Definition
Supply Path Optimization (SPO) is the practice of improving how programmatic ad spend reaches publisher inventory by cutting unnecessary intermediaries and favoring supply routes that are more direct, more transparent, and better performing. A “supply path” is the specific chain of platforms and resellers that carries an impression opportunity from a publisher to a buyer — publisher → SSP → exchange → DSP, with resellers sometimes inserted at any point along the way.
See also: Programmatic Advertising · Real-Time Bidding (RTB) · Private Marketplace (PMP) · Ad Exchange
How SPO relates to marketing
For marketers, SPO is a budget question wearing a technical costume. Every hop between publisher and buyer takes a fee, and the ANA programmatic transparency work has been the industry’s most-cited evidence that the losses are large — analyses of that research put roughly 70 cents of each dollar reaching the publisher on unoptimized paths. The same study family found that a meaningful share of spend was still landing on made-for-advertising sites even where buyers believed they’d blocked it.
So SPO does two things at once. It raises the share of budget that becomes working media, and it tightens the quality controls that sit around brand suitability, fraud, and measurement. Those aren’t separate projects. A path with four resellers is usually both the expensive path and the one where verification coverage gets patchy.
The practice sits with the advertiser or agency, executed through DSP settings and deal structures, though SSPs and curators now build much of the tooling.
How to calculate SPO
SPO isn’t a single metric. It’s evaluated with a set of path-level calculations across cost, auction health, quality, and outcomes.
Supply efficiency and consolidation
- Working media rate = Media cost paid to publishers ÷ Total advertiser spend (Requires fee disclosure. Most buyers approximate it.)
- Intermediary count per path = Number of hops between publisher and DSP, read from SupplyChain object nodes
- Directness rate = Impressions bought via direct publisher or SSP relationships ÷ Total impressions
- Path duplication rate = Duplicate bid requests for the same impression ÷ Total bid requests
Auction and delivery health
- Win rate by path = Impressions won ÷ Bids submitted
- Bid rate by path = Bids submitted ÷ Bid requests received
- CPM by path = (Spend on path ÷ Impressions on path) × 1,000
- Timeout loss rate = Opportunities lost to latency ÷ Total eligible opportunities
Quality and compliance
- Viewability rate by path = Viewable impressions ÷ Measurable impressions
- IVT rate by path = Invalid impressions ÷ Measured impressions
- Suitability violation rate = Violations ÷ Total impressions, scored against your own suitability rules
Outcomes
- CPA and ROAS by path, compared across prioritized and deprioritized routes
- Incremental ROAS where a holdout design is available
The last one is the one most teams skip and most need. A path can post a great CPM and still be buying impressions the brand would have won anyway.
How to utilize SPO
Common use cases
- Collapse duplicate inventory. The same impression frequently reaches a DSP through five or six wrappers. Buying it once, through the cleanest route, is the core move.
- Prioritize authorized, direct supply. Favor publishers’ direct SSP relationships and sellers verified through ads.txt and sellers.json.
- Improve auction efficiency. Concentrate on paths with better win rates, lower latency, and pricing that doesn’t swing wildly week to week.
- Extend verification coverage. Prefer supply that carries consistent app and site identifiers so verification vendors can actually measure it.
- Reduce carbon load. Fewer redundant bid requests means less infrastructure burn, which has moved from a nice-to-have to a reporting line at several large agencies.
A workable sequence
- Pull path-level reporting from DSP logs, SSP transparency reports, and verification data.
- Rank paths by spend and impression volume. Start where the money is, not where the analysis is interesting.
- Score each path on directness, quality, disclosed fees, and outcome performance.
- Consolidate. Prune resellers, cut exchanges that add nothing but a fee, and route the volume to the survivors.
- Implement controls — allowlists, deal IDs, app bundle governance, floor strategy.
- Re-test on a set cadence. Supply relationships change constantly, and a rule set from last quarter is already a little wrong.
Step 6 is where most SPO programs quietly die.
