How smarter site filtering and frequency capping prevent wasted spend

When budgets are flat and pressure on performance keeps rising, simply chasing more impressions is not a strategy.

You're likely expected to lean into upper‑funnel channels like display, connected TV (CTV), and streaming TV (STV) and, at the same time, protect lower‑funnel results. That means you have to do more with the media budget you already have.

One of the smartest ways to do that is to reduce waste so that what you buy, where you run, and how often you show up all become more efficient.

You cannot invest more in where you are seeing results if a meaningful slice of your budget is tied up in places and impressions that do not convert. The work starts with getting very honest about where your upper‑funnel dollars are going today and which of those environments are actually driving outcomes.

Finding the upper‑funnel impressions that are worth keeping

Within the Amazon ecosystem, Amazon Marketing Cloud (AMC) makes it clearer than ever where your upper funnel dollars are going and what is coming out of them. You can see what happens after that first impression and which combinations of audiences, campaigns, sites, and frequency levels move people from impression to product detail page to purchase. That visibility is a big part of what has turned Amazon STV into a more credible performance channel.

The first job is to use that visibility to find the waste. We most often see it when advertisers run display campaigns on sites or CTV/STV campaigns across publishers that do not drive outcomes or hit the same shoppers too many times with the same ads.

In other words, reducing waste starts with where you do not show up and how often you choose not to appear. Once you understand that, you can decide where to reinvest.

Reducing waste starts with asking better questions

For years, the programmatic ecosystem has rewarded scale. Publishers focused on volume. Advertisers chased cheap CPMs and big impression counts, often without a clean way to see if that spend turned into product detail page views, purchases, or new-to-brand buyers. Over time, it has become clear that this approach does not reliably drive the outcomes that matter for brands.

Instead of asking only how many impressions you bought and at what CPM, you can ask something more useful when it comes to your Amazon DSP and STV strategy:

Which impressions actually change shopper behavior and which ones do not?

That question is at the core of Flywheel’s ADSP Site Filtering & Optimization models. We use AMC insights to look past surface metrics and focus on closed-loop retail outcomes, then apply those to two levers that matter a lot more than impression volume: the sites where your Amazon DSP campaigns run and the frequency with which you reach individual shoppers on Amazon DSP and Amazon STV.

The goal is to help your brand stop paying for impressions that never convert and put that budget back into impressions that do.

Using site filtering to focus on the right inventory

On Amazon DSP, not all impressions are created equal.

With AMC, you can see what happens after those impressions serve. Some sites and apps lead to a healthy chain of detail page views, add to carts, and purchases. Others deliver a lot of volume but very little in the way of real outcomes.

Site filtering is the practice of using that downstream performance data to decide where you want your Amazon DSP campaigns to run and where you do not. In our ADSP Site Filtering model, that looks like:

• Using AMC to evaluate Amazon DSP sites and apps at the site level, not just at the campaign level

• Scoring each site on retail performance signals such as detail page view rate, purchase rate, and new to brand rate rather than only CTR or impressions

• Prioritizing delivery on sites that consistently drive those outcomes, and deprioritizing or excluding sites that do not

In practice, that means taking budget out of the cheapest CPM inventory and into the sites that consistently drive real outcomes. Site filtering goes beyond a brand safety check and is a way to shift budget out of weak inventory and into placements where your shoppers are more likely to convert.

Using frequency capping to stop overpaying for the same shoppers

The other half of the waste problem is how often you show up. Everyone has experienced the feeling of seeing the same ad over and over again. It is annoying for the shopper and rarely a sign of smart budget use. At a certain point, more impressions do not make someone more likely to buy. They just make your media plan more expensive.

Frequency capping is the control you use to prevent that. It limits how often an ad is shown to the same person over a specific time period. The concept is not new, but what is relatively new with Amazon is the ability to tie Amazon DSP and Amazon STV impressions directly to sales through AMC. That makes it much easier to answer the real question:

At what point does additional frequency stop changing outcomes for this brand and these campaigns?

Our view on frequency is straightforward:

• There is no single magic number that works for every brand

• The target is incremental, long term revenue, and AMC is the right place to measure it

Using AMC, our ADSP Site Filtering & Optimization models look by brand and by tactic to see when additional impressions stop moving key metrics such as detail page view rate, purchase rate, or new-to-brand rate. For one advertiser, we asked the question: at what frequency did additional impressions stop driving incremental sales?

The answer was clear. Roughly 90 percent of monthly sales came from users who saw ads 18 or fewer times. Users who saw ads between 18 and 49 times accounted for only about 10 percent of sales, while consuming a significant share of impressions and spend. Beyond that range, extra impressions were not nudging people toward purchase, they were just adding cost.

How this works together on Amazon DSP and STV

On Amazon, these levers work best as part of one connected system.

AMC gives you closed loop visibility into who buys, after how many impressions, and from which sites and formats. Those insights inform how you build your AMC Audiences to make sure you reach high intent shoppers where and when it matters most. Site filtering keeps your open web ADSP campaigns focused on placements that drive retail outcomes. Frequency caps (which can inform exclusion audiences) keep you from overserving the same shoppers on Amazon DSP and Amazon STV.

When you use all of that together, you move from a plan built around buying large amounts of media to a plan built around specific cohorts, proven inventory, and a frequency that maximizes incremental sales and new to brand growth. In that world, where you show up and where you choose not to show up are both strategic decisions grounded in what your data is telling you.

A brand-specific solution by nature

Our ADSP Site Filtering & Optimization models were built to flex by brand.

Every category, brand, and objective is different. An emerging brand trying to break through in a cluttered category will have a different ideal frequency and site mix than a mature brand defending share with high repeat rates.

By focusing on the outcomes a brand cares about most, such as incremental sales, new-to-brand growth, or long-term value, we can ignore a lot of the noise in the broader programmatic ecosystem and focus on more deterministic value creation within Amazon’s walls.

The point is not to force every brand into the same rule set, but to use AMC insights to set guardrails that match your reality.

Making your Amazon DSP and Amazon STV budget work harder

If your Amazon DSP and Amazon STV reports make you suspect that part of your budget is tied up in impressions that do not move the needle, you need to analyze where waste is occurring, eliminate it, and reinvest those dollars into campaigns and audiences that will.

That is how you do more with less. You reduce waste at the site and frequency level and reinvest that money into the audiences, placements, and formats that are more likely to convert.

If you want a partner to help you apply this thinking across your Amazon DSP and Amazon STV strategy, let’s connect and dig in together.