If your CFO asked you to find 20% in your streaming TV budget tomorrow, you wouldn't have to cut a single campaign or creative. You'd just have to stop paying to reach the same households over and over.
P&G’s Marc Pritchard has called excess frequency "the biggest waste" in his media plan, and cross-platform spending “flying with a sheet over the windshield.” Sir Martin Sorrell has flagged frequency overexposure as a “front and center concern,” one of advertising’s biggest unsolved efficiency drains.
But the math is worse than most marketers realize. Overexposure is rampant – the highest-investment campaigns deliver average frequencies of 15 or more per household. A recent peer-reviewed study estimated a median exposure waste of 38% on YouTube and 27% on DV360 Display & Video. EDO's internal data shows waste can run as high as 60% on individual campaigns.
Your streaming TV budget has a hidden leak. The good news: it's fixable, the math is already solved, and the path from insight to action is now as easy as hitting a button.
An average campaign frequency of 6 might sound healthy. But averages mislead – the real picture only emerges when you break them down by publisher, segment, and DMA. Sixty percent of your audience saw the ad once, 25% saw it three times, and 15% saw it 30 times. The average looks fine. The distribution is a disaster.
Nutritionists call this “TOFI”: thin outside, fat inside. Your campaign looks lean on paper. Underneath, a long tail of households is quietly eating most of your media budget.
The ad industry has been asking the wrong question for decades — "What is the optimal frequency?" — as if there's a single magic number for every publisher and audience. This is a ghost of Herbert Krugman's 1970s "Three-Hit Theory”: the idea that a consumer needs to see an ad three times before it burns into memory.
The right question today is: what are the optimal frequency caps? Plural. Different for every publisher and every market. This isn't a consumer-psychology problem about retention thresholds. It's an advertiser-side budget optimization problem — answerable with data and math, not gut feel.
Reframed that way, we can find these precise caps with three inputs:
The economics — The dollar value of the outcomes you're driving, and what each publisher charges you per thousand impressions.
The response curves — How ad engagement actually changes with each additional exposure, for that specific publisher (see graph below).
The exposure distributions — What your real frequency tail looks like underneath the average, publisher by publisher (see bar chart above).
Most marketers are missing all three. EDO has them — we connect Convergent TV airings to the ad-driven consumer behaviors most predictive of future sales, publisher by publisher. With those three inputs, computing the right cap is just math – but math at a scale no human can do by hand.
A streaming TV buy can span 20–40 publishers across 100+ DMAs, creating millions of permutations. And although two publishers might charge the same CPM, their response curves almost certainly diverge for your brand. One may saturate fast with engagement dying off after a few exposures, while another flattens gradually as ads keep driving results.
Frequency caps must follow the curve, not the rate card. And eyeballing a curve isn’t enough – two curves can look identical at a glance and produce very different break-even caps once the math is applied. This is exactly the work that Vertical AI does better than people: millions of small decisions, in near real time, adding up to millions in optimized spend you can redirect towards reach (or savings for your CFO).
Even armed with the right math, media planners must put these recommendations into action. Often, this means manually configuring frequency caps across dozens of ad groups in multiple DSPs. Then doing it again next month, while setting up a control group to prove it all worked. The friction between insight and execution can be debilitating.
EDO has been laser-focused on eliminating this friction. Last year, we predicted that 2026 would be the year of the “Easy Button” in TV advertising — automated, outcomes-driven buying that finally makes streaming TV work the way search and social already do. Frequency optimization is one of the easiest, highest-impact Easy Buttons there is.
Here's what it looks like in practice: Our algorithm computes the break-even cap for every publisher and DMA. We hand you a file you upload directly to your DSP, with the right cap pre-set for every ad group — and a control group built in alongside it, so the savings are provable, not theoretical.
Then we keep going. CPMs shift and audiences change, so we rerun the model on whatever cadence the campaign needs — weekly, sometimes daily — with a human reviewing each refresh. As we build direct API connections to major DSPs, that loop will tighten further: from file-based handoffs to closed-loop, agent-to-agent updates.
No spreadsheets. No translation layer. No quarterly project that’s stale by the time it ships. The decision is the deliverable. In one analysis, recoverable spend reached as high as 24% across major streaming publishers.
For too long, frequency waste has been the silent killer of streaming TV efficiency. Now, automated frequency optimization marks a key moment where that promise starts coming true.
Previously, no one in the value chain had the data, the math, and the incentive to put these controls in place. But with the rise of Vertical AI, the Easy Button for frequency capping is now at your fingertips – the only thing left is to press it.
Want to see what your TV distribution actually looks like? Start your frequency audit.