Keyword research when half your competitors rank for nothing

Most keyword research advice assumes a crowded market — that the obvious terms are taken and the work is finding the gaps nobody has spotted. In the sectors we measure, that assumption is wrong.

636 of 1371 businesses rank for nothing at all. Not “rank poorly” — absent from the top 100 for every keyword in their sector’s basket. The median Visibility score across all 1371 is 4 out of 100.

That is not a competitive market with gaps in it. It is a market where most participants are not competing.

What that changes

If you are doing keyword research for a local business, the useful question is not “which long-tail term can I find that nobody has taken”. Almost none of them are taken. The useful question is “which terms do my customers actually use, and am I on the page at all for any of them?”

Concretely, that inverts the usual order of work:

1. Check whether you rank for your own core terms first. Not variations, not long tails — the two or three phrases someone types when they want exactly what you sell. 1275 of 1371 businesses score below 50 on this pillar, which mostly means those core terms are not covered.
2. Then check whether your competitors do. In most of these cohorts, several do not. That is your gap, and it is much larger than any long-tail opportunity.
3. Only then look for untapped terms. This is where conventional advice starts, and for a local business it is usually step three, not step one.

How we measure it, and what that misses

Each index defines a basket of keywords for its sector and town — the terms a customer would plausibly search. We check the top 100 organic results for each, plus local pack appearances, and score the share a business ranks for, weighted by average position.

Two honest limits. The basket is a fixed, human-chosen set, so a business ranking well for terms outside it scores lower than its real visibility. And ranking is measured at one point in time from one location, so personalised or highly local variation is invisible to us. The methodology states the basket construction and the datasets publish every position we recorded.

What actually moves it

In order of how often it is the binding constraint for the businesses we measure:

Having a page for the thing at all. A surprising share of local sites have one services page listing eight services. If someone searches for one of those eight, there is no page competing for it — only a paragraph. That is the single most common cause of a zero.

Naming the town. Local intent is explicit. A page about your service that never says where you do it is competing nationally, which you will lose.

Coverage across a topic rather than a term. A page supported by related pages that link to each other states a subject; one page competing alone states a keyword. This is what topical authority means in practice, and it is also what the internal linking analysis found separates the businesses that rank from the ones that do not.

Being indexable. Worth ruling out before anything else — see indexing and crawling. A page that cannot be crawled cannot rank for anything, and it looks identical to a keyword problem from the outside.

Where your own sector sits

Every published index reports this pillar per business against its sector median: browse the indices. If you appear in one, your scorecard names the keywords you rank for and the ones you do not, which is a faster answer than any research tool.

---

Figures recomputed at build time from the published datasets, 1371 businesses across the South West.