Keyword Research With Free Tools: Search Console + an AI Assistant

Paid SEO suites are worth their money — eventually. But we've watched too many small sites pay $140 a month to research keywords they don't yet have the authority to rank for. Before that invoice, there's a free workflow that covers more ground than most people expect. It's the one our under-€100 stack prescribes for month one, and it runs on two tools you already have.

Part 1: Mine Search Console for near-wins

Google Search Console is the only tool that knows what your site can rank for, because it's reporting what Google already does with it.

Open Performance → Search results, set the date range to three months, and sort by impressions. Now look for the specific gold: queries with high impressions, low clicks, and an average position between 8 and 20. These are your near-wins — pages Google already considers relevant, sitting just below where anyone clicks.

For each near-win you have three moves, in order of effort:

1. Retitle — if the query and your title don't quite match, align them.
2. Extend — add the section that actually answers the query (search the query yourself; the top results show you what's expected).
3. Interlink — point two or three related posts at the page with the query as anchor text.

Moving position 12 to position 6 on a page that already has impressions beats publishing something new, every time. It's free traffic you've already half-earned.

Part 2: Let an AI assistant do the clustering

The classic paid-tool jobs — grouping keywords, guessing intent, drafting outlines — are exactly what a $20 AI assistant does well. Three prompts we actually use:

Expansion: "Here are 15 queries my site gets impressions for [paste from Search Console]. Generate 30 closely related long-tail questions a solo marketer would search, grouped by intent (learn / compare / buy)."

Clustering: "Group these keywords into content clusters, and for each cluster tell me whether it needs one comprehensive page or several narrow ones."

The reality check: "For each cluster, what would realistically outrank a small site here — and which clusters have a 'small site angle' a big publisher can't take?" That last question is the whole game for sites like ours: the qualifier ("for solo marketers," "not enterprise," "under €100") is what makes a keyword winnable.

What the assistant can't give you is search volume or competitor data — it will guess, confidently. Treat its numbers as fiction and its language as signal.

Part 3: Know when free stops being enough

This workflow has a ceiling, and honesty says so. You've hit it when:

• decisions start depending on numbers (real volumes, competitor traffic) rather than judgment,
• you need to know who links to competitors and how you'd catch up,
• you're publishing consistently and want every draft scored against what ranks.

The first two are a research-suite problem, the third is a content-optimizer problem — and they're different purchases at very different prices. That fork is exactly what our Surfer vs Semrush comparison is for: most small sites need the $49 content tool before the $140 research suite.

One more modern wrinkle: a growing share of "best tool" searches now happens inside ChatGPT, Claude and Perplexity — and those systems cite pages with clear structure, tables and dated facts, not pages with the best keyword density. The free workflow above plus disciplined structure gets you into both games at once.