Keyword Difficulty: What the Score Measures and What It Misses

The short answer

Two tools can give the same keyword different scores, and neither is wrong: each calculates difficulty its own way [1]. What matters is using one tool consistently, then checking the actual page-one results yourself.

How Ahrefs calculates KD

In Ahrefs' words, KD "estimates how hard it would be to rank on the first page of Google" on a 0 to 100 scale [1]. Its help center adds that the score uses the top ten organic results only and "does not take into consideration any 'on-page SEO' factors" [2].

The rule of thumb behind the number is the one that matters most. To reach the top ten, Ahrefs says you will need about as many backlinks as the pages already there, "if not more" [1]. Matching them is no promise of the top result, though Ahrefs calls a place somewhere in the top ten "a very good chance" [1]. Ahrefs is also candid about the limit: every tool uses the same 0 to 100 scale but calculates it differently, and a metric that accurately predicts your chances "doesn't exist" [1].

How Semrush calculates KD%

The "authority" part deserves a caution. Google says: "some third-party services provide "reputation" or "authority" scores for sites. These don't correspond to any of Google's own signals nor come from Google." [4] Authority Score is Semrush's measure, useful for comparing sites inside Semrush, not a number Google uses.

What difficulty misses: your site, the map pack, AI answers

• Your site. Ahrefs says the score shows how hard it is for the average website to rank, not yours [1]. A site with strong links and relevant pages can take on phrases a new site can't.
• The map pack. Google says local results are mainly based on relevance, distance and popularity, with links and reviews feeding prominence [5], and that the categories you choose affect local ranking [6]. None of those inputs is in a KD score.
• AI answers. Google says AI Overviews often don't appear [7]. When they do, studies disagree on how often they cite top-ten pages: 16.7% of citations (BrightEdge, September 2025) [8], 76% of cited pages (Ahrefs, July 2025) [9] and about 38% (Ahrefs, March 2026) [10].
• Volume. Difficulty is read next to search volume, and both are estimates. Google Ads averages volume over 12 months by default, groups close variants and rounds the figures [11]. Most keywords are small: almost 93% of those in Ahrefs' US database get fewer than 10 searches a month [12].

How to find low competition keywords: the weakest page on page one

To reach page one you don't have to beat the leader; you have to beat the last result on the page. For each ranking page, record:

1. Intent and format. Is it a service page, a guide, a directory, a list? Match what Google already shows.
2. Topics covered. Google says it looks for "other relevant content beyond the keyword" [13], so list the subtopics each result covers.
3. Referring domains to that page. Count the sites linking to the ranking URL itself, the input both tools use [1][3].
4. Quality. Run it through Google's own helpful-content questions, such as whether it offers "insightful analysis or interesting information that is beyond the obvious" [14].

Links matter, but Google says "We have many ranking signals, and PageRank is just one of those." [16] A page that answers the search better can pass one with more links. This page-by-page look is the core of SEO competitor analysis and a content gap analysis.

Using difficulty to order the work

Ahrefs' timing study is a useful baseline: only 1.74% of random new pages reached the top ten within a year, and high-volume keywords took longer [15]. Those were pages with no targeted work behind them, so the figure shows that page one is not automatic, not what a planned campaign will do. The plan that follows is simple: work the smallest gaps first, in waves, and let links and reviews build toward the harder phrases. Local SEO keyword research covers building the list, and how long SEO takes to work covers the timing.

This is also the arithmetic behind our first page guarantee. It is offered after an assessment, and only if you qualify, and the share we promise is set from the measured gaps, with margin built into the schedule. The share is measured separately in each of the three places: the map results, the organic results and Google's AI answer. The first page math is set out in full in how an SEO guarantee works.

Measure first, then promise

Keyword difficulty is a starting estimate; the real measurement is the page-one competition for each phrase, in each place. That measurement is what our first page guarantee rests on: read the SEO guarantee explained, see the organic SEO guide for the work itself, and get a free SEO audit to see where you stand today.