What is keyword difficulty in SEO? In the version most people use, it is a 0–100 score built from links. Ahrefs, which popularised the metric, describes its own calculation plainly: it looks at “the top 10 ranking pages for your keyword and look[s] up how many websites link to each of them.” That is the whole input. Content, intent and freshness are not part of it.
That is not a criticism of the metric. A link-based score is a useful first filter across thousands of keywords. The mistake is reading it as a forecast of your own chances, which it was never built to be.
What is keyword difficulty in SEO, exactly?
Every tool has its own formula, but the Ahrefs version is the clearest to reason about because the company publishes a rough mapping between the score and the links behind it. In its guide to keyword difficulty, Tim Soulo gives these reference points:
| Ahrefs KD | What the top-ranking pages typically have |
|---|---|
| 0–5 | Minimal backlinks |
| Around 50 | Several hundred backlinks |
| 90+ | Thousands of backlinks |
| Example: “best womens hiking boots”, KD 6 | About 7 referring domains to reach the top 10 |
Read that table as a translation, not a ranking guarantee. KD 6 means the pages in the top ten do not have many sites linking to them. It does not mean seven referring domains will put you there. The gap between the reference points also matters: the jump from a few backlinks at the bottom of the scale to several hundred around the middle is large, so a score in the middle of the scale already implies far more link work than one near the bottom of it.
Why two tools give one keyword different scores
Put the same keyword into two tools and you will often get two different numbers. That is expected. Ahrefs itself notes that different tools calculate keyword difficulty differently: they use different link indexes, weight the top results differently, and some blend in factors other than links.
The practical consequence is that scores are only comparable inside one tool. A 30 in one and a 30 in another are not the same claim. Pick one tool, learn what its scale means for the keywords you have already ranked for, and stop converting between them.
Scores also move. Link indexes are recrawled and the pages in the top ten change, so the number attached to a keyword today is a snapshot. If you built a shortlist months ago, refresh the scores before committing content to it, and expect some keywords to have drifted in both directions.
What is a good keyword difficulty?
There is no universal good score, because the score describes the keyword, not you. Ahrefs is explicit that KD is calculated for an “average website”. A domain with years of links in a topic can win keywords that a brand-new site cannot touch, at the same score.
Some tools do give new sites a starting point. Semrush’s guide to SEO for a new website suggests new domains target keywords under 20 difficulty, on Semrush’s own scale. That is a reasonable rule of thumb for that one tool. It does not translate directly to Ahrefs KD or anyone else’s number.
A more useful answer to “how much keyword difficulty is good for SEO” is relative. Look at the scores of keywords you already rank for in the top ten, and treat that range as your current ceiling. As your site earns links, the ceiling moves up. A long-tail keyword strategy works partly because it sits below most sites’ ceilings from the start.
The score also says nothing about the reward. A keyword with low difficulty and no meaningful demand is easy for a reason: nobody has bothered to compete for it. Pair the difficulty column with search volume, intent and what a ranking page could earn, and judge the keyword on all of them together. A slightly harder keyword with clear commercial intent is often the better target than an easy one that will never send a buyer.
What the score leaves out
Ahrefs lists the limitations of its own metric, and they are worth taking seriously:
- Content quality. The score does not read the ranking pages. A thin page with links and a thorough page with links score the same.
- Freshness. Topics where Google rewards recent content can be easier to enter than the link count suggests.
- Intent. If the ranking pages do not match what searchers want, the links behind them matter less.
- No guarantee. Matching the backlink counts of the top ten does not guarantee a ranking.
Here is the stated limitation of any difficulty score, including the best one: it is an average over ten pages, compressed into one number. Two keywords with the same score can have completely different results pages underneath. One might be ten established brands with a few links each; another might be one heavily linked page and nine weak ones. The score cannot tell those apart. You can, in about a minute.
The better check: who actually ranks
Open the results page and read it. The question is not how many links the top ten have on average, but whether any page in the top ten looks beatable. Signs that it might be:
- Forum threads, Q&A pages or user-generated content ranking for a commercial query.
- Small, specialist sites ranking alongside large ones.
- Pages that answer a slightly different question from the one in the keyword.
- Out-of-date pages that have not been revised in a long time.
A full walkthrough of reading a results page is in our guide to how to do SERP analysis. The short version: a low-authority site in the top five is better evidence of an opening than any difficulty score, because it shows the opening exists rather than estimating it.
That logic also works in reverse. Rather than starting from keywords and checking who ranks, you can start from sites that rank well despite modest authority, and see which keywords carry them. Our piece on finding low-DR, high-traffic sites covers that approach. It treats the ranking sites as the evidence and keyword difficulty as a rough filter on top, which is the right way round.
None of this makes the difficulty column useless. It is a fast way to drop the keywords that are clearly out of reach before you spend time on them. Use it to shorten the list, then let the results pages decide what stays on it.
Frequently asked questions
What does keyword difficulty measure?
In Ahrefs’ version, how many websites link to the top 10 ranking pages for a keyword, expressed on a 0–100 scale. Other tools use their own formulas.
What is a good keyword difficulty for a new website?
Semrush suggests new domains target keywords under 20 on its own scale. On any tool, the keywords you already rank for are the best guide to your ceiling.
Why do SEO tools show different keyword difficulty scores?
They calculate it differently, using different link data and weightings. Compare scores only within one tool.
Is low keyword difficulty enough to rank?
No. The score ignores content quality, freshness and intent, and matching the top pages’ links does not guarantee a ranking.
The takeaway Keyword difficulty estimates the links behind the current top ten. Use it to cut the list, then read the results page: who ranks is better evidence than any score.
