization that is smaller, doesn’t have a data science team, wouldn’t necessarily have all bells and whistles in their own internal reporting, would it be difficult for an agency to then come in with those solutions and those answers? Or would it be easy like how open with a client be? I’m just thinking if we were able to say we can look at that holistic picture for you. Is a smaller client receptive to that or is it going to be a challenge because of their organizational challenges? It really depends on the mindset of the people on board. You can find big companies that you would think would absolutely get complexity and actually weirdly don’t want to know or aren’t they interested, or just pay lip service to it. And you can find small companies with one or two very sharp cookies in it going yep, yep, yep. And they have a real appetite for doing quite high end stuff. It’s very much about the mindset and perhaps the background of the people and the key decision makers involved. And there’s a slight economic element to small players, even if they do want to do something, are constrained by their available budget. But I think there’s always, if the ambition is there, there’s always kind of workarounds. And even if there are limitations to the results, at least they’ve done something and there’s enough time to push and push. So I think it’s about mindset. Yeah, absolutely. I think that’s an absolutely spot on because we at the Kite Factory, we have a range of clients with different apps to advertise and some of our smallest clients actually have some of the most sophisticated reporting. So, yeah, I totally agree with you on the mindset. Sometimes having fewer layers of an organization to change is a help as well. Like, if you’ve only got, you know, a dozen people, half dozen people are convinced it’s simple if you’ve got three departments of underviet to change their minds on, something, that can take years. Yeah, absolutely. So, Nora, you were talking earlier about that more holistic view, stepping back. What does that look like for you in digital and your clients? It really depends on clients. Like you just talked about, if you got smaller clients, you’ve got smaller teams. It’s really, as you mentioned, like, shifting that mindset can be really difficult. And we’ve seen a lot of times if you have a smaller team, especially like with charities, you know, there’s always that pressure that whatever you’re doing, you still need to perform, you still need to bring in the results. And now that we are, like, increasingly not able to rely on the kind of trusted platforms that we knew that worked the best, we’re not able to track the measurements so perfectly. It’s always difficult to, you know, make it might be able to make the argument to the marketing team that you’re working with and they might agree, they know how marketing works and you start showing them, if we’re running some more engagement stuff, this is going to help us. But then it’s really difficult when you essentially, as an advertiser, you have to try and make the argument to the people who like top level, who are looking at they’re just looking at tickets, they’re just looking at you’re putting X amount of money into this, how much you’re getting back. So it’s difficult. But then they’ve definitely done it with some of our clients, which they are certainly bigger kind of agencies, and they have a specific team that kind of do handle their advertising. And for example, we had one client where we ran multiple different suppliers. We ran across the full funnel all the way from kind of broker market display activity all the way down to PPC. And from the get go, we were encouraging the client to look at the full funnel look at how everything plays into what you’re doing. So just by pulling into all of these different suppliers or different platforms, all that data into flight deck, we were kind of every week as we were doing our reporting, looking at performance, we were able to look at the full impact of whatever we were doing across the full funnel, which is really helpful, and it really helps to facilitate that kind of narrative to your client as well. And then the client obviously can take that and bring it forward in their agencies and like, hey, we should be looking at the performance from this point of view. But it’s also really nice because obviously when you mentioned earlier how everything was kind of sometimes it’s done as opposed to campaign analysis and all that. But it’s nice when we can be able to bring all of that in real time into a platform and we can look at it in real time and say, we make this slight optimisation here. The next day we see the impact of that and we can kind of at the same time be reactive in what we’re doing, but then also kind of not just look at the performance of the data in Silo where we’re only looking at how does that affect that one single channel? Yeah, I think because there are so many things you can do with research. As you say, you can optimize within the campaign day by day. If you’ve got a system that if you’ve got your APIs working so that fresh data is turning up every 24 hours, you can optimize within. And then you can also do a bit of an end of campaign look. To say, ideally, what you can give to your client is to say we have done a robust piece of analysis that shows both within the campaign and at the end of the campaign, if you look at it as a whole activity, that these things are definitely linked. For example, above the line and digital or display and search. And that then allows your client to take that kind of almost as a business case upstairs to say next time, next time we’re discussing budgets or for the next quarter campaign. Here is a proven data set that shows that it is worth us investing in these channels here because they benefit these other channels over there. And there is a correlation. And I feel that’s one of the important kind of outcomes of a campaign is to give your marketing client that kind of case study to go to the FD and