agentleFS
kastriasanikastriasani/google-ads-skills/skills/google-ads-research/SKILL.md

google-ads-research

Skill1 starsChanged 5 months agoMITView on GitHub

Structure market and advertiser research before building Google Ads campaigns. Produces a research brief with a prioritized web-search query list, an ICP and persona table, a demand and SERP-landscape checklist, audience-signal hypotheses mapped to Google audience types, and offer and angle hypotheses. Use this first, before keyword research or campaign setup, whenever you are planning paid search for a new brand, vertical, or market. Trigger when the user says research a market, research an advertiser, build a research brief, study the demand landscape, scope a Google Ads account, analyze the competitive landscape for paid search, define the ICP or audience, or asks what should I know before launching Google Ads.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill at the very start of a Google Ads engagement:

    line 21
  2. Very specific:

    Run this before google-ads-keyword-research and google-ads-campaign-setup.

    line 29
  3. Distinctive:

    Run the P1 queries first. They carry the highest decision value: confirm commercial-intent demand exists, capture how buyers phrase …

    line 44
  4. Distinctive:

    Read the actual SERP for the money terms.

    line 49
  5. Distinctive:

    Search Console: existing query demand for the domain.

    line 77

≈ “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

Google Ads Research

Front-load the decisions that waste ad spend. Before a single keyword is chosen or a campaign is built, this skill structures market and advertiser research into a decision-first brief: what to search, who you are buying clicks from, how the auction and SERP look, which audiences to use, and which offers and angles to test. Every section must change a budget, bidding, targeting, copy, or landing-page decision -- if it does not, cut it.

When to use

Use this skill at the very start of a Google Ads engagement:

  • A new brand, vertical, or geo where you need to understand demand before committing budget.
  • Scoping or auditing an account and you need a structured view of the market, ICP, and competitive landscape.
  • The user asks to research a market or advertiser, define the ICP, study the demand landscape, or figure out what to know before launching paid search.

Run this before google-ads-keyword-research and google-ads-campaign-setup. For competitor conquesting and battlecards specifically, hand off to google-ads-competitor-intel.

Workflow

  1. Gather the four inputs: brand or url, vertical, geo, and goal (leads, sales, traffic, awareness). These are enough to scaffold the brief.
  2. Generate the brief scaffold: python3 scripts/research_brief.py --brand Acme --url acme.com --vertical "PM SaaS" --geo US --goal leads This emits five sections (markdown by default, --json for pipelines): a prioritized query list, an ICP/persona skeleton, a demand/SERP checklist, audience-signal hypotheses, and offer/angle hypotheses.
  3. Run the P1 queries first. They carry the highest decision value: confirm commercial-intent demand exists, capture how buyers phrase searches, and read the advertiser's current reputation. Then P2 (competitor and SERP scan, audience mining) and P3 (pricing, objection, seasonality color).
  4. Read the actual SERP for the money terms. Note advertiser density (predicts CPC pressure and how strong your USP must be), SERP features that compress paid space, and the angles competitors have already claimed so you can take an open one. See references/research-process.md.
  5. Fill the ICP/persona rows in the buyer's own language. Ad copy that mirrors searcher phrasing wins Quality Score and CTR. Identify which segment is worth a dedicated ad group, landing page, and higher bid.
  6. Map audience-signal hypotheses to Google audience types using references/audience-signals.md: in-market, custom segments, affinity, life events, your-data/RLSA, and Customer Match. Record campaign role (prospecting, RLSA bid-up, exclusion) and observation vs targeting per signal.
  7. Fold findings back in to fill the brief: python3 scripts/research_brief.py --vertical "PM SaaS" --findings findings.json The findings JSON accepts competitors, personas, demand notes, audience signals, and offers, replacing the TODO placeholders.
  8. Output the offer/angle test backlog. Never launch with a single untested angle; the hypotheses become the RSA and landing-page test plan.

Inputs and data sources

Runs standalone -- paste findings or hand-fill the brief; no network or API required. Optional sources sharpen it:

  • Google Ads Keyword Planner: real search volume and competition.
  • Google Trends: seasonality and rising queries.
  • Search Console: existing query demand for the domain.
  • GA4 / analytics: which segments actually convert.
  • Google Sheets / BigQuery: store and version the brief and findings.

The --findings JSON is the bridge: capture research output in that shape and the script renders a partly filled brief.

Outputs

  • A structured research brief (markdown or JSON) with five sections: prioritized query list, ICP/persona table, demand/SERP checklist, audience-signal hypotheses mapped to Google audience types, and offer/angle test hypotheses.
  • A prioritized search agenda (P1/P2/P3) to execute.
  • A handoff-ready foundation for keyword research, campaign structure, and ad copy.

References

  • research-process.md: the five-layer, decision-first research sequence and how each layer changes a campaign decision.
  • audience-signals.md: every Google audience type (in-market, custom segments, affinity, life events, your-data/RLSA, Customer Match), what it captures, and how to map research signals to it.

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.