Skip to content
Amsby

AI agents & automation

SEO audits and competitor keyword research, automated from one form.

Amsby built an n8n and Google Gemini audit agent plus a keyword pipeline that delivers a prioritised SEO action list and ranked keyword opportunities in Google Sheets.

Client
DataClicks (dataclicks.nl, opens in a new tab)
Sector
SEO & SEA agency, Naaldwijk (Westland)
Type
AI audit agent and data pipeline
Engagement
04 · Project-based

About DataClicks

DataClicks is a search marketing agency in the Westland region, run by its founder and active since 2014. It covers SEO audits, SEO advice and training, migrations, outsourced SEO, GEO (visibility in AI answer engines such as ChatGPT, Perplexity and Gemini) and SEA.

The problem

SEO audits are one of an agency’s main deliverables, and also one of the most labour-intensive. A consultant has to pull data from several tools (Lighthouse/PageSpeed, Google Search Console, a backlink tool), interpret it, translate it into concrete actions, decide who should fix each issue, and present it all in a form the client understands. Competitor keyword research takes even more manual work: exporting, cleaning, matching and ranking thousands of keywords.

What the agent does

Phase 1: AI SEO audit generator

The consultant enters a website URL, and the agent:

  1. Collects data from four sources in parallel.
    • Google Lighthouse (SEO): SEO score, title, meta description, HTTP status, robots.txt, crawlability, link text, image alts and hreflang.
    • Google Lighthouse (Performance): performance score, Core Web Vitals (LCP, CLS, TBT, FCP, Speed Index, TTI), real-user CrUX data including INP, and concrete opportunities such as render-blocking resources, image delivery, caching, server response time and page weight.
    • SE Ranking: a summary of the backlink profile.
    • Google Search Console: search-performance data by query, page, country and device.
  2. Turns raw data into an action plan. A Gemini-powered AI agent converts the combined technical data into a structured list of findings in Dutch. Each finding has a topic (Onderwerp) and finding (Bevinding), a concrete, actionable fix (Oplossing), a priority of High, Medium or Low (Prioriteit), an owner (Verantwoordelijk: the agency for analysis, the web developer for technical changes or the client for business and content decisions), a category (website audit, competitor research, keyword research or site structure) and a status field (Opgelost?), so the sheet works as a live tracker.
  3. Delivers it in Google Sheets. The list is ready to share with the client and to track through to resolution.

Phase 2: Competitor keyword-gap research

The consultant enters the client’s domain, and the pipeline:

  1. Finds the client’s project in SE Ranking and retrieves all tracked competitors.
  2. Loops through each competitor, with rate limiting, to get its top 100 keywords by traffic, with positions and URLs.
  3. Pulls the client’s own ranking queries from Google Search Console: clicks, impressions, CTR and position.
  4. Normalises and deduplicates keywords from both sources and merges them in a full outer join.
  5. Removes branded keywords automatically.
  6. Classifies every keyword as competitor-only (a content gap), both (head-to-head) or client-only.
  7. Scores and ranks each keyword with a priority algorithm that combines gap type, search volume and current position, so the biggest opportunities come out on top.

At a glance

data sources collected in parallel
4
phases (audit and keyword-gap research)
2
top keywords retrieved per competitor
100

Business value

  • Hours of audit work are reduced to one form submission, with data collection, interpretation and the action list automated.
  • The output is consistent, client-ready and in Dutch, with a clear owner and priority for every finding.
  • The result is a living tracker rather than a static PDF, because each finding has a resolution status.
  • Content strategy becomes data-driven, with competitor keyword gaps ranked by opportunity instead of guesswork.
  • The approach scales across the agency, with the same audit quality for every client, every time.

Technologies

  • n8n
  • Google Gemini (AI Agent)
  • Google Lighthouse / PageSpeed Insights API
  • Google Search Console API
  • SE Ranking API
  • Google Sheets API
  • JavaScript
  • Data normalisation and scoring algorithms

Hand repetitive work to an AI agent?

Built on your own data and tools, with a human approving anything important.