Every time a demand-side platform buys a supply-side platform, or a data broker merges with a measurement firm, the press release trots out the same line: “better outcomes for advertisers.” The pitch is simple—fewer handoffs, less latency, tighter integration. But if you actually dig into auction mechanics, log-level data, and the incentive structures baked into these consolidated stacks, you start seeing something else. The platform itself wins first. It captures the upside through reduced competition, opaque pricing, and the ability to internalize arbitrage opportunities that used to be kept in check by separate, competing entities. This article walks through the observable mechanics behind that claim, leaning on documented platform behaviors, auction dynamics, and the structural shifts that have quietly tilted the playing field.
The Consolidation Wave: A Quick Map of Where We Are
Over the past decade, the programmatic supply chain has compressed from a sprawling mess of independent point solutions into a handful of vertically integrated stacks. Google’s end-to-end ownership—ad serving (Google Ad Manager), demand (Google Ads, DV360), and exchange (AdX)—is the most cited example, but it’s hardly the only one. The Trade Desk has built proprietary identity and data tools that cut reliance on third-party DMPs. Magnite merged with SpotX and Telaria to combine CTV supply with a full SSP feature set. Amazon’s advertising business ties demand directly to its own retail data, bypassing traditional audience segments. Even smaller players like AdTheorent or Viant have scooped up DSP and data capabilities to create closed-loop systems.
Each of these moves gets framed as an efficiency play: fewer hops, better data matching, lower latency. And on a purely technical level, that’s often true. A consolidated stack can match identifiers across the bid stream without relying on cookie syncs that fail 40–60% of the time in open-web environments. But the efficiency narrative masks a more fundamental shift—who controls the pricing levers and who gets to see the true cost structure.
How Vertical Integration Changes Auction Dynamics
In a fragmented ecosystem, an advertiser’s DSP bids into an independent exchange, which runs an auction among multiple SSPs, each representing different publishers. The separation creates natural checks: if one SSP takes an outsized margin, buyers can shift spend to another exchange where the same inventory is available. Publishers can route inventory to multiple SSPs and compare net CPMs. The friction is real—latency, sync loss, discrepancies—but the competitive tension keeps take rates bounded.
When a single entity owns the DSP, the exchange, and the publisher ad server, that tension dissolves. The platform can run a unified auction where it sees both the buy-side bid and the sell-side floor simultaneously. In a fragmented system, the SSP doesn’t know the buyer’s true bid; it only sees the cleared price after the DSP has applied its own margin. In a unified stack, the platform knows both sides and can optimize for its own total revenue rather than for either the buyer’s ROI or the publisher’s yield.
First-Price vs. Second-Price: The Shift That Changed Everything
The industry’s move to first-price auctions was supposed to increase transparency. In a second-price auction, the winning bidder pays one cent more than the second-highest bid. DSPs could bid high to win but pay less, which encouraged bid shading—algorithms that reduced bids to avoid overpaying. Publishers hated this because it depressed yields. Buyers hated it because it was a black box. The shift to first-price auctions, where you pay what you bid, was meant to eliminate the shading game.
But in a consolidated stack, first-price auctions create a new problem. The platform running the auction can see exactly how much a buyer is willing to pay and can set floors dynamically to capture that surplus. Google’s move to unified first-price auctions in Google Ad Manager, combined with its ownership of the buy-side via DV360, means it can optimize floors based on historical bid data from its own DSP. An independent SSP doesn’t have that visibility. The result is that floors rise to extract more from buyers, but the buyers don’t necessarily see worse performance metrics—because the platform can also optimize delivery to make it look like ROI is holding steady. The margin just shifts from the advertiser’s pocket to the platform’s.
Data Ownership and the Walled Garden Effect
Consolidation also changes who owns the data and how it can be used. When a DSP and a DMP are separate companies, the DMP has an incentive to make its segments available across multiple DSPs. That keeps the DMP honest—if its data doesn’t perform, buyers switch. When the DMP is absorbed into a platform, the data becomes a competitive moat. The platform can offer “proprietary audiences” that aren’t available elsewhere, which sounds great for advertisers until you realize they can’t take that data to another platform to compare performance.