discuss budgets in light of that. Yeah, I think that also helps in terms of testing. You want to test something new if you’re just looking at we have this sometimes where we obviously want to try something new. We know we have to start testing and experimenting with something else because we can’t just put all our budget into Facebook because it no longer works. Old statement. Yeah, it works but you can’t really prove it in the same way. But when you’re looking at a new channel test and you’re only looking at it from that transactional point of view, it’s really difficult sometimes to prove that it works. But whereas if you kind of place it in the funnel and you look at the rest of the campaign performance at the same time, it’s going to be so much easier to kind of sell in that you’ve tested this and how it’s impacted the whole user journey and the whole user experience, rather than just looking at it in silo. So I think there’s low benefits to kind of almost like taking a step back and looking at it from further back and not like a lot of people are, you know, they love the fact that we are able to be so candidate with data and that’s so great about it. We can see the mini school things and we can track absolutely everything, but it’s not always beneficial and it’s not always like sometimes you almost just need to not touch it for a while because if you’re trying to interfere too much, you do a bit more harm than actually just letting it work for you. Yeah, I feel that from a data hygiene sort of table stakes perspective these days, I think you have to get as much of your data into a single place as possible in order to be able to look at stuff in that holistic way. To take a step back from just looking at a single platform task with a single KPI and step back and look across multichannel if you are running multichannel campaigns to be able to see even if it’s just a sort of simplistic level. We’ve dialed up Channel X and we’re seeing that Channel Y is going up as a result. Is going up as well. And we think that’s a correlation. And then you can keep testing to see if that thesis is actually correct. But to get the ability to do that, you have to be able to sort of get all your data into one place, ideally, which is where tools like Flight, they come into their own because unless you’ve got it in one place, it becomes quite difficult to really do a side by side comparison. Yeah, absolutely. I think maybe just to round off the conversation, maybe a look to the future, how do we feel the reporting landscape is going to go over the next few years? More complex, simpler? What do we think the trends will be? I think we discussed this earlier, but it almost feels like there will be a step backwards into we probably can’t be as granular with our measurement, we can’t be as granular with that data. So it’s almost like taking a step backwards and being looking at advertising. What’s the role of advertising? What are you trying to do? And thinking about kind of what you are aiming to do for the user. You’re trying to build trust, you’re trying to build awareness, you’re trying to engage with the user. So it feels like also there will be different solutions that come in. There was a lot of talk about trying to shift the attention, your focus on attention, attention metrics, trying to shift this universal ideas, different contextual, stuff that we can use, but it’s not going to be as granular as what we’re used to. So it feels like it’s almost going old school, where you’re just going to have to put trust in what you’re doing and shifting. Also not like looking at quick wins, because I feel like now everything is just like you know, you can see, you do an X thing and it’s going to give you an outcome of why. And it’s not going to be that. You’re going to have to take a step back and be like what longer term what we’re aiming to achieve here and then just putting a bit of trust in that. It’s going to happen, it’s not going to happen tomorrow. But obviously once you start looking at the data in a longer term view, you will start to see that trend. Yeah, I think that I think absolutely right. That’s definitely going to we are going to be going far more old school whenever the cookie finally costs its last and dies whenever that is. And so I think that’s one big trend on the one hand in terms of that will force us to actually take it more holistically just because that approach will because the cookies are not going to make us give us an easy, quick win. But I think it will facilitate us taking a more customer centric view and that sort of whole funnel view, like you have to do your upper funnel. And keep an eye on those metrics as they slow down. And I think technology on the other hand will continue to evolve and cheapen and become more widely adopted and become more of a product that everybody uses and not just those companies that can afford it, just the skills will gradually expand as well. More people will have data engineers, people who can build APIs, who can wrangle data cloud data space will get ever cheaper and I think those things will help on the tech side of making intelligent decisions based on data. So I think it’s again, it’s a kind of a push and a pull. We won’t ever get identified. There’s going to be a silver bullet moment where we all go TADA and declare that the war has been won and we can all go and have lunch. I think it’ll be just an ongoing battle and you won’t see Facebook and those guys let their walled gardens break down anytime soon. So I think it’ll just be an ongoing optimization. I suppose it’s optimization, iterate improve, it’s rate improve, test to learn but I think it’s not going to be disaster, it’ll just be an adaptation. And we are nothing if not adaptive in digital. A lovely way of looking at it. Thank you both for your time and your thoughts. It was really interesting to hear both of your perspectives on the topic of reporting. I hope everyone enjoyed listening to the and look out for our next instalment of the Unmuddled podcast.