Meta stopped rewarding better targeting. It rewards better creative.

Before we produce a single asset, we watch your creative and read it against the account: which concepts are carrying spend, which have gone stale, whether the structure is fighting the creative, and whether the offer is the real problem. Most accounts we look at have a creative problem being treated as a media buying problem.

We map your creative against what the account has and hasn't tested, then produce against the gaps. Every asset ships coded and tagged, mapped to a testing hypothesis, with the media buying logic decided before it's shot.

Structured tests, read once the data is actually reliable - usually one to two weeks. Winners scaled into primary spend, losers documented against the taxonomy so the learning survives the ad. Quarterly, the whole set gets analysed at concept level to write the next roadmap.

Less audience targeting than you'd think. Advantage+ and broad targeting have moved most of the performance levers into the creative itself, so the job now is producing enough varied, well-briefed creative to give the algorithm something to work with - and having a system to learn from what it does. Agencies still selling audience strategy as the core service are describing a job Meta largely automated.

We work with brands spending from around $15k/month across paid media. Below that, there usually isn't enough data velocity for structured creative testing to pay for itself, and you're better off with a leaner setup than an agency retainer.

UGC ads run creator-style content from your brand's account - we script and produce ours in-house rather than collecting user submissions. Partnership Ads - Spark Ads on TikTok - run from the creator's own handle instead, with their name and following attached, and the content is usually less brand-planned as a result. Both formats ran in Vitable's account across 2023-24, and at six-figure monthly spend the Partnership Ads came in 51% below the UGC ads on cost of acquisition.

Month one is audit, roadmap and production, so expect diagnosis rather than a step change. Structured testing runs through month two. Compounding usually shows up around month three, once the creative taxonomy has enough data to say something durable about which concepts work for your account.

We produce it. Scripted video, creator content, statics and lo-fi, in-house, unlimited within the retainer. If you have an existing creative team we'll work alongside them, but we'd rather brief and produce ourselves because the media buying data is what shapes the brief.

Yes - they're one platform for buying purposes. Placement strategy differs because Reels, Feed and Stories behave differently - the same idea needs a different cut for each, and some ideas only land in one of them.