This creates a lock-in effect that’s well-documented in platform economics but rarely discussed in adtech earnings calls. An advertiser that builds its targeting strategy around a platform’s proprietary data faces switching costs that go far beyond technical integration. The historical performance data, the audience models, the optimization algorithms—all of it stays behind. The platform knows this and prices accordingly. Take rates on proprietary data segments are consistently higher than on open-exchange inventory, and the performance differential is often impossible to verify independently because the platform controls the measurement.
The Measurement Conflict of Interest
This brings us to one of the most underappreciated consequences of consolidation: the platform that executes your media also grades its own homework. When a DSP acquires or builds an attribution product, it gains the ability to shape the narrative around campaign performance. This isn’t necessarily malicious—attribution is genuinely hard, and every model makes assumptions. But those assumptions can be tuned to favor the platform’s inventory or bidding logic.
Consider view-through attribution windows. A platform that owns both the buy-side and the measurement tool can set default view-through windows that capture more conversions and make campaigns look more efficient. Independent verification firms like Integral Ad Science or DoubleVerify can audit some of this, but they can’t see inside the platform’s auction dynamics. They can tell you if an ad was viewable or served in a brand-safe environment; they can’t tell you if the platform inflated the clearing price by 15% because it knew your bid ceiling from its own DSP.
What the Logs Actually Show
If you have access to raw auction logs—and most advertisers don’t, because platforms rarely expose them—you can start to see patterns that contradict the efficiency narrative. One pattern is floor price inflation in owned-and-operated inventory. When a platform owns both the supply and demand side, floors on that inventory tend to be higher than on comparable third-party inventory, even after controlling for viewability, domain authority, and audience composition. The platform captures the difference as margin.
Another pattern is non-transparent fee layering. In a fragmented system, each intermediary charges a disclosed or deducible fee. In a consolidated stack, the platform can bundle fees into a single “platform fee” or bury them in the media cost. Advertisers paying on a CPM basis may not realize that the effective CPM they see in their reporting already includes a supply-side fee, a data fee, and a measurement fee—all going to the same entity. Log-level analysis sometimes reveals that the net media cost reaching the publisher is 30–50% lower than the gross cost reported to the advertiser.
A third pattern is preferential allocation to owned demand. When a platform operates both a DSP and an SSP, it has the ability to route the highest-value impressions to its own demand sources before exposing them to external buyers. This isn’t always visible in standard reporting, but header bidding data and publisher-side logs have shown instances where a platform’s own demand wins impressions at lower CPMs than external demand would have paid—because the platform captures more total margin by keeping the transaction internal.
Why Advertisers Don’t Push Back Harder
If consolidation is so clearly tilted toward platform economics, why don’t more advertisers demand separation? The answer lies in a combination of convenience, measurement complexity, and organizational incentives.
Convenience is a powerful drug. Running campaigns across five different platforms requires five different UIs, five different reporting schemas, and five different optimization workflows. A consolidated platform offers a single login, unified reporting, and “AI-powered” optimization that promises to handle the complexity. For a marketing team that’s understaffed and over-measured on vanity metrics, that’s a compelling pitch. The fact that the platform is quietly extracting more margin is a second-order concern compared to hitting quarterly CPA targets.
Measurement is genuinely hard. Proving that a platform is overcharging requires counterfactual analysis: what would my results have been if I’d run this campaign through a different stack? Most advertisers don’t have the infrastructure to run clean incrementality tests across platforms. Those that do—typically large, sophisticated buyers—often find that their “optimized” platform campaigns perform no better than a properly managed multi-platform approach, and sometimes worse when accounting for total cost.
Organizational incentives are misaligned. At many companies, the team that negotiates platform contracts is not the same team that measures campaign performance. Procurement cares about rate cards and committed spend discounts; performance teams care about CPA and ROAS. A platform can offer aggressive rate-card discounts in exchange for spend commitments while quietly increasing the effective take rate through the mechanics described above. Both teams report success, and the platform captures the spread.
What a Healthier Market Would Look Like
None of this is an argument against integration per se. There are genuine technical benefits to reducing the number of hops in the supply chain, and some consolidation is a natural market response to the inefficiencies of the early programmatic era. The problem is the asymmetry: when one side of the market consolidates more than the other, the consolidated side captures a disproportionate share of the value.
