Having built his first ad network while he was still at high school, Index Exchange chief executive officer Andrew Casale can justifiably lay claim to being one of the foremost commentators on adtech. Albeit, his stereotypical Canadian modesty would preclude him from making that claim himself.
Speaking with The Drum on the sidelines of his company's inaugural partner event in Miami, Florida, Casale addresses a number of issues facing the online media trading market. The discussion includes: market inefficiencies caused by 'the adtech tax'; how the demand-side of the market is starting to cull its partnerships to help drive more value; and how the now ubiquitous 'year of mobile' is likely to impact product innovation.
The Drum: There is a lot of talk about change in the digital media market, particularly in the area of adtech. What do you think that means for publishers in particular?
Andrew Casale: I think one of the issues that media companies and buyers consistently say is that the adtech tax is too high, which means that building a sustainable business in the programmatic economy is challenging because all of the participants consume some of the media dollar along the way. They’re also a little frustrated at the lack of transparency that exists in a lot of business models.
A change has occurred in the market which we think has created a recipe with all of the ingredients required for efficiency to rise, and it feels that this change is heavily desired by the media companies. That was started by the wrapper, which at the core of its purpose was designed to reset who runs the publisher’s programmatic marketplace [and also helped address concerns over latency and header bidding].
Historically, the publisher would give the keys of their programmatic business to one vendor, and that means they’d say: ‘X is my exchange, Y is my SSP [supply-side platform], now try and get the most out of the market for me.’
They’d then have to hope that those businesses would drive the most value possible. However, the wrapper has changed that. Rather than outsourcing, the publisher should now be in control of their own marketplace, and shouldn’t give anyone that full autonomy, and in doing so, they should also gain a lot more transparency.
So rather than saying one party runs a publisher’s market, the publishers can now work with multiple participants, and have those participants become inputs into a market that they operate. What that has effectively done is remove what we used to call the waterfall, and align it completely horizontally.
Now we have many companies like ourselves all sitting next to each other on a competitive basis endeavoring to drive more of the dollar to the media company. I believe that’s a recipe for consolidation to occur and efficiency to rise, because we all want to win and out-price each other to help transact with that publisher.
Then if you add on to that another factor-in why the wrapper was created is that it has created a massive increase in requests, which are very duplicative. What that means is that if you’re a publisher that works with five exchanges in a wrapper, then those five have access to the same wrapper simultaneously.
That also means you’ve also multiplied the cost to transact by a pretty significant degree, and I think that’s another factor that’s going to drive consolidation, or effectively the commoditization of this business model. It’s going to be unsustainable for that duplication to continue to occur, and a correction is about to be driven into the market.
I also think that’s driving some of the changes that we’ve seen in the industry already, particularly on the supply-side. We’ve had companies big and small, changing dramatically on account of what the wrapper has done.
The Drum: So if the move towards header bidding (which has largely been facilitated by the wrapper) means multiple adtech companies are competing when it comes to assisting the sale of publishers’ media, doesn’t that drive up costs on their behalf? How do you think the buy-side reacting to that?
Andrew Casale: Request load and the increase in duplication isn’t just having an impact at exchange level, it’s also affecting the downstream DSP [demand-side platform], and absolutely that listening cost from the increased request load has gone up dramatically.
So what’s the next step, and how are they going to respond? I believe they’re already starting to, and what the buy-side is starting to recognize is that they can now start to compare all of these routes and access points to the same impression.
So I can now say that one route gives me a better win-rate, but maybe it’s a little more expensive, compared to an alternative route which almost never transacts and has loads of duplication, with very little value. Or maybe I there's even an alternative route has both good and bad, with premium inventory and some sketchy things in there, so they’ll be looking at things like ad quality.
I think that in turn, DSPs are starting to reduce or pick the routes that are most effective for their business, or their clients needs.
The Drum: Does that mean consolidation of the supply chain in your opinion, and how much of that do you foresee over the next 12 months in that respect?
