Show Whats Working and What Isnt

Task Library

Present performance data as an explicit two-column verdict – winners with evidence, underperformers with evidence – so the weekly report drives kill/scale decisions instead of hiding them.

Category: Strategy & Measurement

Use this when assembling the weekly MAA report, or any time a stakeholder asks "how's it going?" and deserves a straight answer instead of a wall of screenshots.

Inputs

  • Current-period metrics for every active item: campaigns, boosted posts, articles, pages
  • The thresholds that define "working" – target CPA/ROAS, engagement benchmarks, ranking/traffic goals (from GCT and SAE Stage 2)
  • Last period's numbers for trend context

Steps

  1. Fix the verdict thresholds first: an item is "working" only if it meets its stated target (e.g., at-or-under target CPA, at-or-above engagement benchmark). Write the threshold at the top of the report so the verdicts are checkable.
  2. Score every active item against its threshold – no item gets to be "too new to judge" for more than one cycle.
  3. Build two lists. Working: each item with the one metric that proves it, plus trend vs last period. Not working: same evidence standard – no euphemisms, no burying.
  4. Lead with winners (they get scaled), follow with losers (they get diagnosed and killed) – but never omit the losers; hiding them is how budgets die.
  5. Keep each line to one sentence: item, verdict metric, trend. Detail lives in the linked dashboards, not the verdict page.
  6. Hand both lists downstream: losers go to analyze-why-underperformers-fail; the combined picture feeds kill-underperformers-scale-winners and the Friday report.

Definition of done (QA checklist)

  • Verdict thresholds stated explicitly at the top
  • Every active item appears in exactly one list – working or not working
  • Each line carries its proof metric and period-over-period trend
  • Underperformers are as visible as winners (equal evidence, equal prominence)
  • Both lists handed into the Friday MAA report unedited
  • Linked back to the definitive article and relevant siblings
  • Complies with Blog Posting Guidelines (if it publishes content)

Example(s)

  • Example needed – run the Meta-Article Prompt after first real run.

File in the zip: skills/strategy-measurement/show-whats-working-and-what-isnt.md. Download the Task Library zip.

Index: Task Library. Article guidelines.

Scroll to Top