Skip to content

meta-launch-architect: the launch spec

Launch & scale Cowork-ready

Tells you exactly how to structure a new Meta launch, and shows the arithmetic behind every number. It encodes the current consolidation playbook:

  • One-campaign CBO structure, rather than the sprawl that starves every ad set of learning.
  • Budget math derived from your CPA, not from what feels right.
  • The objective stair-step, so you are not asking a cold account to optimize for a conversion it has never seen.
  • A creative-count cap with the proven mix, plus partnership-ad allocation.
  • A naming convention, instantiated per ad rather than described in the abstract.
  • Admission-ticket expectations: what the account has to spend before the structure can prove anything.

Every number cites a dated decision rule, so you can see what it is based on and when that was established.

Read-only. The structure ships as an ordered manual runbook you paste in. It does not create objects in your account.

  • Launching a new ad account from zero.
  • A new product or offer that needs its own campaign structure.
  • Your current structure is sprawling and you want to know what to consolidate to.
  • You want a defensible answer to “why this budget, at this level?”
Cowork ✓ Desktop ✓

Both.

  • Your target CPA or cost per lead, and your declared conversion shape.
  • The offer and price.
  • How many creatives you actually have ready, because the plan sizes to reality.
  • A verified signal layer. It expects tracking and signal to have been checked first, and says so if they have not.
  • The full structure: campaign, ad sets, ads, named per the convention.
  • The budget arithmetic, worked out rather than asserted: what the learning floor requires at your CPA, and therefore what each ad set needs.
  • The creative mix and count cap.
  • The objective sequence and when to step up.
  • An ordered, copy-pasteable build runbook, and the handoff into scaling.

You: “How should I structure this launch?” (target CPA $45, 9 creatives ready)

Structure: 1 CBO campaign, 3 ad sets. Budget math shown: the learning floor needs about 50 conversions per ad set per week; at $45 CPA that is $2,250 per ad set per week, so $6,750 weekly campaign budget, or $965/day. Creative: 9 fits the cap at 3 per ad set in the standard mix. Admission ticket: expect roughly 2 weeks and $13.5k before verdicts are meaningful. Objective: start on the mid-funnel event, step to purchase at 50 weekly.

  • The admission ticket is the number people skip and then regret. If the budget cannot fund the learning floor, the structure cannot work no matter how good the creative is. Better to run fewer ad sets properly.
  • Do not add ad sets for tidiness. Every extra one divides the same budget and pushes more of the account below the floor.
  • It does not build anything for you. You paste the runbook in. That is deliberate in this version.
  • Get tracking verified first. Launching into unverified measurement is the hard block in new-client-launch.

Runs inside new-client-launch. Hands off to meta-scaling-copilot once live. Expects tracking-doctor and signal-surgeon to have cleared first.