What are you really paying your agency for?
The platforms have absorbed most of campaign setup. So what is left to pay an agency for — and what would an AI operating layer around your campaigns actually do that siloed reporting cannot?
Most retail marketing teams are not short of campaigns. They are short of clarity and control around them. Reporting cannot be compared cleanly across platforms or reconciled against real sales. Ads keep running against out-of-stock products. And as Meta, TikTok and Google have pulled campaign setup into their own automated products, the manual work an agency once owned has quietly shrunk — while the dependency, the cost, and the back-and-forth of passing briefs and content remain.
It is a fair question to ask: what are you actually paying for? The honest answer is that a lot of value now lives in the operating layer around campaigns — reporting, stock awareness, briefing, institutional memory — and that layer is usually manual, fragmented, and disconnected from the numbers that matter.
An AI operating layer targets exactly that. The highest-value piece is reporting that finally reconciles: a single view comparing platforms like-for-like and tying campaign performance back to actual company sales — something siloed agency reporting structurally cannot do. Next is stock awareness: connect merchandising signals so the system can flag, and pause, spend against out-of-stock products before budget leaks. Then a coordinated lifecycle — signal to decision to brief to creative to distribution to reporting — so each campaign is briefed consistently and each result informs the next.
The part most teams worry about is control, and here the model matters. Most human-in-the-loop AI treats approval as a permanent gate. Real teams do not work that way — they onboard a new agency by reviewing everything closely for a few months, confirm it understands the brand, then deliberately speed up. A well-designed agent earns autonomy the same way: full review at first, then approval requirements dialed back by category as trust is proven — while material budget, launch and brand decisions always stay with a person.
None of this is an agency replacement. It coordinates and sharpens the work around the agency, and gives the team the reporting and control an agency cannot. And it should prove itself, not assert — run as a benchmarked proof-of-concept alongside the incumbent, measured on the metrics you already care about.
Read the full story
Rebuilding the operating layer around retail campaigns