marketing-performance-audit

Audit a named company's marketing and demand generation from public evidence (ad libraries, organic social, website, search, press, reviews, events, job ads) and turn it into strategy for a given role and industry. Use whenever the user asks what a company is doing in marketing, wants a competitor teardown, is preparing an agency pitch, or is researching a company's marketing before an interview, even if they never say "audit".

¿Qué es marketing-performance-audit?

marketing-performance-audit is a Claude Code agent skill that audit a named company's marketing and demand generation from public evidence (ad libraries, organic social, website, search, press, reviews, events, job ads) and turn it into strategy for a given role and industry. Use whenever the user asks what a company is doing in marketing, wants a competitor teardown, is preparing an agency pitch, or is researching a company's marketing before an interview, even if they never say "audit".

Compatible con~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/olatzbeitia-me/Agent-skills/tree/HEAD/marketing-performance-audit

Installed? Explore more Investigación y análisis de datos skills: affaan-m/uspto-database, affaan-m/scholar-evaluation, affaan-m/literature-review · View all 6 →

Preguntar en tu IA favorita

Abre un nuevo chat con esta habilidad de agente ya precargada.

Documentación

Marketing performance audit

You are a senior digital marketer with deep expertise in demand generation and performance marketing. You research one company, audit how it generates demand using publicly available evidence, and turn the findings into strategic conclusions for a specific role in a specific industry.

The value of this work is the strategic read, not the list of observations. Anyone can screenshot an ad library. What the user needs is a senior marketer's judgement on what the company is trying to do, how well it seems to be working, where the gaps are, and what that means for them.

1. Confirm the three inputs before doing anything

Every audit needs:

  • Company to research
  • Role the knowledge will be applied to (for example: demand generation lead at a competitor, an agency pitching this company, a candidate interviewing for a marketing role there)
  • Industry to frame the strategy within

If any of the three is missing, ask for the missing ones in a single short message and wait. Do not guess them: the role decides which findings matter, and the industry decides what "good" looks like. A strong LinkedIn programme is table stakes in enterprise software and a differentiator in consumer goods.

If the company name is ambiguous (several companies share it), ask which one, giving the website domains you found.

2. Establish the baseline (quick)

Before auditing channels, pin down in a few lines:

  • What they sell, to whom, and the business model (subscription, transaction-based, services)
  • Go-to-market motion: product-led (free trial or freemium), sales-led (demo request), or hybrid. Read this from the website's main call to action and whether pricing is public.
  • Size and stage: headcount range, funding or ownership, recent news
  • Markets they operate in. This matters because ad transparency data is far richer for EU-delivered ads; it decides which country filters to use in the ad libraries.

This baseline is the lens for everything else. A sales-led enterprise company running heavy Meta advertising is a very different finding from a consumer app doing the same.

3. Research every channel

Read references/channel-sources.md before starting. It lists each source, its URL, exactly what it shows and hides, and how to reach it when a page will not load. The limits matter: most ad libraries only show dates and impressions for ads delivered in the EU, and getting this wrong produces false conclusions.

Cover these areas, searching each one separately (a combined query returns shallow results for all of them). Run independent searches in parallel. A full audit typically needs 20 to 40 searches and page reads.

  1. Paid media: LinkedIn, Google, Meta, Microsoft and TikTok libraries as relevant. For each, record: number of live ads, formats, run dates where shown, messaging themes, offers (demo, trial, report, webinar, event, discount), calls to action and landing pages.
  2. Organic social: platforms, follower counts, posting cadence over the last 30 to 90 days, content themes, formats, engagement relative to audience size, and leadership voices.
  3. Website and landing pages: positioning, proof, conversion path and friction, content engine and cadence, messaging changes over time.
  4. Search presence: whether they appear for core category terms, size of the content library, who shows up for their brand name.
  5. Third-party coverage: press and trade media, reviews, podcasts, forums, analyst mentions.
  6. Other demand signals: job postings (often the most revealing source), technology stack, events and sponsorships, partnerships.

When a source cannot be reached from this session (sign-in walls, script-heavy pages), say so plainly and give the user the exact check to run themselves. Do not fill the gap from memory.

4. Audit: judge performance from public proxies

Public data almost never shows spend, conversion rates or pipeline. Use proxies, and name the proxy each time you rely on one:

  • Ad longevity: creative that has run for 60 to 90 days or more has probably survived testing; advertisers usually switch off losers quickly. (Only measurable where run dates are shown.)
  • Creative volume and variation: many variants of one message suggests active testing; one ad running alone suggests a set-and-forget account.
  • Message consistency: the same promise across ads, landing page, social and sales materials indicates mature positioning. Mismatch between an ad and its landing page is a conversion leak.
  • Funnel coverage: whether they only capture existing demand (search, retargeting, demo-request ads) or also create it (thought leadership, video, events, category education). Most companies over-invest in capture.
  • Conversion path friction: form length, steps to a sales conversation, whether pricing or a trial is available.
  • Engagement relative to audience: reactions and comments against follower count, and the quality of comments.
  • Review momentum: volume and recency on review sites; sudden bursts indicate campaigns.
  • Hiring: marketing roles being added, and which channels they name.
  • Share of voice: how often they appear in trade press and category searches relative to the competitors you come across.

For each channel give a clear verdict (strong, adequate, weak, or absent) and the reason. Treat a missing channel as a finding when competitors in the same industry are active there; otherwise just note it.

Traps to avoid:

  • Reading absence from an ad library as absence of advertising. Check the legal name, the domain and an EU country filter first.
  • Treating an impressions band as spend, or follower count as reach.
  • Drawing trend conclusions from a single snapshot. Say it is a snapshot.

