How-To

Set up a Meta Sales campaign with AI

One task, a few steps. Run the prompts in your approved AI workspace.

Metasetup
← All how-tos

Where to run this

Your approved ChatGPT, Claude or Gemini workspace, files uploaded first.

Upload first

  • Approved campaign brief and KPI
  • Budget, markets, languages
  • Landing pages and product catalog
  • Pixel and Conversions API status
  • Historical performance and creative assets, if reusing or cloning a campaign

Steps

  1. 01

    Pick your starting point before you build a blueprint

    Do this

    New advertiser, new market, new product launch, no suitable historical campaign: build fresh. An existing sales campaign with credible performance history: audit it, don't redesign a winner without evidence. A proven campaign that needs to launch in a new market: clone and adapt it.

  2. 1a

    New campaign: generate the blueprint

    AI prompt
    You are designing a complete Meta Sales campaign. Using only the approved materials provided, generate: a campaign blueprint, campaign structure, audience framework, catalog framework, creative framework, and measurement framework. State whether the account meets a reasonable bar for Meta's automated Advantage+ structure (broadly: established conversion data, a catalog with meaningful SKU coverage, Pixel and CAPI both live) versus a manually-structured campaign, and flag if there isn't enough evidence supplied to make that call. Do not invent products, audiences, offers, claims, promotions or budgets. Finish with READY TO BUILD or NOT READY.

    Do this

    As of February 2026, Meta merged its manual and Advantage+ flows into one unified campaign creation interface, AI-driven optimisation (audience, placement, budget) is enabled by default with individual toggles, not a separate campaign type you opt into. What used to be called Advantage+ Shopping is now Advantage+ Sales, and it now supports services and digital products, not just physical catalog items.

  3. 1b

    Existing winning campaign: audit it, don't just redesign it

    AI prompt
    Review the attached Meta Sales campaign and the supplied performance evidence. First identify what appears to be working, what should remain unchanged, and what is not supported by sufficient evidence. Then identify: catalog opportunities, creative refresh needs, existing-customer budget allocation issues, and scaling opportunities. Use only: KEEP, TEST, EXPAND, REMOVE, HUMAN REVIEW, NOT EVIDENCED. Do not redesign a successful structure without evidence. Do not invent performance results. Do not approve the campaign. Finish with STRUCTURE CAN BE REUSED or STRUCTURE REQUIRES REVISION.
  4. 1c

    Cloning into a new market: adapt, don't rebuild

    AI prompt
    Use the attached high-performing Meta Sales campaign as the source blueprint. Preserve: campaign structure, audience framework, catalog framework, and measurement framework. Review whether changes are required for: market, product or catalog availability, language, currency, shipping and local regulatory restrictions, and creative availability. Classify every recommendation as: KEEP, LOCALISE, REMOVE, ADD, HUMAN REVIEW, NOT EVIDENCED. Do not redesign the campaign unless the supplied local-market evidence requires it. Do not approve the campaign. Finish with READY TO CREATE LOCAL BLUEPRINT or LOCAL EVIDENCE REQUIRED.
  5. 02

    Build tracking and measurement before anything else

    AI prompt
    Review my Meta Sales tracking setup against the attached blueprint. Validate: Meta Pixel, Conversions API, domain verification, Aggregated Event Measurement, conversion event hierarchy, and catalog connection. Use only: MATCH, MISMATCH, MISSING, NOT EVIDENCED. Do not infer missing settings. Do not approve setup.
  6. 03

    Build campaign and audience structure

    AI prompt
    Review this Meta Sales campaign and audience structure against the attached blueprint. Review: campaigns, ad sets, prospecting audiences, retargeting audiences, customer lists, catalog audiences, and exclusions. If this is an Advantage+ structure, confirm the existing-customer budget cap is set deliberately, not left at a default, since it directly controls how much spend can go to retention versus new customers. Use only: MATCH, MISMATCH, MISSING, CONFLICT. Do not infer missing information. Do not approve setup.

    Do this

    An Advantage+ campaign runs as a single unified ad set with a campaign-level budget only, there's no ad-set-level budget tier the way a manual structure has, don't look for one.

  7. 04

    Build creative assets

    AI prompt
    Review these Meta Sales creative assets against the approved guidelines and catalog. Check: images, videos, carousels, collection ads, catalog ads, primary text, headlines, descriptions and CTAs for relevance, duplication, unsupported claims, and missing assets. Use only: PASS, EDIT, REMOVE, HUMAN REVIEW, NOT EVIDENCED. Do not invent offers, claims or promotions. Do not assume policy approval. Do not approve the campaign.

    Do this

    Upload genuinely distinct creative, not minor variations of one asset, an automated structure needs real variety to test against, a handful of near-identical images gives it nothing to learn from.

  8. 05

    Reconcile the build against the blueprint

    AI prompt
    Compare the approved blueprint against the completed campaign build. For every setting, provide: SETTING, APPROVED VALUE, ACTUAL VALUE, STATUS, EVIDENCE, REQUIRED ACTION. Use only: MATCH, MISMATCH, MISSING, NOT EVIDENCED, NOT APPLICABLE. Do not infer missing settings. Do not modify the campaign. Do not approve launch.
  9. 06

    Final launch readiness review

    AI prompt
    Review the final Meta Sales campaign evidence pack. Verify: Pixel implementation, CAPI implementation, audience implementation, prospecting coverage, retargeting coverage, catalog readiness, creative coverage, landing-page readiness, and policy status. Use only: PASS, BLOCKER, HIGH RISK, NOT EVIDENCED, CONDITION ACCEPTED. Missing evidence is not a pass. Do not approve launch. Do not recommend optimisation actions.

    Do this

    Have a second person check the campaign before it goes live.

Practitioner notes

Notes from other practitioners

What actually happened when you ran this. Gotchas, better prompts, platform quirks. Notes post immediately and are visible to everyone.

No notes yet.