Key Takeaway

  • A Turbulent Ad Tech Landscape: Publishers everywhere are experiencing eCPM degradation, with trends varying across platforms and strategies.
  • Quality is King: Efficiency and quality are increasingly important, with buyers rewarding publishers who provide value and clean inventory.
  • The Cookie Saga Continues: Cookie degradation is ongoing, prompting publishers to explore alternative ID solutions and focus on authenticated user bases.
  • Increased Curation: Curation and Deal ID-based transactions are gaining popularity as ways to cut through noise and improve targeting.
  • Consolidate to Make Your Voice Heard: Industry consolidation is an emerging trend, with publishers benefiting from aligning with larger aggregators or sales houses.

Are you a publisher feeling the squeeze of changing ad tech trends? If so, you’re not alone. 

As we enter 2025, the digital advertising landscape is shifting beneath our feet. Staying ahead of the curve has never been more crucial for your bottom line, whether you’re dealing with fluctuating eCPMs or navigating the murky waters of cookie deprecation. 

In this edition of Playwire Live, we’ve brought together a panel of industry experts to break down the major trends shaping the programmatic ad landscape in 2024 and beyond, and more importantly, to offer practical strategies for thriving in a rapidly evolving ad tech ecosystem.

Scott Schroeder, Playwire SVP of Yield Operations is joined by Ashley Wheeler, SVP at Magnite, overseeing their DV+ platform, and Playwire VP of Yield Optimization, Myles Engler.

Watch the discussion below, or keep scrolling to read along.

 

eCPM Degradation and How Publishers Are Responding

Scott: The topic du jour will be major trends we're noticing in 2024. We're going to have a casual conversation around that, and then we're also going to speak about how publishers can best manage these trends that we're seeing to not only protect but grow their revenue. 

So without further ado, we have a couple of topics we want to dive into. The first will be the general trend of eCPM degradation. Ashley, what do you think? Can you talk a little bit about what you're seeing when it comes to display and online video CPMs?

Ashley: One of the benefits of working at a supply-side platform (SSP) is that you get to see all of these trends in aggregate, and start to normalize and understand a little bit about what's driving them. One of the prevailing things we've seen not just recently, but over the last couple of years has been this gradual degradation of CPMs. 

It has played out in a couple of different ways. The most noticeable is that there are just a bunch of DSPs that have been very upfront about the fact that they use floors as guides now. They fit below floors, they take them as advice, but they don't really follow them. And there's probably a lot of different factors that are driving that change.

Buyers are increasingly looking for efficiency, and that's always going to be the case. I also think there's just a lot of noise in the ecosystem, so there's a surplus of supply. A lot of it's great, but some of it's not as great. And I think buyers are sort of building that into pricing decisions, particularly when they're looking for specific audiences. 

I also think there are systemic complexities that make it difficult for publishers to manage floors. So if you think about flooring strategies, GAM, and UPRs, they don't always talk to Prebid. Sometimes, depending on how complex your flooring strategy is, it's difficult to go in and maintain that and maintain consistency across all your partners. So I definitely think that complexity adds to some of this.

But overall, we've seen this play out the last few years, display and video CPMs specifically have been on this downward trajectory.

Have you seen that from the publisher's side?

Scott: Myles, what do you think about the CPM trends we're seeing?

Myles: CPMs in general for us this year have been growing quite heavily, actually, particularly on the display side, but that's the result of a very specific quality initiative that we built. 

But Ashley, to your previous point, the industry from the buy-side feels like it's gearing up to transact on this concept of efficiency, which has not historically been the primary method with which buyers and sellers introduce themselves to each other. And so, net-net, if you're doing the same things you were doing to monetize your websites in 2022, 2021, or previously, you're not really going to see growth this year in the core unit metrics like CPM.

So really what's going on, in my opinion, is the buy-side is asking for some efficiency to come back to the auction from the sell-side. There’s a bunch of different things here like auction duplication; having the same SSP partner calling the same DSP partner 60 or 70 times for the same opportunity, the buy-side wants us to knock that off because, from a cost perspective, they're just lighting money on fire at that point. 