A healthier market would have at least two characteristics. First, log-level transparency would be standard, not a premium feature negotiated by the largest buyers. Advertisers should be able to see the full chain of fees, the gross and net media costs, and the auction dynamics that determined their clearing price. This is technically feasible—the data exists in platform logs—but it’s withheld as a competitive moat.
Second, interoperability mandates would prevent platforms from locking advertisers into proprietary data ecosystems. If audience segments, optimization models, and performance data were portable across platforms, the switching costs that enable margin extraction would collapse. This is the logic behind initiatives like the IAB Tech Lab’s Data Transparency Standard, but adoption remains limited because the largest platforms have little incentive to participate.
FAQ: AdTech Consolidation and Advertiser Impact
Does consolidation actually reduce costs for advertisers?
Not in practice. While consolidation can reduce some technical overhead—fewer ad server calls, less latency, simplified reporting—the savings are typically captured by the platform, not passed through to advertisers. In many cases, the platform’s ability to control both supply and demand allows it to increase its total take rate without visibly degrading campaign performance metrics. Advertisers may see stable or even improved CPA/ROAS while the platform extracts more margin from the transaction.
How can advertisers detect if a consolidated platform is overcharging?
Detection requires access to log-level data, which most platforms do not provide by default. Advertisers with sufficient scale can negotiate for log-level access or work with independent verification partners to run supply-path optimization analyses. Practical signals include comparing net CPMs (what the publisher actually receives) against gross CPMs (what the advertiser pays) across different supply paths, and running controlled experiments that isolate the platform’s proprietary inventory from open-exchange inventory with similar characteristics.
Are independent adtech companies necessarily better for advertisers?
Not automatically. Independence removes the conflict of interest inherent in owning both sides of the transaction, but it doesn’t guarantee lower fees or better performance. Independent platforms still need to make a profit, and they face higher costs of capital and data acquisition than their consolidated competitors. The key advantage of independence is verifiability: when the buy-side and sell-side are separate, each party has an incentive to audit the other, and the advertiser can triangulate between them. The value isn’t in independence itself—it’s in the competitive tension that independence enables.
What should advertisers ask their platform partners about consolidation risks?
Advertisers should ask three specific questions. First, “Do you operate both a buy-side and sell-side business, and if so, how do you prevent preferential routing between them?” Second, “Can you provide log-level data showing every intermediary, every fee, and the net amount received by the publisher for each impression?” Third, “Are your optimization algorithms designed to maximize my ROI or your total platform revenue?” The answers—and the willingness to provide them—are often more revealing than the platform’s marketing materials.
Practical Steps for Advertisers
While structural change to the adtech market is slow, advertisers can take immediate steps to protect their interests within the current system:
Demand log-level data. Even if you can’t get full auction logs, push for as much transparency as your spend warrants. At minimum, request a breakdown of the media cost, data cost, platform fee, and any other line items. Compare the net media cost against publisher-reported revenue when possible.
Run supply-path optimization (SPO) exercises. Map out every path your spend takes to reach inventory. Identify redundant or high-cost paths. Consolidate spend toward the most direct, transparent paths—even if that means using multiple platforms instead of one consolidated stack.
Separate measurement from execution. Use an independent attribution and verification partner that doesn’t share a parent company with your DSP or SSP. This creates a check on self-reported performance metrics and makes it harder for a platform to grade its own homework.
Test incrementality, not just last-touch attribution. Consolidated platforms optimize for metrics that make their campaigns look good—often last-touch conversions. Run holdout tests to measure the true incremental value of platform-driven media. You may find that a significant portion of attributed conversions would have happened anyway.
The consolidation trend in adtech isn’t going to reverse. The economic forces driving it—network effects, data advantages, margin pressure—are too strong. But advertisers who understand the mechanics can make more informed decisions about where to concentrate their spend, what questions to ask, and how to structure their measurement to see past the platform’s preferred narrative. The goal isn’t to avoid consolidated platforms entirely; it’s to use them with eyes open, knowing exactly where the incentives point and what’s being traded away in exchange for that single-login convenience.