Show-Level CTV Optimization in 2026: What Advertisers Can Actually Buy, Report On, and Optimize

A version of this article was originally published in AdExchanger

Show-level reporting in CTV is becoming real. Show-level optimization (continuously buying, targeting, and steering budget against individual programs in real time) is not, at the moment. 

Three things are true at the same time, and conflating them is where buyers get burned:

If you only read one line, it’s this: reporting where an ad ran is not the same as being able to transact against it.

What Publishers Told Tatari About Show-Level Optimization

I run programmatic at Tatari, which means I spend most of my week very close to the thing everyone else is writing about. We buy across the largest premium streaming publishers every day. Roughly 80% of the media on our platform is transacted direct, and the remaining 20% is programmatic where it genuinely earns its place. So when a reporting limitation shows up, I don't read about it, I see it first-hand.

That's the only reason I have a strong opinion here. I've spent fifteen years on the optimization side of this business at Google and YouTube, at Yahoo, at Matterkind, and the pattern is always the same. A new signal arrives, the word "optimization" gets attached to it, and it takes the market about eighteen months to figure out that a signal you can see and a signal you can act on are different products.

So when the show-level optimization claims started circulating this summer, I did the boring thing. Over about a week, I called our day-to-day and technical contacts at several of the biggest publishers and exchanges in CTV and asked them, plainly, what they could actually deliver. 

What I heard is below. If you're a performance buyer, an investment lead, or a brand-side media owner about to sit through a show-level pitch, I'd rather you walk in with my notes than take the headline at face value. I'll cover what the underlying data actually looks like, what the largest premium streamers can deliver right now, why linear solved this problem decades ago almost by accident, and the five questions I'd ask any partner making the claim.

Why Do CTV Buyers Want Show-Level Data?

Because most of them don't trust what they're buying.

• 43% of CTV buyers say they have little or no confidence in the inventory they're purchasing, per a recent IAB report.

• 91% of programmatic traders say they often struggle to explain CTV performance because of limited content-level data.

• 86% of US media planners say they would move more linear dollars into CTV if they had show-level targeting.

• 53% say show-level data would improve campaign optimization.

The demand signal is unambiguous. That's exactly why the supply-side claims deserve scrutiny.

What Does Show-Level Data Actually Look Like?

The first problem is data quality at the source.

Raw bidstream data does carry content-level signals, but those signals are incomplete and inconsistent. The clearest tell: a significant share of content titles arrive as "unknown". A dataset with heavy "unknown" density isn't a transparency breakthrough, it's a partial picture with a confidence problem. Properly licensed, metadata-matched reporting including Gracenote-based data being the common example, remains the more reliable standard for understanding where ads actually appeared.

But better reporting does not automatically become an optimization capability. The major exchanges we consulted do not currently offer scaled show-level optimization as a standard product. In most cases, that functionality is still in early design and development.

Bidstream Data vs. Licensed Metadata: What's the Difference?

What Can Premium Publishers Actually Deliver Today?

The gap gets clearer when you talk to publishers directly. Three representative conversations:

1. A top-five streamer told Tatari its current SSP integration has limitations that prevent pulling show-level reporting at all.

2. A second premium publisher confirmed it does not broadly share show-level data.

3. A third was candid that while show-level reporting is available — typically as top 25/50/100 rankings, without impression counts — the ability to optimize against it is directional at best.
   
   

What You Can and Can't Do With Show-Level CTV Data in 2026

You generally can:

• Receive post-campaign, aggregate reporting on which programs your ads appeared in

• Validate brand suitability and content context after the fact

• Use historical program performance to inform the next flight's planning

• Buy at the genre or audience level

You generally cannot:

• Target individual shows programmatically without a programmatic-guaranteed (PG) deal or sponsorship

• Optimize against individual shows inside a PMP — the workhorse of performance CTV buying

• Get impression counts alongside most show-level rankings

• Apply real-time, in-auction bid decisioning based on the specific title
  
  

Why Does Linear TV Already Offer Program-Level Transparency?

The contrast between streaming and linear is instructive — and it isn't a knock on streaming. It's structural.

Linear TV has offered full program-level transparency for decades. Linear ads are bought and measured against pre- and post-roll program logs: records confirming exactly what aired, plus an aggregate count of who watched. Because that log-based reporting isn't tied to identity and can't be connected to specific households, there's no privacy trade-off in sharing it. Advertisers know precisely which programs their spots ran against.

Streaming inverted that model. CTV knows far more about who is watching — often down to the household and device. But that same richness is what brings the Video Privacy Protection Act (VPPA) into play. When show-level viewing data is combined with household IDs or IP addresses in an ad transaction, VPPA exposure is real. That's precisely why premium streamers have leaned into contextual, content-ID-based approaches.

One compliant approach is to match anonymized content IDs to program metadata without exposing an individual household's viewing habits. That delivers valuable program-level insight while maintaining privacy protections, but the resulting signal is naturally better suited to aggregate reporting than to ID-level optimization.