Publishers employing older monetization strategies are going to see a slow, continual decline. But for publishers in general, who are really focused on the true value exchange with the advertisers, there's an opportunity for large growth this year, particularly over last year because for most, 2023 was fairly painful. 

If you adapted your strategy, it has not been that painful this year. However, if you're still using your old tricks, my guess is your CPMs are quite a bit lower than last year.

Ashley: It kind of fits a broader theme that we'll talk about. This idea that we're finally at this point in the industry where we're being incentivized to do the right things, I think is a really important thing. And it's great to see that finally start to come to fruition where focusing on the premium supply, doing the right things, and making sure that you're building towards those buyer outcomes is what's ultimately going to help publishers drive more revenue in the future.

Scott: That's a theme that'll be coming up quite a bit. So if you're a publisher and you're aware of inefficiencies from SSPs or DSPs, you're aware that DSPs are trying to access your inventory below the price points you're trying to set in the auction, what do you do? What strategies can a publisher employ to combat this?

Myles: Zooming out at a high level, anytime you have a user coming to your webpage, you're essentially introducing advertisers to that user by loading ads on the page. And if you zoom out and ask the question, is it reasonable to expect the advertiser can return their investment on buying the ad based on how you're injecting the ad, where you're injecting the ad, and what your users are doing on the site? You will tend to have good outcomes if that's your North Star. 

You don't want to create those connections to each user if you know there's no shot that an advertiser is going to return. 

A good way to think about this is, let's say you have a blog about running, right? About track and field. And let's say on any given day, two users visit your site in the market to buy a pair of running shoes. If you only send one click to a shoe company like Nike or Adidas or whoever, then you've under-monetized your visits on that page. But if you send three or four clicks to those shoe companies and there are only two users inherently who are willing to buy shoes, you've degraded the value of their ad buys. 

Really thinking about the tactics and strategies utilized on your page in terms of when, where, and why you're placing the ads to maximize the relationship between that particular user and advertisements that they may see or advertisers they may get introduced to, is ultimately going to be a winning strategy.

Ashley: I would just add that you should try to be really buttoned up in your flooring strategy, ensuring that when you set a price floor, you're setting it across all partners equally, and you're not creating that ability for leakage or for buyers to find it cheaper through here versus another place. 

That's a hard thing to do, but I think that's actually a really important thing to do because buyers are bidding below floors, particularly if you've got direct integrations with them. You want to ensure that you've got these consistent floors in place so that you're upholding the value of your supply.

Myles: Yeah, that's a very good point. When I zoom out on floors specifically, flooring strategy is not really about maximizing the value of any particular impression. It's about training a vast number of algorithms around what the general value of your audience and users is. 

Your goal with your flooring strategy should not be to eke out an extra $0.01 or two-cent eCPM on any given impression. It should really be about rightsizing the value exchange between the advertisers and you as a publisher. If you think about it that way, you're not trying to get advertisers to pay the highest possible price on any given impression. 

What you're trying to do is say, the inherent value of my users on display is $3, but I'm getting bid back at a dollar. What are a flooring strategy and inventory strategy that I can implement to try and bridge that gap between what I'm getting paid and what the inherent value is for those users? 

And if you step back, you're trying to do that across thousands of different algorithms. Every DSP has a different bidding algorithm. And so when you're stepping back as a publisher and thinking about your flooring strategy, really what you want to be doing is twofold. 

You want to be thinking about your inventory quality for the advertisers and then through flooring. How do you train the advertisers about the increased quality that they're getting? 

And they will come as you improve metrics for them.

Ashley: Yeah, I agree with that wholeheartedly.

Scott: That was well put. If you feed SSPs and DSPs what they want to eat, your inventory quality will remain clean. You can afford to use flooring tools as a mechanism to train algorithms. 

Overcoming Cookie Degradation

Scott: What trends are you seeing around cookie degradation this year and what are publishers doing to combat it with magnet?

Ashley: Degradation is different from deprecation. Because now we're in this new world where they're (cookies) not going away, but they are going to continue to degrade over time. And that's not any different from the trends that we've seen. 

