Keyword Research Productivity Tips: The 90-Minute Cap

Key takeaways

• Cap keyword research at 90 minutes per content cycle and batch it into one weekly session instead of researching per article.
• Skip keyword research entirely when Search Console shows no impressions, no community posts rank for your topic, and no one has asked the question anywhere.
• Apply a three-signal filter before opening any tool: real people asked the question, you have something competitors don't, and the intent matches a page you can ship.
• Mine Search Console queries where you rank below position 20 with growing impressions, since these come with your domain's credibility attached.
• Run each 90-minute session in four timed blocks: pull existing data, collect and filter, validate the shortlist, and write briefs before closing the sheet.

Keyword research productivity tips for solo founders boil down to one rule: cap research at 90 minutes per content cycle, batch all your queries into one weekly session, reuse queries you already rank for in Search Console, and apply a three-signal filter before opening any tool. If a topic fails the filter, skip research and write anyway.

keyword research productivity tips

The highest-leverage tip for a solo founder: cap research at 90 minutes per content cycle and spend the rest of that time writing. You shipped on Lovable or Cursor in a weekend, but keyword research is where momentum dies, because it expands to fill whatever time you give it.

Why research time scales badly for solo builders

Research feels productive, so it eats your evenings. You open a tool, pull 500 suggestions, export a spreadsheet, and ship nothing. The data itself is cheap: Google's Performance report already shows which queries give you impressions. The cost is the unstructured browsing around it.

The core move: research once per cycle, not per article

Batch it. Pick your keywords in one 90-minute session, then write from that list for two weeks without reopening a tool. Our competitor-driven keyword research guide breaks down the full workflow; the rest of this article shows each step.

When should you skip keyword research entirely before writing?

Skip keyword research when you have no existing content to mine, no Search Console data to reuse, and no evidence anyone searches for what you built. That is most zero-user founders on day one. Spending 90 minutes on a tool that returns empty volume estimates for a niche nobody queries yet is research theater, not distribution.

The skip test: three questions before opening a tool

Ask three questions. First, does Search Console already show impressions for anything on your domain? Second, do competitor or community posts rank for the topic you want to write? Third, will this article support a keyword you could realistically target later? If all three answers are no, write anyway, but write for a person, not a query.

When building in public beats keyword targeting

Before your first 100 paying users, some content has no search job at all. Build-in-public posts, launch notes, and changelogs exist to be shared where your target users hang out, not to rank. Write them without a keyword, ship them, and save your research time for pages that can compound. The 45-minute gap-to-draft workflow works best once you have a gap to find.

What three signals justify spending time on a keyword?

Most solo founders over-research keywords that will never convert. Before opening any tool, check three signals.

Signal 1: a real person asked it somewhere you can reach

The keyword mirrors a question someone actually posted on Reddit, Indie Hackers, or a niche Discord you can join. If no human has asked it anywhere, you are writing for a keyword list, not for users chasing their first 100 paying users.

Signal 2: you have something to say that competitors don't

If the top results already answer it fully and you have no build-in-public proof, unique data, or lived experience to add, skip it. Google's own guidance on how search systems reward helpful, people-first content makes this explicit: content needs to demonstrate first-hand expertise, not restate what exists.

Signal 3: the intent matches a page you can actually build

"Best Reddit marketing tool" fits a comparison page you can ship in a day. "GEO strategy" fits nothing you can credibly write yet. Match the keyword to a page type you can finish, then move on. For the broader system this filter plugs into, see increase productivity with AI: build a solo growth system.

How do you reuse Search Console queries to shortcut new keyword research?

If your site has any Google impressions at all, Search Console already did half your keyword research. Open Performance, filter to queries where you rank below position 20 but impressions are growing, and you have a list of topics Google already associates with your site. That beats a fresh tool export because these queries come with your domain's credibility attached.

Mining impressions you already have but don't rank for

Export queries with impressions but no clicks, sort by impression volume, and pick the ones matching your product's actual use case. A page that ranks 15 for a query is one internal link and a content refresh away from page one, which is cheaper than ranking a new URL from scratch. This is the fastest organic growth lever a zero-traffic site overlooks.

Turning one ranking page into a cluster of follow-ups

Any page that ranks for one query usually ranks for five to ten related ones. Check the query list under each URL, group them by intent, and write one follow-up piece per group. Those follow-ups become your roadmap for the cycle, and each one feeds your AI search visibility, since getting cited by ChatGPT starts with pages that already answer a real query. For why distribution, not more code, is the bottleneck at this stage, see AI search visibility is a launch problem, not a marketing one.

How do you batch keyword research so it happens once per cycle, not per article?

Keyword research per-article is the fastest way to burn your writing hours. Batch it: one sitting per content cycle, feeding several briefs at once. Timeboxing works because a fixed container forces you to stop researching and start writing.

The weekly research block: one sitting, five briefs

Pick one 90-minute block per cycle. Run the three-signal filter across a list of candidate queries, then output five content briefs in a single doc. Five briefs from one session means each article inherits research for free, and your cost per brief drops by roughly 80%.