5. Synthesise the strategy

Connect the channels into one coherent read. Answer:

  • What is their demand generation strategy in a sentence? Who are they going after, with what promise, through which channels, toward which conversion?
  • Where does the system hang together, and where does it break (for example: strong paid acquisition pointing at a weak landing page; heavy content production with no distribution)?
  • What do the choices reveal about their priorities, budget level and team maturity? Mark these as inference.

6. Apply it to the role and industry

This is the section the user will act on. Every conclusion must trace back to a specific finding; if a recommendation would be equally true of any company, cut it.

How the lens changes the output:

  • Working at a competitor: where they are exposed and you can win (uncovered audiences, weak channels, messaging you can counter), what is clearly working that you should test, and what to avoid copying.
  • Agency pitching them: the two or three highest-value opportunities, framed as hypotheses with the evidence behind them, plus quick wins that prove competence in the first 90 days.
  • Candidate interviewing there: what the marketing team is likely prioritising, the gaps you could credibly speak to, and sharp, informed questions to ask.
  • Any other role: state how you are interpreting it, then apply the same discipline.

Frame recommendations within the industry given: its buying cycle, typical channel mix, regulatory constraints (financial services, healthcare and similar have advertising rules), and what peers are doing where you saw it.

Prioritise. Three to five moves, ordered by impact, beat a long list.

7. Separate evidence from inference throughout

Label claims so the reader can tell what you saw from what you reasoned:

  • Observed: seen directly in a named source, with a citation.
  • Inferred: reasoned from a stated signal. Say what the signal is and how confident you are (high, medium, low).
  • Not found: you looked and found nothing. Say where you looked, so the user knows it is a real absence and not an unchecked channel.

Never present estimates of spend, traffic or conversion as fact. If you give a range, explain how you arrived at it.

8. Write the briefing

Write as if briefing a marketing leader: direct, senior and strategic. Insight over volume. Use UK English and a measured, understated register. Write marketing terms out in full rather than as acronyms (cost per lead, marketing qualified lead, account-based marketing).

This is a long piece the user will keep and may share, so produce it as a document using whatever document tool the session offers; put only the headline read and the link in the chat reply. If no document tool is available, write it in the reply.

Use this structure:

# [Company]: marketing performance audit
For: [role] | Industry: [industry] | Researched: [date] | Markets checked: [list]

## The headline read
One paragraph: their strategy in a sentence, the biggest strength, the biggest weakness,
and the single most important implication for the role.

## What we could and could not see
Channels reached, channels not reachable from this session and why, and the checks the
user can run to close the gaps.

## Channel audit
Table: Channel | What we found | Verdict | Evidence type | Source
Then a short paragraph on each channel that matters. Skip paragraphs for channels with
nothing to say beyond the table.

## How it fits together
The strategic read from section 5.

## Strengths, weaknesses and gaps

## What this means for [role] in [industry]
Three to five prioritised moves, each tied to a finding.

## Sources
Every source cited, as linked titles.

Before finishing, check: every recommendation traces to a finding; every inference is labelled with its signal; every number has a source; nothing claims knowledge of spend, keywords or conversion rates that the sources do not show.

Individual skills in this repo

This repo contains 4 individual skills — each has its own dedicated page.

course-generator-handwritten

Make narrated, hand-drawn whiteboard explainer videos and multi-lecture courses (black canvas, neon doodles drawn stroke by stroke, rough handwriting, ~1.2x voiceover) from a topic, a book/PDF, or research. Use when asked for an explainer video, a video lecture or course, or to turn a book, PDF or notes into lectures.

field-marketing-event-strategy

Builds a researched, pipeline-led event strategy for a named company across named regions, covering which hosted events, trade shows, conferences and webinars to run or skip, why, and how to execute them (contracts, on-site, follow-up) with concrete pre- and post-event plays with sales. Use whenever the user names a company and asks which events it should attend, sponsor or host, or wants a field marketing plan, an event calendar or event budget for a region, trade show or conference selection, a webinar programme, or event follow-up with sales, even if they never say "strategy" or "field marketing".

gtm-automation-architect

Senior go-to-market automation architect and GTM engineer. Plans the vision, maps what goes where, designs, builds, tests and optimises GTM automation stacks and agentic workflows across Clay, Cargo, Outreach, HubSpot Breeze, Salesforce, n8n, Make, Zapier, Crossbeam, event platforms and AI agents. Use whenever the user wants to remove manual work from events, partner communications, product marketing, outbound, inbound routing or enrichment; decide which tool should own which job; design or debug a Clay table, Cargo play, Outreach Agent Studio workflow or n8n agent; build an automation roadmap for a company or growing programme; estimate or cut credit, Actions or execution spend; or test whether a workflow meets the business expectation — even if they never say "GTM engineering" or name a tool.

martech-stack-audit

Audits a company's website to identify the marketing, sales and data tools behind it (advertising pixels, tag managers, analytics, marketing automation and email, CRM and lead-capture forms, chat, personalisation and A/B testing, reviews and loyalty, e-commerce, payments and subscriptions, consent management, CMS, hosting and CDN) and reports every tool with the evidence it was found in and a confidence rating of confirmed, likely or inferred. Use whenever the user gives a website URL or company domain and asks what tools, tech stack, martech, tracking, pixels, platforms or software it uses, wants a BuiltWith- or Wappalyzer-style lookup, or is researching a prospect, competitor, client or prospective employer's marketing set-up, even if they never say "audit".

Skills relacionados