Comparison to similar approaches
| Topic | Supply Path Optimization (SPO) | Demand Path Optimization (DPO) | Curation | Allowlisting / Blocklisting | Header Bidding |
|---|---|---|---|---|---|
| Primary focus | How spend traverses the chain to reach inventory | How publishers route and prioritize demand sources | Pre-packaged supply plus data, sold as deals | Which sites and apps are eligible | How publishers expose inventory to auctions |
| Main goal | Cut waste, raise transparency and quality | Raise yield and reduce demand-side redundancy | Deliver performance-ready inventory with less buyer effort | Control suitability and safety | Increase publisher yield |
| Primary owner | Advertiser or agency, plus DSP | Publisher, plus SSP | Curator, often an SSP or data company | Advertiser or agency | Publisher |
| Typical levers | Reseller pruning, exchange selection, direct paths | Demand partner selection, auction rules | Deal IDs, audience enrichment, inventory filtering | Lists, category exclusions | Wrapper config, timeout settings |
| Best fit | Programmatic programs with real scale | Publishers with many demand partners | Buyers wanting SPO outcomes without the audit work | Brand safety governance | Publisher monetization strategy |
SPO and curation get conflated a lot, and it’s worth separating them. SPO is the buyer’s own audit and consolidation work. Curation is a productized version of the same outcome, sold by a third party — which is convenient, and also means you’re trusting someone else’s definition of quality.
Best practices
Start with the money. Rank by spend before anything else. A path carrying 0.3% of budget doesn’t deserve a two-week investigation.
Use ads.txt, app-ads.txt, and sellers.json as baseline hygiene. These are cheap to check and they catch spoofing and unauthorized resale. The SupplyChain object gives you the full node list per bid request, which is what makes hop-counting possible at all.
Measure overlap before you cut. Identify inventory reachable through multiple routes, then consolidate deliberately. Cutting an exchange without knowing what unique supply it carried is how buyers accidentally lose reach.
Treat latency as a cost line. Timeout losses erode win rates quietly, and the damage rarely shows up in a standard report.
Segment by channel. CTV supply chains look nothing like display supply chains — different reseller structures, different verification constraints, different pricing behavior. One global rule set will be wrong somewhere.
Connect SPO to suitability, not just fraud. Supply quality includes content adjacency. A clean path to a bad context is still a bad buy.
Keep a governance registry. Approved sources, deal IDs, ownership, review dates. Boring, and it’s the difference between a program and a series of one-off cleanups.
Future trends
Curation absorbs SPO. By 2026 the practical center of gravity has moved toward SSP-led curated paths, where the filtering happens upstream and buyers activate through pre-approved deal IDs. Equativ’s guidance to buyers frames shorter paths as adding 15–20% to working media, which is the number doing most of the persuading in agency meetings.
Agentic curation. The newer development is supply-side systems that select paths in real time rather than applying static rules written last quarter. That reframes the buyer’s job — less list maintenance, more defining what good optimization intent actually means. It also creates a governance problem nobody has fully solved, since an automated path selector optimizing for cost can quietly drift away from suitability standards. Related work is underway through the Agentic Advertising Standards effort and the Ad Context Protocol (AdCP).
Cross-platform pressure. The IAB’s 2026 Outlook Study found cross-platform measurement rising to a top priority for 72% of advertisers, up from 64% the prior year. Optimizing supply inside one channel while retail media and CTV report separately produces a tidy result and an incomplete picture.
Sustainability as a path criterion. Duplicate bid request processing is now a measurable emissions cost, and deduplication via GPID and schain analysis has become part of the standard SPO toolkit rather than a separate ESG exercise.
Outcome linkage. Expect continued movement from CPM and viewability toward incrementality as the deciding test for whether a path earns its place.
FAQs
What’s the difference between SPO and curation? SPO is the buyer-side discipline of auditing and consolidating supply paths. Curation is a packaged product — usually from an SSP or data company — that delivers similar outcomes through pre-filtered deal IDs. Many buyers now do both, using curated deals for the bulk of spend and running their own path analysis to check the curator’s work.
How many supply paths should an advertiser keep? There’s no universal number, and anyone quoting one is guessing. What matters is that each remaining path serves a function the others can’t — unique inventory access, a data enrichment layer, or a fraud control — rather than existing because nobody removed it.
Does SPO reduce reach? It can, if you cut without measuring overlap first. Done properly, most consolidation removes duplicate access to inventory you were already reaching, so reach holds while cost drops. The failure mode is pruning an exchange that carried genuinely unique supply.