We've already seen a huge degradation of cookies and addressability in the open Internet that's going to continue. It might have moments in time when it accelerates. We're all kind of still sitting and waiting to see how this all unfolds. 

But in general, what I've seen is when a request is not addressable, it mostly impacts fill. It's not so much a CPM issue as it is that these aren't as valuable to buyers. They don't bid on them as much.

We have certainly seen, as this degradation has happened over time, more and more publishers who are working with alternate ID partners. We've really seen a rise in alternate ID partners. It's been really interesting to watch, actually. Individual DSPs all have an alternate ID. And so I think we're starting to. We have kind of been testing for years to see which ones drive lift. 

I think we're starting to have a much better sense now of which ones of these IDs are actually valuable and which ones buyers and DSPs are listening to. I also think, and we'll talk about this a little bit later, the focus on an authenticated user base is going to be critical. These are the truest alternatives of a deterministic identifier that we're going to be able to find in a cookie-less or less-cookie ecosystem.

As publishers start to think about their long-term strategy, it's not going to be a one-size-fits-all silver bullet. It's going to be a mosaic approach where you're working across a bunch of different solutions because you want to be able to transact however your buyers want to transact with you. 

Focusing on that authenticated user base is important. But also as cookies degrade, there's going to be this opening for some of these more premium differentiated types of audiences, which are first-party data contextual. 

If we can do that at scale, in standardized ways, I think that adds a really nice solution to the portfolio of things that publishers and buyers can choose to transact with one another on.

And that's sort of what we're seeing. There's been a lot of change in this space, even recently. As we adjust to the changes and watch how this all unfolds, being nimble is certainly going to be important. 

But I also think it doesn't really change much. Because cookies are still going away and we are moving into this more privacy-centric world. So we need to come up with these more privacy-centric solutions in order to transact with one another.

Scott: You started with IDs, and that's a very interesting one. Essentially what we saw with Google's announcements early this year and late last year was publishers scrambling to find what works in a world without third-party cookies, including us here at Playwire. 

We're all throwing stuff at the wall through A/B testing to understand what is working, and what isn't working. Are there slices of inventory that are a little bit better to put through an ID and so on? That was the common theme I saw early this year: a lot of publishers scrambling and testing different ideas. We even saw some attempts at very early productization around solutions aimed to address this. 

But the reality is we never really saw the efficacy of those things. So testing is really how we're looking at this as well.

Ashley: Yeah, but testing is tricky. We've struggled with it on our end, too. 

It's somewhat hard to separate one singular EID from the rest of them and to understand which one of these is actually driving value. That's been a challenge. 

We're lucky that we've got some pretty good A/B testing tools, particularly through our demand manager integrations. And we've been able to narrow in on there. But, from the publisher's perspective, I can imagine that it's not just a straightforward way to understand which of these are valuable.

Scott: No, it is not easy. Reporting on IDs, unless you can construct your own config-level split testing, getting actual information that's not your own, is difficult to come by, honestly. 

Then you mentioned something else. Beyond IDs, there's this idea that third-party cookies are going away. How should publishers think about emphasis on this? 

In our case, we are aware of it and continuing testing, but it's not necessarily the fire that we were trying to jump on back in April (2024). I still think it's worth the investment for publishers to continue exploring other options.

Ashley: Oh yeah, I 100% agree with that. The issue of addressability is not going away. The landscape has changed a little bit, but the investments that you were making before should by and large still be the same. 

On our end, our focus is on Magnite audiences, which are these standardized segments that we can deploy at scale and work with buyers through curated packages to transact with publishers or with contextual. I think that's going to continue to be really important. And I think that's a point of differentiation and value. 

To Myles’s earlier point, creating value and outcomes for buyers, this is an important thing that we can offer in the ecosystem that provides differentiated value and those outcomes.

As it relates to some of the privacy sandbox work that we're all working on, we can take the gas off of that and see how that solution unfolds and evolves. Certainly, the timelines and the speed with which we needed to execute against something like that have eased. 

