For most of its existence, supply path optimization was understood as a cost exercise. Find the cheapest route to the same impression. Cut the intermediaries taking a margin. Get more of the dollar to the publisher.
That framing is now demonstrably wrong, and there is data to prove it.
The ANA’s Q1 2026 Programmatic Transparency Benchmark compared higher-performing and lower-performing advertisers across log-level impression data. The difference in transaction costs between the two groups was 2.4 percentage points. The difference in media productivity losses was 19.4 percentage points.
Fees were almost a rounding error. What separated the winners was whether the impression at the end of the path was worth buying.
SPO in 2026 is not about the toll. It is about the road.
What SPO actually is
Any given impression can usually be bought through several routes. The same app inventory might be available through four SSPs, two of which resell from the other two, with a reseller layered on top of one of those.
Same impression. Different paths. Different prices, different transparency, and, critically, different probability that the impression is real, viewable, and measurable.
SPO is the discipline of choosing deliberately.
For advertisers, it means concentrating spend on paths that are direct, authorised, and measurable, then holding them accountable on quality rather than CPM.
For publishers, it is the mirror image: making your inventory the path buyers choose. That means being easy to authorise, easy to verify, and easy to trace.
The three files that make SPO possible
SPO is not a philosophy. It runs on three IAB Tech Lab standards, and most of the industry’s transparency problems are just these three being ignored.
ads.txt / app-ads.txt is the publisher’s public declaration of who is authorised to sell their inventory. If a seller is not in the file, they are not authorised, and any bid you win through them is inventory nobody vouched for.
sellers.json is the other side: each SSP publishes who its sellers are and whether each is the direct owner of the inventory or a reseller. It turns an anonymous seller ID into an identifiable company.
The SupplyChain object (schain) travels inside the bid request and lists every node the request passed through, in order. It is the receipt.
Together they answer: is this seller allowed to sell this, who are they, and how many hands did this touch on the way to me?
The standards have existed for years. The gap between advertisers is not access to them. It is enforcement.
What the 2026 data changes
The ANA found that higher-performing advertisers operate with significantly more concentrated supply footprints. Fewer paths, deliberately chosen.
The rest of their profile follows from it. A 13.3-point advantage in measurable inventory. A 6.7-point advantage in viewability. Lower average CPMs, not higher.
That last one deserves emphasis, because it breaks the assumption underneath most SPO resistance. Quality did not cost more. The advertisers running concentrated, direct, measurable supply paid less per thousand impressions than the ones spraying across everything available.
And once you adjust for waste, the gap becomes absurd. TrueCPM, the ANA’s quality-adjusted measure, came in at $7.46 for the higher-performing cohort and $19.04 for the lower-performing one. A headline CPM difference of $1.95 became a real difference of $11.58.
The cheap path was two and a half times more expensive.
Meanwhile, the market-level TrueAdSpend Index, the share of programmatic spend producing a fraud-free, measurable, viewable, MFA-free impression, sits at 43.3 percent. Most of a programmatic dollar still does not arrive.
What this means for advertisers
Stop optimising to CPM. It is the metric that makes long, opaque supply paths look good. Ask every partner for a quality-adjusted cost and compare on that.
Concentrate. Every additional path is another integration to monitor, another set of reporting to reconcile, another surface for MFA and fraud. Breadth of supply is not a strength. It is unmanaged risk that shows up as media productivity loss.
Enforce schain and sellers.json pre-bid. Not in the quarterly audit. If the path cannot be resolved to authorised, direct sellers, do not buy it.
Prefer direct. A path with one hop to a publisher who owns the inventory is structurally safer than a path with four, regardless of what any vendor is layered on top.
Watch MFA separately. MFA exposure rose from the 0.4 to 0.6 percent range across 2025 to 1.1 percent in Q1 2026, and the ANA named AI slop as an emerging subtype. Generative tooling has made low-quality inventory dramatically cheaper to produce. This number is not going down on its own.
What this means for publishers
The instinct when SPO gained traction was defensive: fewer paths means fewer buyers means less demand. That has not been how it played out.
Buyers are concentrating. The question for a publisher is whether you are in the set they concentrate into.
Audit your app-ads.txt. Stale entries authorising partners you left two years ago are not harmless. They are unmanaged resale paths carrying your name.
Reduce your own resale layers. If your inventory reaches a buyer through three intermediaries, you are the one paying for two of them in the price they are willing to bid.
Make yourself verifiable. Clean sellers.json declarations, accurate direct/reseller flags, complete schain. Buyers are filtering on this pre-bid. Ambiguity is not neutral, it is an exclusion.
Sell quality, not just reach. Viewability, session depth, retention, real users. That is what the top cohort is paying a premium for and what everyone else is failing to find.
The structural point
SPO is converging on something simple. The industry spent fifteen years adding intermediaries, each promising optimisation, and is now measuring what that actually produced: 43.3 percent of spend arriving as a qualified impression.
The correction is not that intermediaries are evil. Exchanges, DSPs and SSPs do real work. It is that intermediaries who only add distance are being priced out, and the data has finally caught up to make that visible.
This is the reason AdSpin is built the way it is. Our exchange runs on owned-and-operated supply plus direct publisher integrations, on our own oRTB infrastructure. Not because vertical integration is fashionable, but because it is the only structure where full supply-path visibility is a property of the system rather than a promise in a deck. When you buy Adspin AdX, you are connecting to the source of the supply. There is no chain to audit because there is barely a chain.
For advertisers, the shortest verifiable path is the cheapest one, once you count what you actually received. For publishers, being that path is the position worth holding.
SPO stopped being an optimisation tactic. It became the definition of who gets the budget.
See how AdSpin’s owned supply and proprietary exchange give buyers a verifiable, direct path to mobile inventory at adspin.io.