What is the SupplyChain object? An OpenRTB specification that lists every entity involved in selling a given bid request, as a set of nodes. IAB Tech Lab’s OpenRTB Working Group finalized it alongside sellers.json in July 2019. It’s what lets a buyer count hops rather than guess at them.
How is SPO different from ads.txt? ads.txt is a publisher-published file declaring who’s authorized to sell their inventory. It’s an input to SPO, not SPO itself. SPO uses that authorization data along with performance, fee, and quality signals to decide where spend should actually flow.
Can smaller advertisers do SPO? Yes, though the economics differ. Below a certain spend level, building a path-level analysis practice costs more than it recovers. Smaller buyers usually get most of the benefit by working through curated marketplaces and PMPs rather than running their own audits.
Does SPO apply to CTV? Very much so, and CTV is often where the waste is worst. Streaming supply chains carry more inventory-sharing arrangements and more resellers than display, and verification coverage is thinner. IAB Tech Lab has published a CTV-specific extension of app-ads.txt for exactly this reason.
How often should supply paths be reviewed? Quarterly is a common cadence for full reviews, with monitoring in between. Anything longer and the analysis is describing a supply chain that no longer exists.
Does SPO help with walled gardens? Not really. SPO is an open-web and open-programmatic discipline. Walled garden environments don’t expose the supply chain in a way that permits path analysis, which is part of why cross-platform comparison stays difficult.
What’s the single highest-value first step? Pull the SupplyChain object data for your top 20 paths by spend and count the nodes. Most buyers find at least one route they didn’t know they were paying for.
Related Terms
- Programmatic Advertising
- Demand-Side Platform (DSP)
- Ad Exchange
- Real-Time Bidding (RTB)
- Private Marketplace (PMP)
- Connected TV (CTV)
- Retail Media Network (RMN)
- Walled Garden
- Agentic Advertising Standards (IAB Tech Lab)
- Ad Context Protocol (AdCP)
Sources
- IAB Tech Lab, sellers.json Supply Chain Transparency — https://iabtechlab.com/sellers-json/
- IAB Tech Lab, sellers.json Specification (PDF) — https://iabtechlab.com/wp-content/uploads/2019/07/Sellers.json_Final.pdf
- IAB Tech Lab, ads.txt 1.1 Implementation Guide (PDF) — https://iabtechlab.com/wp-content/uploads/2022/04/Ads.txt-1.1-Implementation-Guide.pdf
- IAB Tech Lab FAQ, Supply Chain Transparency — https://interactiveadvertisingbureau.github.io/TechLabFAQ/docs/faq-supply-chain-transparency.html
- IAB, Sellers.json Guidelines — https://www.iab.com/guidelines/sellers-json/
- Equativ, A 2026 Guide to Programmatic Curation & Curated Marketplaces — https://www.equativ.com/blog/2026-programmatic-curation-guide
- AudienceX, Fixing Programmatic Waste: A 2026 Supply Path Optimization Blueprint — https://insights.audiencex.com/fixing-programmatic-waste-a-2026-supply-path-optimization-blueprint/
- AI Digital, What’s Supply Path Optimization (SPO) in Programmatic Advertising — https://www.aidigital.com/blog/what-is-supply-path-optimization
- AI Digital, Sustainable Programmatic Supply Paths: SPO, Efficiency & Carbon Impact — https://www.aidigital.com/blog/sustainable-programmatic-supply-path
- Experian, Why Programmatic Curation Is Becoming the Standard — https://www.experian.com/blogs/marketing-forward/why-programmatic-curation-is-becoming-the-standard-2026/
- Index Exchange, Understanding Supply Chain Transparency: Ads.txt, Sellers.json, and SupplyChain — https://www.indexexchange.com/en-au/index-explains/understanding-supply-chain-transparency-aus/
- PubMatic, Unlocking New Layers of Control Through Supply Path Curation — https://pubmatic.com/blog/unlocking-new-layers-of-control-through-supply-path-curation/
- Admozart, Agentic Curation for Smarter Programmatic Supply Paths — https://admozart.com/agentic-curation-for-smarter-programmatic-supply-paths/
