Keyword research: 20 to 50 article candidates per pillar
Pick a pillar from your content map. Or skip the pillar and type any topic. Pillar mode anchors the run to your strategy. Free-form handles ad-hoc research, competitor lookups, long-tail questions.
What I do, behind the scenes
1. Read your brand strategy. Your audience, expertise, opinions, refusals, the country you rank in.
2. Pick the scope. A pillar from your content map, or any topic you type in.
3. Score your authority. New, growing, established, or mature. Sets the ceiling on what you can realistically rank for. Most blogs lose months chasing keywords above their authority. I keep you under the ceiling.
4. Wide discovery. I pull keyword candidates from five places at once:
- What people actually search (volume data on real queries)
- What you already rank for, and what your real SEO competitors rank for that you don't
- Live Google results, People-Also-Ask, related searches
- Real questions from Reddit and Quora
- ChatGPT-style conversational queries (because that's how Perplexity and ChatGPT get asked)
Everything tied to your country and language. No global averages.
5. Filter for your brand. I drop anything that doesn't fit your audience or hits your refuse list.
6. Pull volume and difficulty for the survivors.
7. Drop what you can't win. Anything above your authority ceiling is out. If zero candidates survive, I tell you straight: your authority isn't there yet for this pillar.
8. Check live Google. For the highest-volume survivors I read who's ranking right now, and whether Google is showing an AI summary at the top of the page.
9. Group into articles. One primary keyword per article. Variations grouped underneath. Two articles fighting for the same Google result get merged so you don't cannibalize yourself.
10. The four verdicts. Every article candidate gets one:
- Transactional. Write it. Humans buy.
- Defensible. Write it. You can win on depth, opinion, or original data.
- Defensible with AI summary. Write it, but pivot the angle so you get quoted inside the AI summary instead of clicked from below it.
- Non-defensible. Don't write it. I put it in the "Don't write this" tab so you learn what NOT to chase.
Every keeper needs an angle that cites something only YOU know: your case studies, your data, your strong opinions. If it doesn't, it's non-defensible.
What you get back
A three-tab document.
- Strategy. The narrative, the pillar map, and the breakdown of how many candidates landed in each verdict.
- Write this. 20 to 50 article candidates with primary keyword, variations, verdict, angle, volume, difficulty, and the AI-summary flag.
- Don't write this. The on-topic-but-non-defensible candidates with the reason. The most valuable tab over time.
About 5 minutes per pillar.
