Why creative needs managing as actively as media

A paid social account has two variables, media and creative, and the industry manages one of them. Media has a strategy, a setup, a weekly optimisation rhythm, a measurement standard and an owner, and it has all five correctly. Creative, on a medium where it sets the price of every impression and decides who the ad reaches, needs the same five things, and in most accounts it has none of them, not because anyone is careless but because the operating model that handles creative was designed for a medium where the work could not change what the media cost. On paid social it can, so the two variables have to work together in the same way. Just as media has a strategy and is optimised regularly by adjusting settings, creative now has to be actively managed, measured and adjusted for a campaign to work.

The two-column test

Draw two columns on a page, media on the left and creative on the right, and five rows down the side, strategy, setup, weekly optimisation, measurement, owner. Then fill in the left column from your own account, which takes about a minute because the answers exist.

Media strategy is reach and frequency goals by audience, flighting, the budget split by objective and platform, written down and signed off. Media setup is the campaign structure, the objective on each campaign, the audiences, placements, bids and budgets, built before launch and checked against the plan. Weekly optimisation is pacing, frequency management, budget shifts between campaigns and audiences, bid changes and placement trims, made from the account's data by someone whose job it is. Measurement is CPM, reach, frequency and cost per result, against the plan and against last quarter, in a dashboard someone reads. The owner is a media agency or an in-house buying team, with a name, a weekly meeting and a contract.

Now the right-hand column. Creative strategy is not the campaign idea, the idea belongs to the brand platform, it is the portfolio, the twenty to thirty distinct hypotheses about which category buyers the ads will reach and which category entry points they will meet them in, each a different person, situation and opening. Creative setup is each ad paired with one job and one objective at the brief, and a pretest gate every ad passes before spend attaches. Weekly optimisation is the pause, remix and replace decision, per ad, from the account's own data, with the learning written into the next brief. Measurement is stop rate, hold rate and watched rate per ad on one platform and one threshold, read after effectiveness and before cost, against the brand's own prior creative. The owner is a name.

In most accounts I have seen, the left column comes back full and the right column has a campaign idea in the top cell, a delivery schedule in the second, and three empty rows. That is not a verdict on the team. It describes where the operating model stopped, and the empty cells are the unlock, because they describe management that has not yet been applied to the account's other variable.

The platforms say the hard part for us

The reason the right column now matters is mechanical, and the platforms have said it in their own words. Meta describes its auction as ranking each ad on three factors, the bid, the estimated action rate, meaning the probability the person takes the action the campaign is optimised for, and ad quality. Two of the three are properties of the creative, so an ad people watch wins impressions it would otherwise have lost and wins them cheaper, and an ad people skip pays more for what it still gets. The creative sets the price of the media, per impression, in real time.

Delivery went the same way. Meta's engineering account of Andromeda, its 2024 ads retrieval system, says that "increased ad diversity can improve people's experience with ads and drive better advertiser outcomes", and its business guidance the following year said that with AI-enabled delivery "the focus has shifted from niche targeting to creative diversification as the best lever to find the most relevant audiences". Its planning advice to advertisers is broad audiences, target frequency and diverse creative, which is the platform telling you where the decisions moved. As the automation takes on more of the audience and placement choices that used to be set by hand, the creative becomes the lever the advertiser still holds that steers who sees the ad. The platform that sells the media is saying the right-hand column is where the account is won.

The spread inside one media plan shows what is at stake. Meta's own test data has the winning creative in 235 brand lift tests at 13% lower cost per ad recall than the average of the other creatives in the same test, and the winner across 14,780 A/B tests at 21% lower cost per website conversion. The plan, the budget and the audience were the same in every test, the work was not. A variable that moves cost that much under a fixed media plan needs managing weekly, in the same way the media is.