Keeping a running backlog so the block starts warm

Keep a single doc where you dump questions from Reddit threads, Search Console queries, and AI search visibility checks all week. Your signals that prove visibility is moving belong in the same backlog. When the block starts, you filter a pre-built list instead of facing a blank page.

What does a 90-minute keyword research session look like step by step?

A 90-minute session runs in four blocks: pull existing data (0-15), collect and filter candidates (15-45), validate the shortlist (45-75), and write briefs (75-90). The timer matters more than the tools. If a block overruns, cut candidates, not validation.

Minutes 0-15: pull existing data

Open Search Console and export every query where your site already ranks, even at position 20-80. These are pre-validated topics Google already associates with your pages. Note impressions and current position in one sheet.

Minutes 15-45: collect and filter candidates

Run two or three seed queries through one keyword tool and dump everything into the same sheet. Now apply the three-signal filter from earlier sections: search volume, relevance to your product, and evidence of buying or problem intent. Kill anything failing two of three. Aim for 15-20 survivors.

Minutes 45-75: validate the shortlist

Search each surviving keyword manually. Check who ranks: if page one is all DR70+ sites, drop it. If you see Reddit threads, thin forums, or stale content, keep it. Record the top three URLs and one line on what they miss.

Minutes 75-90: write briefs and stop

For each validated keyword, write a five-line brief: target query, angle, the gap in current results, and one call to action. Then close the sheet and start writing. Reopening research mid-article is how the 90-minute cap dies.

How do you stop keyword research from becoming procrastination before writing?

Keyword research becomes procrastination the moment it feels safer than publishing. The fix is structural: cap the tools, cap the clock, and force a draft out of every session.

Hard stops, tool caps, and the one-draft rule

Set a hard stop: when the 90-minute timer ends, you open your editor, not another tab. Cap yourself to one research tool per cycle so you are not cross-checking five dashboards for the same answer. Then apply the one-draft rule: no session ends without a publishable draft, even a rough one. A keyword with no article attached has zero value to your first 100 paying users.

What to automate so research shrinks over time

Automate the repeating parts: pulling Search Console queries, tracking which keywords already rank, and drafting briefs from your three-signal filter. Each automated step returns minutes to writing, so every cycle should take less research time than the last. If your research time is flat or growing, you are collecting, not shipping.

Which tools fit a 90-minute solo workflow (and where they fall short)?

Honest criteria: time cost, data reuse, skip support

Judge any tool against the workflow, not feature lists: does it cut your research time toward 90 minutes, does it reuse data you already have (Search Console queries, existing rankings), and does it ever tell you to skip a keyword? Most tools fail the third test. They surface more keywords, not fewer, and more keywords is how research becomes procrastination.

Where HeyCatch fits and where it doesn't

HeyCatch is built for exactly your stage: bootstrapping builders going from zero to their first 100 paying users, not funded growth teams or agencies. It plans and executes across channels, so keyword research happens inside a daily growth loop rather than as a separate ritual, and it can recommend the skip when a keyword fails your three-signal filter. Pricing starts at $18.95 for the 2-week trial (reduced from $41.00), with longer plans at $28.95 and up, so trying it costs less than the hours it saves. Where it falls short: it won't replace Search Console as your source of truth for what already ranks, and if you want raw keyword volumes to sort yourself, a direct competitor may fit better. The honest test is the same for all of them: run one 90-minute cycle with the tool and one without, then count which cycle shipped more posts.

FAQ

How long should keyword research take for a single blog post?

Nothing per post. Research happens once per content cycle in a single 90-minute session that produces several briefs, so each article inherits its research for free. If you find yourself opening a keyword tool before writing one article, that per-article ritual is how research time expands to fill your evenings and kills your shipping momentum.

Can I do keyword research without paying for a tool as a bootstrapped founder?

Yes. Search Console's Performance report shows which queries already give your domain impressions, which beats a fresh tool export because those queries come with your site's credibility attached. Manually searching each candidate to check who ranks on page one replaces most validation features. A paid tool only helps once you have a backlog to filter.

Is it bad to write content without keyword research when I have zero traffic?

No, and it is often the right call. If Search Console shows no impressions, no community posts rank for your topic, and no one has asked the question anywhere, write for a person, not a query. Build-in-public posts and launch notes exist to be shared where your users hang out, not to rank.

How many keywords should I target in one research session?

Aim for 15 to 20 candidates surviving your three-signal filter, then validate down to 5 to 8 keywords and write a brief for each. Five briefs from one session means your cost per brief drops by roughly 80%. Anything beyond that is collection, not shipping, since keywords with no article attached have zero value.

What's the difference between SEO keyword research and GEO (getting cited by ChatGPT)?

SEO research targets queries people type into Google so your page ranks on page one. GEO targets getting your content cited inside AI answers, which starts with pages that already answer a real query clearly. The two compound: your Search Console follow-up clusters feed AI visibility, since models favor sources demonstrating first-hand expertise over restated summaries.