The Quality Open Web: Why Only a Fraction of Programmatic Inventory Makes the Cut
“The Open Web is supposed to offer almost unlimited inventory. So why does scaling become so difficult as soon as we apply a few basic quality safeguards?” A French Media Trader asked me that question. We decided to investigate.
1. The Open Web paradox: huge supply, limited usable inventory
The Open Web contains an extraordinary range of publishers, formats, devices and selling paths. On paper, programmatic buyers have almost endless choice. In practice, the picture changes when a campaign needs high viewability, credible publishers, clean supply paths and sufficiently prominent ad formats.
The challenge is not simply “finding impressions.” It is finding impressions that meet the advertiser's actual quality requirements at a workable scale and price. This is closely related to the Open Web's attention problem: an impression being available does not mean it will receive meaningful exposure.
2. Our French-market planning exercise
We used SSP/DSP planning tools to estimate the available pool of display bid requests associated with French IP addresses, excluding in-app inventory. The planner indicated approximately 53,000 million requests per week (53 billion). We normalized that starting pool to an index of 100 to make the effect of each filter easier to see.
We then applied six restrictions in sequence. Each reduction below refers to the volume remaining after the previous filter, not to the original total (the source percentages are rounded, and different planning tools or settings can produce different estimates).
3. Six filters: from 100 bid requests to approximately 0.24
| Successive filter | Reduction at this step | Index remaining (from 100) |
|---|---|---|
| Starting pool: French display bid requests, excluding in-app | — | 100.0000 |
| Brand safety and ad-fraud controls | −4% | 96.0000 |
| Minimum 75% MRC viewability criterion | −88% | 11.5200 |
| Top 600 websites whitelist | −19% | 9.3312 |
| Exclusion of suspicious MFA websites | −40% | 5.5987 |
| SPO: publisher SSP seats | −64% | 2.0155 |
| Screen-share proxy: mobile device + large ad format | −88% | 0.2419 |
In our original analysis, the planner's headline result was 0.23%. Recalculating from the six rounded percentages published above gives 0.2419%, or about 0.24%. The small difference is consistent with rounding or underlying unrounded planning figures; the original raw outputs would be needed to reconcile it exactly.
Put differently, for every 10,000 initial bid requests, the rounded waterfall leaves about 24 that satisfy all six conditions. That is an eligibility estimate—not a forecast of impressions won, people reached or campaign outcomes.
4. What does “only 0.23%” actually mean?
It does not mean that 99.77% of the French Open Web is fraudulent, worthless or unsuitable for every advertiser. It means that the overlap between these particular filters was extremely small in this particular planning exercise.
Some restrictions are deliberate strategic choices rather than objective measures of bad inventory. A top-600-site whitelist excludes many legitimate smaller publishers. A 75% viewability threshold can be demanding depending on how the planning tool estimates eligibility. And restricting buying to direct publisher SSP seats is a supply-path preference, not proof that every reseller impression is poor quality.
Similarly, mobile plus large-format inventory is a practical proxy for screen prominence, not a direct measurement of attention. For more on this distinction, see our analysis of why viewability alone is not enough.
5. And we haven't even applied audience targeting
Our six-filter exercise stops before three additional constraints that matter in real campaigns:
- Audience targeting: an interest, demographic or intent segment can narrow the pool further.
- Cookie and ID matching: audience addressability varies across browsers, consent states and identity solutions.
- Delivery optimization: bids, floors, pacing, frequency caps and performance rules affect which eligible opportunities actually turn into impressions.
That is why a seemingly enormous supply estimate can turn into a surprisingly difficult campaign to scale. Our article on reaching Safari inventory explores one part of the addressability challenge.
6. Why GreyFox Ads still champions the quality Open Web
Here in France, we believe advertisers deserve better than buying whatever happens to be cheap and abundant. What brand really wants impressions that never appear on screen, microscopic placements, pages overloaded with ads, unsafe editorial contexts or unnecessary intermediary fees?
That is the thinking behind our focus on a quality-first Open Web: stronger inventory selection, attention-aware formats, brand suitability, MFA exclusions and more transparent supply paths. We have also explored why made-for-advertising websites are a problem and the complexity introduced by reseller-heavy supply chains.
Quality does come with trade-offs. Tighter filters may improve inventory characteristics while reducing scale and increasing the clearing price. The right balance depends on the campaign's objective, not on a single universal whitelist.
7. A practical checklist for Media Traders
- Build a waterfall before launch. Estimate the marginal volume removed by each targeting and quality rule.
- Check overlaps, not just individual exclusions. Six reasonable filters can interact to leave an unexpectedly small eligible pool.
- Separate safety rules from buying preferences. Fraud prevention, publisher whitelists and supply-path optimization serve different purposes.
- Measure quality against outcomes. Compare viewability, attention proxies, completion, qualified traffic and effective cost where appropriate.
- Test controlled relaxation. If delivery stalls, identify which constraint limits scale and assess whether loosening it harms the campaign's real KPI.
- Keep your supply chain understandable. Prefer transparent paths and investigate unexplained intermediaries rather than assuming every indirect path is automatically invalid.
8. The takeaway: quantity is not the same as quality
The French Open Web may generate tens of billions of bid requests, yet the intersection of strict brand-safety, viewability, publisher, MFA, SPO and format criteria can be tiny. In our planning scenario, the reported figure was 0.23% of the initial pool.
For Media Traders, the lesson is not to abandon quality safeguards. It is to understand the cost of each filter, measure the trade-offs and build a buying strategy that can actually deliver.
Because the best Open Web campaign is not the one that buys the most bid requests. It is the one that buys the right impressions.
Activate the quality French Open Web with GreyFox Ads
Buying France from a UK, US or EMEA trading desk? Our Master Curated Deal brings French-market expertise, quality controls and simplified activation together in one deal.
Discover the Master DealContact GreyFox AdsOriginal analysis: Read the original LinkedIn post. Planning volumes and filtering estimates are drawn from that post; the underlying planner export was not supplied for independent verification.