The media half, kept honest

None of this makes media easy or makes it matter less, and it is worth saying in the same breath. The Nielsen marketing mix meta-analysis Meta commissioned across 119 CPG brands, modelled and presented by the platform, put frequency, placement and reach as the three biggest executional drivers of sales impact, at 31%, 24% and 22% of relative importance, with continuity at 12%, targeting at 9% and the choice of optimisation goal at 1%. Those are media decisions, and they are the necessary condition, because a creative nobody is served does nothing and a good ad served at the wrong frequency in the wrong placement is wasted. The media team's remit does not change in any of this. Bidding, placement, audience settings, budget allocation and campaign setup stay where they are, and the left column stays full. The argument is that the right column should be as full, not that the left one should empty.

And vice versa

The dependency runs both ways, and poor setup starves the best creative as surely as poor creative wastes the best setup. The wrong objective is the common case. A brand-building ad run under a video views or ThruPlay objective sends the system looking for people who watch anything, so the ad collects completions among the most tolerant viewers in the market and reach narrows, which is why the creative is optimised for attention and the objective stays set for the brand's goal, reach for brand building. We read attention, we do not buy it. A capped audience does the same damage differently, an interest stack or a lookalike built up over years draws the fence so tight that the delivery system has nowhere to take a distinct ad, and the portfolio's variety is spent on one slice of the market. A consolidating structure, budgets pooled so that delivery concentrates behind whichever ads win the first day's auctions, leaves the strategically important ads live and barely served. Frequency spent on skips is the quiet one, a plan buying a frequency of eight, say, is buying eight impressions, and if seven are skipped in the first second the watched frequency is one, so the plan's frequency and the brand's are not the same number. And under-weighted media starves everything. The Advertising Research Foundation's 2026 validation work found that "media weight plays a critical role in translating attention into outcomes" and that attention "should be considered alongside media weight, reach and frequency rather than treated as a stand-alone proxy for outcomes", which is the right caution for anyone reading this as a case for creative over media. It is a case for both columns being managed, by their owners, in the same week, against the same scoreboard.

Won't the platform put all the spend behind one or two ads?

Yes, if you let it, and it is the first thing a media team asks when a portfolio of twenty to thirty ads arrives. The delivery system has a narrow brief, for each person find the ad most likely to earn the optimised action at the lowest cost, and it does that superbly. Give it thirty ads that look alike, open alike and earn the same response from the same kinds of people and it will, in practice, put the spend behind the two or three that win the first auctions and leave the rest live and starved, because from the system's point of view they are one signal. Meta's published wording does not describe it that way, this is how the system behaves in accounts that have tried it.

Two things stop it, and both sit in the right-hand column. The first is distinct hypotheses, because an ad that opens on a different person in a different situation earns a different response from different people, so the system has a reason to deliver it rather than consolidate it, which is the whole argument of Why you need a creative portfolio, not more content. The second is the weekly read, because the difference between an ad the market rejected and an ad the structure starved is visible in the data, the first has impressions and no watched rate, the second has no impressions, and only someone looking per ad every week can tell them apart and act on each correctly. Consolidation is the platform doing what it was told. Managing creative is deciding what to tell it.

The asset view

The counter-position is the way almost every brand is organised and it was correct for seventy years. Creative is what you hand to media. The agency makes the work, the brand approves it, the media team places it, and the account is run from the media side because that is where the money moves. On every medium priced before the work existed this was the right design, the quality of the film could not change what a rating point cost, so there was nothing to manage after the handover except the media, and procurement built the working and non-working split on it.

It stops being true at the auction. Once the price of every impression and the choice of who sees it depend on the work, the handover is the moment the account's largest variable leaves management, not because anyone handed anything over, nobody did, but because the structure has a place for creative to be delivered and no place for it to be run. The asset view is correct right up until the auction, and the auction is where paid social starts.

