Beyond The Billboard
00:19:57 · Remi Du Preez
[00:19:57] S3: No, I think it's definitely probably one of the most hot topics at the moment because, as you say, guys are quite comfortable with programmatic insofar. But people want a little bit more. That sort of excitement about EDU is sort of not died down. It's just stabilized. So for me, it's just the measurement is sort of the new space. Our measurement tools are not perfect. And where we want to get to is you want to, again, look at what the internet fraternity did well. And it's scalable across publishers, the same kind of methodologies. And that doesn't quite exist just yet. Right now, if you want to do and there's sort of two elements to it. So let's talk it's like the planning phase. Before I drop my money into this DSP, how many people am I actually going to reach? And the answer is we don't really know. Let's just be perfectly honest. It is exceptionally difficult. Not impossible, but you can do a bit of media maths on some tools that you have at your disposal to have some indication of what you're going to have. So I do think that the industry, the tech players specifically do need to do a little bit more work. And when I say the tech players, again, it's look at the internet fraternity when they started introducing programmatic as a general buying mechanism. They didn't have all the answers. Guys were standing on the sidelines going, hold on, I think I can solve that. And everybody started rushing, developing solutions. And then eventually, something stuck, and it just became the norm. And then that was sort of the way to measure. And then new little features that the market required these things. So right now, nobody has come up with a formidable, scalable solution that can actually help us understand how many people am I going to reach in a campaign before I serve out the ads? Yes, we have our impressions goals, but we don't really know like unique reach and a couple of other things will be important for planning. So there's a bit of a shortfall there. But then sort of how do we serve in market audiences? Now this thing has started working quite well. Not nearly perfect just yet. And this is sort of embedding and using cell phone data. So what happens is all of us have a cell phone, and we subscribe to hundreds of different apps. And within those apps, especially the free ones, mean, they need to make money.
00:22:07 · Remi Du Preez
And the way they make money is that you need to give you a location setting to be able to use the app in many instances. And in that instance, that information is shared with someone in the background who pays for that info and eventually us as clients or media owners, we actually pay for that luxury to track audiences. That's one way that we can currently track audiences, sort of people that have been seen inside certain environments. We want to know which billboards they go pass at whatever time of day. And then sort of, for lack of a better word, spam a billboard because historically, we see these kinds of people. So we're using past behavior to sort of indicate future behavior. And that sort of determines the scheduling. The mechanisms, I don't think are bad at all because they do optimize and refine themselves. However, the sample sizes are still quite small. So you could then refine about 10% of cell phones are normally coming through there. One big barrier is the attic guys want way too much money. So minimum spend is a good couple of R100,000, which then means only the guys with those kinds of budgets can actually spend it. Which if you again, if you look at the digital space, mom and pop shops have the opportunity with a few thousand rand to be able to run maybe not effective campaigns, but they have the tools. We to break down those cost barriers. And then on the flip side, on the other side, you have the measurements, the let's say attribution rather. What do people do when they actually come into contact with an ad? Likewise, SDKs can be used from mobile apps. Were people exposed to a billboard when that ad flighted? There's a timestamp on that billboard. There's a timestamp of your cell phone in front of that billboard. And then did that cell phone show up inside of my customer's store? We don't know if you bought anything, but hey, we saw him inside of the store. It's very likely that something happened there. So that's sort of the general attribution that we're currently working on. When it gets to online attribution, it can now start getting a little bit deeper. The ad tech guys haven't quite nailed that. But there are a couple of lease operators on the side where you can drop in tags and clients' websites within big publisher environments, and then track audiences in that kind of way.
00:24:16 · Remi Du Preez, Livia Brown
But again, they're not really embedded inside of the environment. So guys, media partners, owners, they people are sort of left to their own devices to figure these things out. So we're definitely moving in the right direction. But for it to be accessible to everybody, I think it needs to be available via the DSP that buyers are going to use. And that cost needs to come down a lot. [00:24:37] S0: As the programmatic campaigns are evolving and changing and different media owners are coming on board, I understand that the measurement metric, depending on what formats you have and which environments can change. But what role is the OMC data playing in measurement currently from a programmatic perspective for the media owners that are part of the OMC?
00:24:54 · Remi Du Preez
[00:24:54] S3: So I mean, the OMC data right now is really just used to inform what we call impressions multipliers, or how many people were potentially exposed to the ad when the ad flighted at any given time of the day. The OMC has done great work. I love their However, it is unfortunately just stuck with roadside. And I think they're starting to move into open either mall spaces that may be very soon. So again, out of home is not just roadside and forecourts and malls and airports. There's retail. There's gyms. There's hotel lobbies. There's moving digital inventory. I mean, we are moving into a space imagine when you leave your home, there are screens everywhere. It'd be beautiful to be able to connect all of these in an environment so that the client has an opportunity to follow the consumer really throughout their journey wherever they may be. Now, coming back on the ONC stuff, one little sort of caveat, and I've been advocating for this quite strongly, is that in the position that I am at I have about 25 media partners that we're currently connected in today. We have 1,500 screens. And some of these guys are AMC affiliated. And then I get the impressions data. And everybody's using the same system, but everybody's numbers look different. So everybody's got the same info, but they're just not using the system correctly. And I don't think people are doing it from a place of, call it, fraud. But there's definitely requirements, I think, that we need some standards, even maybe some verification. Nobody's watching what a media owner does when they're embedding this data into their system. When you're going to go buy a campaign, how do you know that there's 100 people outside of that screen versus 50 when the media owner down the road for the same billboard, same size, he says it's 50, they can't say that's But one they both use MC data as an example. So again, it's just media owners have a commercial bias to push the numbers that make sense for them. I think we need a referee on the side to go, we are going to audit this way. This stuff gets embedded in natural fact, and another person is in a bid. Into the system, you get a little blue tick, and everybody knows you. You're buying verified impressions. Not perfect, but at least everybody's on the same field and clients feel safe. Because when they feel safe, they're going to spend more money.
00:27:12 · Livia Brown
[00:27:12] S0: Yeah, 100%. And I think what you've touched on now, and like I said, I know that it is a hot topic and they will please, we invite our audience to please comment via our LinkedIn page through us. But I think that it's an evolving space. And I think that the measurement and the OMC information that's being used as a proxy, like I said, there are a lot of different other types. But your point is that sometimes the data is different is the hottest point that I've had the most comments for around programmatic digital out of home. People have stopped asking like, what is it? And they're more like, how is it measured? And then often a tirade carries on in that space around some of the challenges in that space. And like I said, I think also to the point is that there are some formats like retail or hotel lobby that are not measured from an OMC perspective. So how does that how does that play out? Maybe it is a discussion point to have from an industry perspective is that there needs to be like a, I could say, a third party or a referee to say like, this is verified so that everyone's playing with equal cards almost.
00:28:09 · Remi Du Preez, Kirsty Carlson
[00:28:09] S3: Last sort of comment I had, Liv, was, you know, we need to start moving into some bit more real time analytics. At some of the measurement bodies overseas, they've even started incorporating real time analytics now. That makes sense for the market. But we need to start moving into the real time analytics space. Like the survey data is fantastic and it's fine. But if you're not going to start moving into real time analytics, we're going to start excluding a lot of environments that absolutely will add value to clients' lives as well. [00:28:35] S1: Yeah, I completely agree, Remi. Particularly for retail environments, I think that would be really, really beneficial. What I wanted to ask you Remi, just because I want to close with this question, is what has been your best programmatic design of home or most effective or favorite programmatic design of home campaign you have seen or executed in South Africa?