TomHow / Search visibility
Keyword research that starts with what you already rank for
Most keyword research starts with a tool and a blank page. That skips a useful first-party starting point, which is the list of searches your site already appears for.
Reviewed
The overview
What this work covers.
Keyword research maps buyer searches to pages with a clear purpose. TomHow starts with available Search Console queries, then uses competitor pages and tool estimates to explore gaps. The output links each priority to an existing or planned page, with its intent, evidence and uncertainty stated. This work is part of the retainer. A new site without search history starts with a testable plan, not a forecast presented as fact.
01 / First-party demand
Your Search Console data is the best keyword tool you own.
Every site with any history has a list of searches it already surfaces for. Grouping those by position helps identify pages worth investigating. A term at position twelve with real impressions is a different proposition from a term at position eighty, and no third-party tool can tell you that about your own site.
- Every query the site has surfaced for, over the full sixteen months Search Console retains
- Position bands, so page-two terms surface as the priority they are
- Queries with impressions but few clicks, reviewed for position, intent and result appearance
- Terms the site ranks for by accident, which often reveal a market you had not considered
02 / Intent, not volume
High volume with the wrong intent is worse than useless.
A term with strong volume that brings people looking for a free tool when you sell a retainer costs you attention and converts nothing. Every candidate term gets classified by what the searcher is trying to do before it is allowed near a content plan.
- Commercial, comparison, informational and navigational intent, classified per term
- Whether the current results are dominated by publishers, tools or providers
- Whether a smaller site can realistically hold position, given who is already there
- Buyer prompts for AI answer systems, which phrase differently from typed searches
03 / Mapping to pages
One page per distinct question, and no orphans.
Research that ends in a spreadsheet has not finished. Every term that survives the filter gets attached to a specific page, either one that exists or one that needs building, and every planned page gets internal links from somewhere that already has authority.
- A term-to-page map covering existing pages and gaps
- Terms merged where they share intent, so two pages do not compete with each other
- Internal linking routes for anything new, so it is not published into isolation
- A deliberate not-targeting list, with reasons
Keep exploring
Find the right next step.
Before you decide
Straight answers.
Can I buy keyword research on its own?
No. It is part of the TomHow retainer. A keyword list with no pages built against it is one of the most common pieces of unused SEO output there is.
Do you use Semrush or Ahrefs volume figures?
They are used to size opportunities, not to find them. Both tools estimate, and both are regularly wrong for lower-volume commercial terms. Your own Search Console impression data is measured rather than modelled, so it leads.
What about a brand new site with no Search Console history?
Then tool data and competitor page analysis carry more weight, and the first plan is treated as a hypothesis to be corrected once real query data starts arriving. The timing depends on when useful query data becomes available.
How many keywords do you target?
Fewer than most plans propose. The plan prioritises useful pages and distinct buyer needs rather than a target keyword count. Progress is judged from the evidence, not the size of the spreadsheet.