But even then, if there's a viable solution that's going to help, with valuable data that will help buyers achieve outcomes that derive from that solution, I feel like we as an industry will be all in on supporting it. At least from the SSP perspective, it will always be a portfolio approach. Our job is to support all of the ways that buyers and sellers would want to work together.

Having this sort of library of things that everybody can choose from in order to affect those best outcomes is what's going to be most important.

Curation and Efficiency in Ad Buying

Scott: You mentioned curation, too. That is a trend I've seen really become popular in the last two months. I think with Google's announcement initially, it's like, “All right, how fast can we get our contextual product up? How fast can we scrape together demand and then sell it at scale?” 

We jumped through a lot of hoops to get into position to have that set up. We didn't really see the dollars flowing through it early on. Now it seems like buyers are more leaned into this idea. Have you noticed that on the Magnite platform?

Ashley: Oh, yeah. We have a huge team that's invested in curating supply and working with our buyers again to achieve those outcomes, to really, as we were saying earlier, cut down on that noise. 

Curation has been an effective tool for targeting certain publisher lists and certain KPIs. It's actually been a really helpful benchmark for us to share with publishers. Like, this is what buyers are looking for, these types of KPIs, this viewability threshold. And it helps set that standard that the industry then needs to get behind. And there's money behind it, which is the best part.

If you're able to create these packages that perform well and deliver on the value of the supply, it's kind of a win for all parties. And so we have certainly seen a massive uptick in the curated deals that transact through Magnite. 

That's something that we are actively investing in. We've got a whole demand facility team. It's like 100 people who are working through that on the demand side and then obviously on the supply side. We work very closely with our publishers to create that bridge and to make sure that we're informing our partners of what's important, what buyers are looking for, and how we can continue to create value within the ad tech ecosystem.

Scott: Yeah, it's perfect. And you know what? At the risk of sounding like a homer, self-service here — on the platforms that do curate demand well, it can be pretty high. 

For example, we saw curated PMPs (private marketplaces) go from about 9% or 10% of our total revenue at low viewability times. After the efficiency initiatives, we saw sites reach five-zero percent, 50% curated. And Magnite is one of the platforms where we see really strong curation. So I'm fully bought into that idea.

Myles: It makes sense because it's one of the main tools that the buy side has to quickly and efficiently cut out noise. If you have a group of 100 sites and you're not hitting your KPIs for the media buy, the easiest thing to do is to cut out the handful of sites that are performing the lowest on that. 

And so when you look at the total open marketplace as a whole, the easiest mechanism that buyers currently have to expand what they're looking at, so to speak, is to pick a handful of publishing partners or a large handful of publishing partners and look at everything they have.

Because the other option is to look at everything in the open market and throttle what they're looking at so they don't get drowned in the noise of low quality. The buyers have a big incentive to use curated deals. 

Now that there are dollars behind it in a more meaningful way than a few years ago, it aligns the incentive on publishers to deliver the metrics with their advertising that those buyers are looking for. So it's a great tool that the buy side has to hold the publishing side accountable to make the value exchange beneficial for buyers.

Ashley: Transacting these curated deals, at the end of the day, they're Deal ID-based transactions. They're auction packages. They’re different from your typical pub-managed deals, but they also give buyers a mechanism to control and direct spending a little bit better. 

Open-market, you work with the DSP algorithms and what those algorithms are tied to. But within this space as a buyer, you can actually leverage your weight a little bit more. You can cut more strategic deals, you can lean in more heavily with partners that deliver value, and then this gives them a mechanism to be able to direct that spend. So if there are spend thresholds tied to certain deals, it's easier to achieve that, it's easier to track that.

And so I think we're sort of, we're seeing a bunch of different trends that are driving into this broader one, which is curation. But there are a lot of different reasons right now why these are really beneficial tools in our ecosystem. And we've seen it accelerate a ton this year. I can only imagine that trend is going to continue.

Scott: And also a quick reminder, if you are a publisher, now is a really good time to make sure your signal fidelity is in order. Your signals are healthy, they're being passed and ingested correctly by SSPs. Very important. 

