agentleFS
aaronpdohertyaaronpdoherty/meta-ads-genius/skills/opportunity-spotting/SKILL.md

opportunity-spotting

Skill1 starsChanged 3 months agoMITView on GitHub

Find growth opportunities in and around a Meta Ads account — underexploited audiences and angles in account data, competitor gaps via Meta Ad Library research, seasonal/auction timing windows, offer and market expansion. Use when the user asks where to grow, what to test next, for competitor/market research, new audience or angle ideas, or why performance has plateaued.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every opportunity gets: estimated audience size (range) · expected cost per result range (analogous segments or vertical benchmarks, …

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Meta Ads Opportunity Spotting

Opportunities are found where evidence already points and competitors aren't looking. The output is never "ideas" — it's an opportunity brief: a falsifiable hypothesis, graded evidence, a size estimate, and a bounded test design ready for the campaign-management skill.

The five lenses (run the ones the question calls for; a full sweep runs all)

1. Account mining (strongest evidence — start here)

Work from performance data (analysis reports if the workspace has them, else pull --daily insights):

  • Efficient-but-starved segments: strong cost per result, low spend share, and a flat marginal curve (daily CPA not rising as spend grows — the breakdown-effect test from the ad-analysis skill distinguishes real headroom from already-harvested pockets).
  • Coverage gaps: compare live creative against the buyer-situations list (creative-playbook) — situations with zero ads are free test ideas.
  • Saturation signals as redirects: frequency climbing + CTR decaying in a segment = its budget needs a destination; that destination is the opportunity.
  • Untested rungs: audience tiers never tried (next lookalike %, new seed source), formats absent, placements excluded historically for reasons nobody remembers.

2. Competitor recon (medium evidence)

Meta Ad Library shows every active ad for any advertiser, free, no auth. Protocol in references/ad-library-recon.md. Core signal: longevity ≈ money — an ad running 3+ months is very likely profitable. Map competitors' angles/offers/formats; the white space (angles nobody runs, offers nobody makes) is yours to test first.

3. Timing & auction windows (medium evidence)

  • Q1: cheapest CPMs of the year — the aggressive testing/scaling window; front-load creative in December.
  • Q4: CPMs +30–80%; lead-gen accounts bank budget and creative for January.
  • Vertical buying cycles (from _config/account.md seasonality notes): plan campaigns to land when the ICP decides, launch 2–3 weeks earlier for learning.
  • Platform shifts: new placements/formats typically under-priced early (early-majority CPM discount) — worth a bounded test when format-fit exists.

4. Audience & market expansion (weaker evidence — size carefully)

Expansion ladder, nearest first: next lookalike tier → new seed (different best-customer slice) → adjacent segment/vertical → new geo. Prefer the top-5% in-market framing: who is buying now? Target them first; expand to demand creation only when capture saturates.

5. Offer innovation (weaker evidence, highest ceiling)

Match offer to traffic temperature: cold traffic wants low-commitment value (benchmark/report, ROI calculator, assessment) — not a demo — unless the product is low-ACV/self-serve where direct trial/demo can work cold. Warm: case study, comparison, product tour. Hot: tailored assessment. A quality plateau across many creative tests usually means the offer is the ceiling — the "nuclear option" escalation from the optimization skill lands here.

Sizing & the brief

Every opportunity gets: estimated audience size (range) · expected cost per result range (analogous segments or vertical benchmarks, sources labeled) · budget to significance (3 × target per concept tested — from the optimization skill's data floor) · payback logic. Grade every evidence item: account data > competitor behavior > market inference.

Brief template in references/opportunity-lenses.md — hypothesis, evidence table, sizing, test design (audience/creative/budget/duration), success & kill criteria, go/no-go recommendation. A brief that can't be tested within the workspace's test budget guardrail gets shrunk until it can.

What to avoid

  • Trend-chasing with no offer/audience fit; novelty is not evidence.
  • Copying a competitor's creative (copy the angle, differentiate the execution — and never copy much-larger advertisers whose economics you don't share).
  • Point-estimate sizing theater — ranges with stated assumptions.
  • Opportunities requiring account-structure upheaval to test — smallest viable test wins.

References

  • references/ad-library-recon.md — the competitor research protocol, step by step, with what to record.
  • references/opportunity-lenses.md — full lens checklists, sizing method, brief template.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.