What to do with this

Draw the two columns this week and fill in both from your own account, with the people who own each. Where the right-hand cells are full, you are running a creative system, whatever you call it. Where they are empty, name the owner first, because the other four rows follow from having one, and make the owner's first job the weekly read, stop rate, hold rate and watched rate per ad, one platform, one threshold, with the pause, remix and replace decision made in the same meeting. Then put the two columns side by side in the weekly review the media team already runs, so that a creative signal and a setting change are decided together, against the same numbers. Media creates the opportunity to see your ads. Creative determines whether they work. Both halves of that sentence are now managed weekly or neither is, and the strategy the right-hand column runs on is set out in How to apply brand science to social media.

Questions

Won't the platform just put all the spend behind one or two ads?

It will if the ads are similar, because the delivery system treats ads that earn the same response from the same people as one signal and concentrates spend behind the winners, leaving the rest live and barely served. That is observed behaviour rather than Meta's published wording, and it is the system doing its job rather than a fault.

It stops when the portfolio is made of distinct hypotheses, different people, situations and openings that earn different responses, and when someone reads delivery per ad every week and can tell a rejected ad from a starved one. The first has impressions and no watched rate, the second has no impressions, and they need opposite decisions.

Does this mean media matters less than it used to?

No. Frequency, placement and reach were the three biggest executional drivers of sales impact in the Nielsen meta-analysis Meta commissioned across 119 CPG brands, and the ARF's 2026 work found attention translates into outcomes only alongside media weight, reach and frequency. Setup is the necessary condition and the media team's remit is unchanged.

What changed is that creative became the second variable that moves cost and delivery inside that setup, so it needs the same five things media already has, a strategy, a setup, a weekly rhythm, a measurement standard and an owner. The argument is for two managed columns, not for one at the expense of the other.

Who should own the weekly creative decision?

Someone with a name, the account's data and the authority to pause an ad that is spending. In most rosters that is nobody yet, because the creative agency works at campaign cadence, the media agency has the data and no creative mandate, and the in-house team makes the adaptations, so the decision falls between them.

The owner can sit in any of those places or in a creative system built for the job, what matters is that the role exists, meets weekly and shares a scoreboard with the media team. Where a creative system fits with your agencies maps the roster around that decision without moving anybody's existing remit.

Sources

• Meta Business Help Centre, "About ad auctions". Ranking on bid, estimated action rate and ad quality. Platform documentation, Meta sells the media.
• Engineering at Meta, "Meta Andromeda: Supercharging Advantage+ automation with the next-gen personalized ads retrieval engine", 2 December 2024. "Increased ad diversity can improve people's experience with ads and drive better advertiser outcomes." Platform, Meta sells the media.
• Meta for Business, "Demystifying Creative Diversification", April 2025, as quoted by Jon Loomer, 2025. "The focus has shifted from niche targeting to creative diversification as the best lever to find the most relevant audiences." Platform wording via an independent consultant, Meta sells the media.
• Meta, "Demystifying attention. How to unlock brand growth", 2025, planning guidance. Broad audiences, target frequency, diverse creative. Partner-facing sales deck, Meta sells the media.
• Meta, "Creative and Media Excellence", circa 2023. Winning creative at 13% lower cost per ad recall across 235 brand lift tests (April 2021 to 2022) and 21% lower cost per website conversion across 14,780 A/B tests (April 2021 to April 2022), against the average of the other creatives in the same test. Meta internal test data, Meta sells the media.
• Nielsen custom MMM meta-analysis commissioned by Meta, "Global CPG Execution Best Practices To Maximize Meta Performance", 119 CPG brands, 2019 to 2021, as presented in Meta, "Creative and Media Excellence" and "Demystifying attention". Relative importance of frequency 31%, placement 24%, reach 22%, continuity 12%, targeting 9% and optimisation goal 1% in predicting sales impact. Modelled, commissioned by Meta, which sells the media.
• Advertising Research Foundation, Attention Measurement Validation Initiative Phase 3, press release 9 June 2026. Media weight, reach and frequency as conditions for attention translating into outcomes. Member-funded industry body.