Every Q4 gets billed as the most competitive one yet, and most years that is just seasonal drama. This year we think it actually holds up, and midterm season is a good illustration of why.
Political advertising has always been loud in an election year, but what's different now is where those ads are going. Connected TV is the fastest-growing media type in political advertising, holding about 23% of a cycle now projected at $11.6 billion. That’s the most expensive on record, and a figure AdImpact revised upward mid-year as Senate races heated up.
Those dollars are not competing with you for customers, but they are competing with you for streaming inventory, in the same weeks you were planning to buy it. And political spend is only one of four shifts landing in the same auctions, making this quarter structurally different rather than just busier.
We spend a lot of the fall helping brands and agencies pressure-test Q4 plans. Almost nobody asks us how much they should spend. What they want to know is how to stay visible when everything is pulling on the same inventory at once. Our answer is that most of it comes down to decisions you make before you go live.
If political spending was the only thing happening, you could plan around it. But the other three shifts make the picture a little more complicated.
Retail media keeps growing, but the growth is concentrated. EMARKETER projects that Amazon Ads and Walmart Connect will take 89% of the incremental US retail media dollars this year ($9.42 billion of $10.53 billion in net-new spend). Growth in a category does not mean growth is available to everyone in it, and that concentration is the planning constraint.
Streaming crossed a real threshold this year. Nielsen's Q1 Ad Supported Gauge put streaming at a record 46.6% share of ad-supported TV viewing, and connected TV upfront commitments passed primetime linear for the first time. Ad spend still has not caught up to where people's viewing time actually sits, which means there’s opportunity there.
Search is the one most people underestimate. SparkToro, working with Similarweb clickstream data, found that 68% of US Google searches ended without a click in the first four months of 2026, up from 60% in 2024. Fewer clicks reaching your site means more upward pressure on what the remaining clicks cost.
Political CTV budgets, retail dollars, and brand budgets are all chasing the same streaming inventory in the same weeks. Teams that start planning in September tend to be much calmer in October, when retail media season really kicks off. Teams that wait for Black Friday are already behind.
Someone watching the game on Sunday night has a phone in hand, scrolling social through the commercial breaks and probably during the game too. That person is both a streaming viewer and a mobile social user in the same 15 minutes, which means you have two chances to reach the same person inside one moment.
Plans that treat the TV buy and the phone buy as one connected journey can sequence the message across screens, so the second impression builds on the first instead of repeating it. Your own first-party data is what makes that coordination possible. Setting exclusions deliberately by platform keeps your campaigns working together rather than competing for the same impression, which protects your costs.
The same thinking pays off in your creative. Think about how differently you would pitch the same product to someone browsing from the couch versus someone already out running errands. "Order now, arrives tomorrow" speaks to the first, while "in stock, 10 minutes from you" speaks to the second. The creative covers the same product in the same season. It’s just differentiated to meet shoppers where they are.
And there’s room to work with here, because shoppers move between those modes constantly. Salsify's 2026 consumer research found that about 67% of shoppers “webroom,” which means they’re researching online and buying in store the same day. Meanwhile, about 53% “showroom,” checking a price on a phone while standing in the aisle. Having both versions of the creative ready before the quarter opens means you can meet either behavior as it happens.
All of that movement between online and the store is visible to somebody, and increasingly that somebody is the retailer. Sponsored spots on a retailer's own site are finite and usually claimed by whoever has been bidding there longest. The real growth opportunity comes from the fact that retailers are letting their first-party purchase data travel beyond their own sites, activated wherever the shopper already is.
Amazon is the obvious example, but what’s newer is how much company it has now. Walmart, Kroger, and Instacart all sit on an abundance of purchase data of their own, and they are also now offering offsite activation. Each one knows its own shoppers as well as anyone does, and each is cheaper to test at Q4 volumes, which makes the interesting question not whether to run Amazon but which second network earns the next dollar and whether you can hold both to the same closed-loop measurement standard.
Streaming is where most of that portable data is landing. Picture someone watching a cooking show while an ad for the ingredients arrives on their phone through a grocery retailer's app at exactly the right moment. That’s one shopper reached across two screens using data the retailer already had. This is why planning retail media and connected TV together, rather than as separate budget lines, can be so impactful.
Once data is flowing across multiple retailers and multiple screens, no single platform captures all of it. Each major demand-side platform (DSP) holds its own exclusive inventory and its own exclusive signal, which is why most strong plans buy across more than one.
None of these is better than the others. They are built for different roles, and the platform you reach for should be whichever one holds the inventory or signal your objective actually needs. If you are weighing whether your current setup still fits, we walked through the signals to watch in a separate post on re-evaluating your primary DSP.
The real complication with multiple platforms is measurement. They report differently by default, so comparing performance across them is not quite apples to apples. An ad server like Campaign Manager 360 (CM360) helps consolidate that view, and settling on a single source of truth early keeps every platform comparison honest.
Fragmentation is not only a programmatic problem. It shows up in search and social, too, for a completely different reason, and it changes how you should split budget between them. AI is pulling the two apart, because they were never really the same kind of behavior to begin with.
Someone whose dishwasher just broke searches for an answer and acts on it immediately. That same person, scrolling later that night, sees an ad for a dishwasher brand they have never heard of and stops. Search meets demand that already exists. Social creates it. When both are judged on the same metric, one of them always looks like it is failing at a job it was never doing.
AI Overviews are what turn this from a nuance into a budget decision. You’ve probably noticed it in your own searching. You ask something, you get the answer, and you never click anything. Your intent was real, but it just got satisfied higher up the page. The clicks still coming through are fewer and later, but they tend to carry more intent when they arrive.
This is a good conversation to have with your stakeholders before the quarter. Lower organic sessions year over year reflect a change in how the results page works rather than a drop in performance, and our post on AEO and GEO foundations covers how to measure visibility.
The same care also applies on social, where TikTok skews younger and Meta's audience has been trending older, so the platform that reaches your actual audience matters more than the one with the biggest reach number.
Budgets are getting locked earlier this year even as shoppers start buying earlier, so sequencing has to be right before anything goes live. There is very little room to fix it in flight. Before launch, get agreement on:
Every one of these shifts is a place where a plan can get better, and the teams who treat them that way tend to walk into October with a lot more confidence than the ones scrambling in November. Working through it does not have to be all or nothing either.
If your team owns execution and wants the plan pressure-tested, we can take a look at your Q4 channel mix, platform access, and where the gaps are. If you want channel experts running activation alongside you, across every platform we just walked through, we can also help with media strategy. And we’re always happy to help with measurement, which sits underneath both, because a reporting standard set before launch will make your Q1 reporting life easier.
Two things we lean on a lot in a quarter like this:
We covered all of this in more depth in our full session. Watch the Q4 media planning webinar if you have some time, and reach out if you would like a second set of eyes on your plan before flights go live.