For advertisers who want reliable, program-level insight today, linear remains the most transparent option available. Not because streaming is behind, but because the privacy architecture is genuinely different.

Reporting ≠ Optimization: Why the Distinction Matters

Post-campaign, aggregate show-level reporting is becoming more standard, and that's genuinely useful. It lets you validate brand suitability, understand content context, and justify streaming spend internally.

But reporting where an ad ran is fundamentally different from dynamically optimizing a campaign at the individual show level.

Matching content IDs to metadata offline or mid-flight and then steering budget toward historically stronger programs can improve decision-making. It should not be confused with the ability to continuously buy, target, and optimize against individual shows programmatically in real time. Those are different products with different infrastructure requirements.

Five Questions to Ask Any Partner Pitching Show-Level Optimization

1. How complete is the data? What share of impressions return a known title versus "unknown"? Is it bidstream-derived or licensed metadata?

2. How fast is it available? Post-campaign only, daily, or in-flight?

3. Can I actually transact against it? Or is it a reporting layer with no buying control attached?

4. Through which deal type? PG and sponsorship are a different commitment profile than PMP. If PMP isn't supported, the capability doesn't fit most performance buying.

5. At what scale? How many publishers, what percentage of your CTV footprint, and with what impression volume behind each title?
   
   

Where the TV Industry Goes From Here

Scaled, compliant show-level optimization is a goal worth pursuing, and several ad tech companies across the ecosystem are actively building toward it. FreeWheel's Video Content Report and IAS Total TV are both real steps toward closing the visibility gap, and Tatari is engaged in those conversations. We'd be first in line for a version that works at scale.

But the market still has a meaningful gap to close between greater program-level visibility and true program-level optimization. As infrastructure, data quality, and privacy frameworks mature, that gap may well close. Until then, show-level reporting can provide valuable signals for campaign planning and performance analysis. Buyers should simply understand exactly where reporting ends and optimization begins.

Want to learn more about understanding what publishers can provide? Let’s talk!

Frequently Asked Questions (FAQs)

What is show-level optimization in CTV?

Show-level optimization is the ability to buy, target, and adjust CTV media against individual TV programs in real time — steering impressions and budget toward specific titles while a campaign is live. It is distinct from show-level reporting, which tells you after the fact which programs your ads appeared in. As of 2026, the reporting side is maturing; the optimization side is not available as a scaled, standard product from major exchanges.

Can I target specific TV shows in a CTV campaign?

In most cases, no — not programmatically at scale. Targeting individual shows generally requires a programmatic-guaranteed (PG) deal or a direct sponsorship with the publisher. PMP-based optimization, which is the workhorse of performance CTV buying, does not support show-level targeting today. Genre- and audience-based deals remain the practical ceiling for most buyers.

Why can't streaming platforms share show-level data the way linear TV does?

Privacy architecture. Linear program logs confirm what aired and how many people watched in aggregate, with no tie to identity — so there's no privacy trade-off in sharing them. CTV data is tied to households and devices, which brings the Video Privacy Protection Act (VPPA) into play when show-level viewing data is combined with household IDs or IP addresses in an ad transaction. That legal exposure is why premium streamers have moved toward contextual, anonymized content-ID approaches instead.

Is bidstream show-level data reliable?

Not on its own. Raw bidstream content signals are frequently incomplete and inconsistent, with a telltale prevalence of "unknown" content titles. Licensed, metadata-matched reporting — such as Gracenote-based data — is a more reliable standard for understanding where ads actually appeared. Ask any vendor whether their dataset is bidstream-derived and what their known-title match rate is.

What is the difference between show-level reporting and show-level optimization?

Reporting is retrospective and descriptive: an aggregate list, often a top 25/50/100 ranking without impression counts, showing which programs carried your ads. Optimization is prospective and transactable: the ability to bid, target, and reallocate against specific titles in real time. Reporting is increasingly available. Optimization, at scale and compliantly, is not.

Does show-level reporting include impression counts?

Usually not. Most publishers that share show-level reporting deliver it as ranked lists — top 25, 50, or 100 programs — without impression volume attached to each title. That limits how precisely you can weight performance by program.

What is the VPPA and why does it affect CTV targeting?

The Video Privacy Protection Act is a US law originally written to protect video rental records. In CTV, it becomes relevant when show-level viewing data is combined with household identifiers or IP addresses in an ad transaction, creating real legal exposure for publishers. It is the primary reason premium streamers limit how broadly they share show-level data and why compliant approaches rely on anonymized content IDs matched to metadata.

If show-level optimization isn't available, what should performance CTV buyers do instead?

Focus on the levers that actually work today: direct publisher relationships that provide transparency without the bidstream data-quality problem, genre- and audience-level deals, outcome-based measurement tied to business results rather than content proxies, and — where program-level certainty genuinely matters — linear TV, which has offered full program-level transparency for decades.