But Ashley, just to put a cap on your last point there, this is the second topic we've talked about where efficiency and clearing out noise is becoming very important. The Deal ID-based transaction is a very, what's the right word? Romantic way to purchase inventory. It kind of skirts outside of the noise that elsewhere exists.

Ashley: It actually speaks to a broader trend of consolidation, as we think about some of the inefficiencies in the ecosystem. And it's not just buyers who are looking to cut down on that noise. I'm sure as a publisher you're looking, you're looking to lean in more heavily with a few strategic partners. 

From the DSP side, they're looking at specific publisher lists now. Direct integrations are a mechanism for consolidation. We're seeing this in a bunch of different areas of the business and it all plays out in different ways, but it speaks to the same trend. 

Five, or six years ago, header bidding came onto the scene, and all of a sudden we saw this absolute inundation of supply. And we've seen a lot of that proliferate since then. Now, we're coming full circle where it's like, “These are the partners that add value. This is how they can add value. And these are really where I want to direct my dollars or where I want to lean in heavily in that partnership.” 

That's what we're starting to see play out, which is quite frankly, a healthy thing for our industry. We're finally being incentivized to do the right thing, which is a great place to be in.

Scott: I can say this year publishers are actually getting rewarded for having desirable behavior, more than ever. At least that's being felt in the numbers. But you did mention consolidation. The six-month initiative that we took to clean our inventory also had implications for demand. 

In earlier days, let's say five years ago, theoretically, you could add a new SSP, and that SSP might be 1% incremental. You could end up adding 100 SSPs and getting 100, you know, single points of incremental revenue. But at what cost? 

I think that's what we started listening to this year. What implications exist if you run 30 SSPs in your private stack? It's not good. You're duplicating bids to DSPs that have QPS limits. You're costing compute money to process the bid requests.

So that's one of the things that we did last year; we cut 40% of our SSPs.

Ashley: Did you see a major hit to revenue in doing so?

Myles: In the short term, it was 1%.

Ashley: I think we were talking about flooring strategy. You're training things over time. It’s absolutely the same thing here. It takes weeks and months for DSP algorithms to really hone in and hit a sweet spot once you make a change that drastic. 

As we consolidate, as we start to do more of the right things, you're rewarded for it, and then that just helps to fight against some of the other trends, like CPM degradation. If there's less noise and then what you're presenting to the market is more valuable, your CPMs will be more resilient as a result.

Myles: That's been one of the interesting things that's come out of the initiative is looking at typical seasonal patterns and how our predictions on demand fall off when that typically happens. 

Let's take the end of the month and the end of the quarter, for example. It used to be, ten years ago, that your demand fell off half a day before the month ended, or half a day before the quarter ended. And then you zoom out to 2023, and your demand is shrinking away like a day and a half, sometimes two days before the month ends and the quarter ends. That's a direct result of the glut of waste – spending the money before the IO is over, technically. 

And one of the things that we noticed as we went through our efficiency initiative, which internally we call QPT, is that our CPMs and our seasonal patterns are much more resilient up to the changeover points. We see stronger performance than we did last year on the last two days of the month. And the drops that we get when a quarter turns over or when traffic surges around a holiday or dies around a holiday, a lot of those things are ending up much more resilient. 

The theory around that is that we've reached a better equilibrium on displaying the true value of a user to advertisers who actually want to buy that specific user. And so there's less of a reason for those budgets to pull back or not start as quickly on the change of a quarter or change of a month.

Demand Consolidation and the Future of Online Advertising

Myles: Most industries consolidate over time as waste, and for lack of a better term, arbitrage, opportunities dry up as an industry matures. So from a consolidation standpoint, there are some significant challenges, a lot of them self-created by publishers, that just exist in the industry. 

A good example of this: It’s fairly expensive on a unit economics term to listen for an ad opportunity and then try and win it. If you're going to try and win that opportunity as a bidder, you're competing against thousands, millions of advertisers and campaigns and line items via thousands of DSPs. And if you're a publisher, every publisher is unique and every publisher is special. But also there's like millions of publishers out there. 

