Keyword difficulty is a provider-specific metric that summarises the competitive strength of the pages currently ranking for a query. It is expressed on a 0-to-100 scale where higher values indicate stronger incumbents and, therefore, more effort required to displace them.
How it is calculated
Most providers derive the score from the referring-domain profiles of the URLs ranking on page one. Additional signals can include the average domain rating of the top ten, the topical relevance of ranking pages, the presence of well-known brands, and the density of SERP features that push organic results below the fold.
Example
The query "buy running shoes" typically returns a difficulty score above 80 because the top ten is dominated by high-authority retailers with tens of thousands of referring domains. A long-tail query such as "how to lace running shoes for wide feet" often scores below 20 because the ranking pages have thinner backlink profiles.
How SEM Optimiser reports it
SEM Optimiser labels every difficulty score with the source provider so scores from different vendors are never mixed in the same comparison. Missing scores are shown as null rather than substituted with zero.
- Difficulty is a relative estimate, not a guarantee.
- Two providers can return different scores for the same query — this is expected.
- Difficulty should be paired with search intent before deciding what to target.
Related terms
- Search Volume
Search volume is the average number of times a specific query is searched per month in a given country and language.
- Keyword Intent
Keyword intent is the classification of a search query by what the user is trying to accomplish, typically informational, navigational, commercial, or transactional.
- Domain Rating
Domain Rating is a proprietary Ahrefs metric on a 0-to-100 scale that measures the strength of a domain's backlink profile.
- SERP Feature
A SERP feature is any non-standard element on a search results page beyond the classic ten blue links, such as a featured snippet, People Also Ask box, or local pack.