The story. Show-level CTV inventory control is blocked less by missing data than by two structural walls: the Video Privacy Protection Act, and a deal structure that doesn’t support show-level programmatic buying at all. A new IAB report found 43% of CTV buyers have little or no confidence in the inventory they’re buying. Tatari’s Tom Discepola argues the fix isn’t more reporting — it’s the plumbing underneath it. (AdExchanger, “CTV’s Transparency Gap: Why Better Show-Level Data Still Can’t Buy You Show-Level Control,” 18 September 2026)
01What happened
- Confidence in CTV inventory is thin. A new IAB report found 43% of CTV buyers have little or no confidence in the inventory they’re actually buying.
- Linear TV never had this problem, for a structural reason. Program logs confirm exactly what aired, with no household-level identity attached — so there’s no privacy trade-off in sharing them.
- CTV’s richer data is exactly what blocks it. CTV knows viewers down to the household and device, and combining show-level viewing with household IDs or IP addresses in a transaction creates real VPPA exposure — why premium streamers lean on contextual, content-ID-based approaches instead.
- Publishers themselves confirm the gap. One top-five streamer told Tatari its SSP integration can’t pull show-level reporting at all. A second doesn’t broadly share show-level data. A third offers only top 25/50/100 rankings without impression counts, calling optimization against them “directional at best.”
- The deal structure is the other wall. Buyers can’t target individual shows programmatically without a programmatic-guaranteed deal or sponsorship — PMP-based optimization, the workhorse of performance CTV buying, doesn’t support it, leaving genre- or audience-based deals as the realistic ceiling.
- Gracenote-matched data is the more trustworthy standard. Raw bidstream content-level signals are often incomplete and inconsistent; properly licensed, metadata-matched reporting holds up better.
02What it means inside a GAM network
The IAB’s 43% confidence figure reads, at first, like a reporting problem — buyers want a better dashboard. It isn’t one. A publisher can hand a buyer a flawless show-level report and still have nothing to sell against it, because PMP-based optimization, the mechanism that actually executes most performance CTV buying, has no show-level lever to pull. That’s the distinction worth holding onto inside a GAM/MCM CTV stack: reporting transparency and buying control sit in different layers, and fixing one doesn’t touch the other. VPPA is the reason the gap won’t close through better tooling alone — a publisher who wires household IDs into show-level bid data to satisfy a buyer’s ask is taking on real privacy exposure, not just an engineering task. The practical middle ground several vendors are reportedly building toward is exactly what genre- and audience-tier PMPs already approximate: aggregation above the individual show, specific enough to buy against, generic enough to stay clear of VPPA.
03What publishers should do about it
04Offer tie-in
For publishers selling CTV inventory into agency and brand demand, this is precisely the gap APH’s CTV monetization work is built to close: packaging genre- and audience-tier deals that give buyers real control today, without exposing a publisher to the VPPA risk a show-level identity match would create.
05The bottom line
Better show-level reporting was never going to fix CTV buyer confidence on its own, because the confidence gap sits downstream of two things reporting can’t touch: a privacy law that makes household-level show data risky to assemble, and a deal structure built for genre and audience tiers, not individual shows. The 43% figure is a symptom. The fix, per Discepola’s argument, runs through the plumbing — PMP structure and identity-safe metadata matching — not the dashboard.