So there are literally tens of millions of opportunities to buy an ad every second. And most DSPs, outside of a handful, cannot physically listen to that at any given time. It’s very simple, like the dogs at a water bowl analogy: your water bowl is finite. Everyone's trying to drink water out of it. If you are a big, healthy, likable, strong dog, you're going to have a seat at that water bowl. And if you're loud, obnoxious, don't provide value, and don't return any benefit to the pack, you're not going to have a seat at that water bowl. 

So, what does this mean for publishers? Let's say you're a publisher that's small, medium, or even large to some degree. There's a limit to how big you can be as a single publisher. If you were to pool your resources with a bunch of other like-minded publishers, all of a sudden you're a big dog at that dog bowl. And all of the DSPs are more incentivized to listen to you because they can get more of their water through to you. You just have a bigger mouth and a seat at the table. You're not fighting for scraps. And so as time goes on, we'll see continued consolidation. 

The flavor that it's taking right now, and hopefully this doesn't sound too self-serving, is with sales houses. If you're a sales house like Playwire, your seat at the dog bowl by definition is bigger than even most premium publishers that are out there, just from the scale of requests, the scale of quality requests. We’re going to be larger because at the end of the day, we are an aggregator. 

Aligning yourself with an aggregator can be a very good shortcut for publishers to get a better seat at the dog bowl, to get better demand, fresher water, so to speak. And so as time goes on, we'll just continue seeing consolidation.

Consolidation ultimately is scary, but on the publisher side, it is going to reward publishers who make sure they're doing things above board and following what advertisers want. Because the advertisers ultimately are the ones filling the bowl and deciding which dogs get to drink and which dogs don't get to drink. So, aligning yourself with sort of the biggest dog that is returning what the advertisers want is going to be a very solid long-term strategy. 

Ashley: Particularly some of these solutions that you really do need to employ get more and more technically complex. We've talked about a whole range of different things, flooring strategies, audience strategies – there's a lot, and some of it is fairly technical. So it does make a lot of sense as this ecosystem consolidates, as there is this focus on leaning in more heavily with fewer partners and fewer partners who are doing things right. 

That consolidation, even within the publisher landscape, is something that we're seeing a bigger and bigger trend. And it is something that I think is very helpful, particularly for mid to long-tail publishers, to be able to have that strength in numbers to come to the table with full scale, with premium supply. That's a really important point.

Myles: At the end of the day, if an advertiser has a dollar to bid and you improve their efficiency, they're not going to spend the whole dollar. It's not like you create more efficiency and they're like, oh, I only want to advertise with $0.80 now. They want to spend the full dollar. That's what their marketing budgets allow. And so, as consolidation happens, if you take an approach to consolidate around efficiency, your share of that single dollar will go up because the advertisers aren't going to pull change out of that dollar just because efficiency comes. In fact, if you actually create a lot more efficiency, they might turn around tomorrow and decide to spend two dollars instead of one dollar.

And so really, how the industry is consolidating on the publisher side, how it's consolidating on the sales house, SSP and DSP side is still very much in line with this idea that the demand side is asking for some efficiency back and they're very willing to pay for it when they see it.

Ashley: The theme remains the same. Like, you're incentivized to do the right thing, which is a great spot for our industry to be in.

Scott: It’s a good way to explain to a publisher that might be scared of consolidation, that it doesn't have to be scary. If you have deep relationships and you're focused on the challenge of how to get more with less, then you'll do fine. 

Ashley, I wanted to circle back to one topic, which is the increase in Deal ID transactions. We spoke a little bit about curation. What are you seeing at Magnite? What's driving the shift away from open market spending into more DLID spend?

Ashley: It's just this ability to have these targeted publisher lists and to be able to overlay valuable KPIs on top of them, to cut through the noise, to be able to layer things that provide unique and differentiated value, like audiences and contextualization, and then again, to be able to channel spend and control where budgets go. 

You can create these stronger, more leaned-in partnerships, and really drive value just from the fact that you're working with fewer.

New call-to-action