B2B keyword research fails when it is run with B2C instincts. Sort by volume, apply a minimum threshold, and you will delete every term that would have generated revenue while keeping the ones that generate traffic.
The arithmetic is genuinely different, and it is worth doing explicitly before you set any thresholds.
The volume trap
| Keyword | Volume/mo | Realistic conversion | Deal value | Monthly value at pos 3 |
|---|---|---|---|---|
what is crm software | 12,000 | 0.1% | £4,000 | ~£480 |
nhs procurement crm requirements | 40 | 4% | £4,000 | ~£160 |
crm for housing associations | 90 | 5% | £4,000 | ~£450 |
Illustrative figures — the pattern is the point, not the numbers.
The 12,000-volume term produces roughly the same revenue as a 90-volume term, at maybe fifty times the difficulty. That is the entire case for B2B keyword research being a different discipline: your minimum volume threshold should often be zero.
A term with 20 searches a month, where five of those searches are procurement leads at organisations that match your ICP, is a better target than anything in the head of your market. It is also usually KD 0, which means you can have it this quarter rather than in three years.
Where the good B2B keywords come from
Tools are poor at B2B because low-volume terms fall below the reporting floor. A keyword with 30 real searches often shows as 0 or does not appear at all. If your research is tool-only, you are researching the 5% of your market the tool can see.
Sales call recordings. The single best source. How prospects describe the problem before they know your category name. This language rarely appears in keyword tools and almost always converts, because someone using it has the problem you solve.
Your own Search Console, filtered to 11–30. Queries you already earn impressions for but do not rank well enough to click. In B2B these are frequently long, specific and unqueryable in any tool.
Lost-deal reasons and RFP documents. The requirements language buyers use is the language they search with. “SOC 2 compliant” and “single sign-on support” are keywords in B2B in a way they are not elsewhere.
Support tickets and onboarding questions. What customers ask after buying is what prospects search before buying.
Competitor comparison pages. Not to copy, but to find the alternatives-and-versus space, which is where the highest-intent B2B traffic lives.
The buying committee changes what you target
B2C SEO targets one person with one intent. B2B targets four or five people with different questions about the same purchase, at different times, and any of them can stop the deal.
| Role | What they search | Content that serves it |
|---|---|---|
| End user | ”how to do X faster” | Practical how-to, comparison |
| Manager / champion | ”X software for [industry]“ | Category page, use case |
| Finance | ”X pricing”, “cost of X” | Transparent pricing, ROI |
| IT / security | ”X SOC 2”, “X SSO”, “X API” | Technical documentation |
| Procurement | ”X vs Y”, “X alternatives” | Honest comparison |
The rows most B2B sites skip entirely are IT and procurement — and those are the two that kill deals late, after significant sales cost has been spent. A security-questionnaire page ranking for compliance terms has almost no search volume and removes a common blocker at the point it appears.
Map your keyword list across these roles. Gaps in the table are gaps in the funnel, not just gaps in traffic.
Thresholds that work for B2B
Adapt the standard keyword research checklist as follows:
- Drop the volume floor to zero. Replace it with a relevance test: would a searcher using this phrase plausibly be in your ICP?
- Keep the domain-strength checks. These do not change. Low volume does not mean low difficulty — some B2B terms are contested by very strong vendors.
- Weight commercial proximity heavily. In B2B the distance between the query and the purchase matters more than volume.
- Accept longer feedback loops. With 40 searches a month you cannot A/B your way to insight. Judge on pipeline over two quarters, not clicks over two weeks.
That last point is the hardest one organisationally. A page that produces four visits a month and one qualified opportunity looks like a failure in any traffic dashboard. Report B2B SEO on pipeline influenced, not sessions, or the work gets cut for looking unproductive while producing revenue.
Where B2B keyword research usually goes wrong
Writing for the category term. Every vendor targets the head term and the SERP is owned by review sites and analysts. You are competing with G2 and Gartner, not with other vendors.
Ignoring the unbranded problem language. Buyers search the problem long before the category. “Spreadsheet version control nightmare” precedes “document management software” by months.
Treating comparison keywords as beneath you. “X vs Y” and “X alternatives” are the highest-intent terms in B2B and vendors avoid them out of squeamishness about naming competitors. That squeamishness is why they are winnable.
Publishing without a destination. Same failure as anywhere else — see what to do after keyword research for the mapping step that prevents it.
Adjacent contexts
If your B2B product is software, the alternatives, integrations and jobs-to-be-done patterns are specific enough to be worth their own treatment — see keyword research for SaaS. If you sell across several European markets, the same term behaves differently per market and translation is not research: international keyword research covers that.
For the underlying method, the keyword research guide and long-tail keywords both apply — B2B is close to being entirely a long-tail discipline.
If you want this run against your own pipeline data rather than a tool export, our fractional SEO consultant engagement is usually the right shape for it — B2B keyword work needs access to sales calls and deal data, which is a different relationship from a delivery retainer.