Andrew Casale: Exactly, so right now we’ve gone from having no contrast, to where we had one vendor running a publisher’s market, to a place where we have too much contrast.
Right now we are going to see a lot of people narrow-in that vendor count to figure out what’s best for their business and what’s sustainable so that listening cost stops growing the way it has been.
In a conversation I had at dinner recently I heard one DSP say that today they have 94 exchange connections – that’s a lot of duplication – and they said that they’re looking to reduce that to single digits. You can draw from that the exact number, but that is a pretty substantial change.
Their rationale was very clear, every one of those parties has a very different standard and very different taxes and inventory quality volumes, and their listening costs have grown dramatically – they need to start to curb it. And I think that’s a pretty true depiction of a lot of the other platforms, and what they’re going through right now.
The Drum: Is there a fork in the road between publishers opting between open auctions and striking direct deals, such as private marketplaces (PMP)?
Andrew Casale: I think that the open auction has already started to improve, and will continue. Even as we reduce the participant count of exchanges, and implement more controls on the open market (with things like TAG and the trusted supply chain effort) we’re going to find that the open market is going to continue to get better, and continue to be more trusted.
However, many people are recognizing that 80% of their spend is directed towards a relatively small number of publishers, and those are publishers that they know their brands want.
So I think that PMPs and audience guarantees (or some form of a deal) will continue to command more-and-more of that, because I think it’s going to be insertion order, manual deals moving in to the programmatic pipes, versus just money pouring into the open market.
The open market isn’t going anywhere and it will improve, but I also think that there are few in the open market that represent the vast majority of media spend, so that would represent us having more intelligent deals going forward, I don’t think that’s going to stop.
The Drum: So how can publishers be more smart when it comes to their use of adtech?
Andrew Casale: When things like PMPs first emerged, publishers gave those opportunities to their sales teams. They then got excited about the idea of doing deals, and this basically led to people doing too many deals, plus deals that weren’t designed to do anything conducive to value generation, and this just generated complexity.
But what we’re starting to see publishers do is recognize that if you’ve got hundreds and hundreds of deals that you’re either killing it [in a good way] or that you have deals that aren’t of any value. So now a lot of them are starting to cull them down, and making decisions between what deals that are productive, and which ones are driving value for the buyer, then starting to learn from those deals so they can create more.
So what you’re starting to see is publishers look at things, and think: 'Okay, if I had 500 [trading] deals, and only two of those produce any money, maybe I should just have two deals in place?'
From there, they’re starting to think: 'what can I learn from those two deals, so maybe I can make 20 more that are more conducive to value-generation, rather than 500 more?'
The Drum: The latest IAB figures show that mobile now accounts for over 50% of all media spend. What are your thoughts on that?
Andrew Casale: This is the tipping point, every one’s been talking about it for years and now we’re starting to realize that ‘the year of mobile’ is forever, and it’s no longer now because it is the superior market, similar to how TV crossed over with digital.
It’s going to mean sweeping changes in the adtech space and where platforms focus their innovation and efforts, because if you’re focused on desktop, then you’re in a minority business that’s in decline.
The Drum: If we also look at the latest analysis of those numbers from Pivotal Research’s Brian Wieser, he says that Facebook and Google account for the majority of that increase in digital spend. What’s your opinion on that?
Andrew Casale: A lot of that number [the increase in mobile spend] is going to be Facebook, and all I can say is that one positive in the ecosystem that is changing, is that Facebook is opening its wallet more to publishers. Its recent move to open up its FAN product (Facebook audience network) to header bidding, which means that some of the historical ad spend that would have been on Facebook is now eligible to go to other publishers.
Another good point from that is that it’s good to have both the major titans effectively competing with each other to get inventory on publishers sites, which should effectively drive CPMs up, which is the biggest concern. That’s pretty important.
Click here for more of Index Exchange's advice on header bidding, including consideration to be taken with the migration from client-side auctions to server-to-server