MARKETING

Detect CTR Decay in BigQuery and Auto-Draft Figma Refresh Briefs

Scans daily ad performance in BigQuery for creatives whose click-through rate has decayed past a threshold.

CategoryMarketing
Enginesim
Difficultyintermediate
Triggerschedule
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerDaily schedule fires
  • ActionQuery per-creative CTR (7d vs 28d) in BigQueryGoogle BigQueryBigQuery
  • LogicFilter to creatives past the decay threshold
  • ActionDraft refresh brief from decay patternOpenAI
  • OutputCreate brief page in Figma refresh boardFigmaFigma

What it does

Each morning this workflow reads your ad-level performance table in BigQuery, compares each creative's rolling 7-day CTR against its launch-window baseline, and flags any creative whose CTR has dropped past your fatigue threshold (default 25%). For every flagged creative it generates a refresh brief — the underperforming hook, the angle that's worn out, and the audience it's still running against — and creates a brief page in your Figma project so the design team can begin a replacement variant immediately.

When to use it

Run this when you have steady ad spend across many creatives and can't manually watch every one decay. It catches fatigue before wasted spend piles up and removes the lag between "this ad is tired" and "someone is briefed to fix it."

How it works

  1. 1A daily schedule fires the workflow.
  2. 2It queries BigQuery for per-creative spend, impressions, and CTR over the trailing 7 and 28 days.
  3. 3A decay check filters to creatives whose CTR fell past the threshold while spend stayed material.
  4. 4For each flagged creative, OpenAI drafts a concise refresh brief from its metadata and decay pattern.
  5. 5The brief is written into Figma as a new page in the team's refresh board.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect BigQueryDatasets, queries, schemas.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect FigmaFiles, frames, comments, assets.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.