Keyword difficulty (KD) is a 0 to 100 score that SEO tools use to estimate how hard it is to reach Google’s first page for a keyword. Higher means harder. The catch is that every tool builds the score from different data: Ahrefs counts referring domains to the top 10 pages, Semrush blends links, authority and SERP features, and Moz uses Page Authority, Domain Authority and estimated click-through rate. So the same keyword can look easy in one tool and hard in another. Use KD as a first filter, then judge the actual results page against your own site.
I’ve seen people abandon good keywords because one tool said 60, and chase terrible ones because another said 12. Neither number was wrong. They were answering different questions.
Why Do Keyword Difficulty Scores Disagree Between Tools?
Most people compare KD across tools as if it were a shared unit, like kilograms. It isn’t. Each vendor picked its own inputs, its own weighting and its own link index, then squeezed the result onto the same 0 to 100 scale. Two scores of 40 from different tools don’t describe the same amount of work.
Three things drive the gap:
- Different inputs. Ahrefs looks almost only at links. Semrush adds domain authority and SERP features. Moz adds page-level authority and a click-through estimate.
- Different link indexes. Each company crawls the web separately, so the link counts behind the scores differ before any formula runs.
- Different scales. Ahrefs openly says its scale isn’t linear. A move from 70 to 80 means far more links than a move from 10 to 20.
Picture one results page to see how this plays out. Say the top 10 holds a few big brand pages with modest backlinks, plus several small blogs with almost none. Ahrefs averages the referring domains and may call it fairly easy. Semrush sees the high Authority Score of the brands and pushes the number up. Moz weighs page and domain authority plus how many clicks organic results get, and lands somewhere else again. Three honest formulas. Three different answers.
The metrics that decide whether you can win are the ones none of the formulas can see: how good the ranking pages actually are, whether your site is relevant to the topic, and whether your page type matches what Google shows. I’ll come back to those after the tool breakdown.
How Does Ahrefs Calculate Keyword Difficulty?
Ahrefs keeps it deliberately simple. The company says KD is based on the number of referring domains the top 10 ranking pages have, organic results only. It doesn’t take any on-page factors into account, and it estimates your chance of reaching the top 10, not the top 3.
The useful part is that Ahrefs publishes how its scale maps to links. On its free Keyword Difficulty Checker page, the relationship looks like this:
| Ahrefs KD | Approximate referring domains needed for the top 10 |
|---|---|
| 0 | 0 |
| 10 | 10 |
| 20 | 22 |
| 30 | 36 |
| 40 | 56 |
| 50 | 84 |
| 60 | 129 |
| 70 | 202 |
| 80 | 353 |
| 90 | 756 |
Look at the top end. It explodes. KD 80 suggests about 353 referring domains, while KD 90 suggests about 756. That’s why I read Ahrefs KD as a rough link budget. Then I compare it with how many sites already link to the page I’d rank. If my page has 5 referring domains and the keyword says 84, I know the gap before writing a word.
The curve matters in the middle too. KD 10 and KD 20 sit 10 points apart and only 12 referring domains apart. KD 50 and KD 60 are also 10 points apart, but the gap is 45 referring domains. So “just a bit harder” means very different things depending on where you are on the scale.
One more caveat, and it’s a big one. Those figures describe the pages ranking today, and they’re averages. A single weak page sitting at position 8 won’t pull the score down much, yet it might be the exact slot you can take.
How Does Semrush Calculate KD%?
Semrush casts a wider net. According to its knowledge base, KD% considers the median number of referring domains to the ranking URLs, the median ratio of dofollow to nofollow links, the median Authority Score of the ranking domains, and the SERP qualities of the keyword itself.
Semrush also labels its ranges, which makes the score easier to explain to a client:
| Semrush KD% | Label |
|---|---|
| 0–14 | Very easy |
| 15–29 | Easy |
| 30–49 | Possible |
| 50–69 | Difficult |
| 70–84 | Hard |
| 85–100 | Very hard |
The more interesting metric is Personal Keyword Difficulty (PKD) in the Keyword Magic Tool. Semrush says PKD estimates how hard a keyword is for a specific domain, using the topical relevance between that domain and the keyword, the competition in the topic, and your domain’s metrics against the sites already ranking. That’s the closest any mainstream tool gets to the question you actually care about.
In practice, I’d read a Semrush KD% of 35% and a PKD in the same range very differently from a KD% of 35% with a much higher PKD. The first says the SERP is open and your site fits the topic. The second says the SERP is open, but Semrush doesn’t see your site as relevant yet, which usually means you need supporting content before the main page can compete.
You’ll also see KD next to intent, volume, CPC and competitive density in Semrush’s keyword tables. Competitive density is a paid-search metric on a 0 to 1 scale that tracks advertisers. It says nothing about organic difficulty. Don’t mix the two.
How Does Moz Calculate Difficulty?
Moz describes Difficulty as a 1 to 100 score that estimates how hard it would be to rank in the top ten. In its help center, Moz says the score is calculated by taking into account Domain Authority, Page Authority and the estimated click-through rate for the pages already ranking.
What I like most about Moz is not the formula but the workflow in that same article. You filter the keywords your site already ranks for in positions 1 to 10, put them in a list, and look at which Difficulty range holds the most of them. That range becomes your benchmark. Moz’s example site landed at 21 to 40, which meant keywords in that band were realistic now and anything higher belonged in a longer plan.
That benchmark idea works in any tool. Steal it shamelessly. It turns an abstract score into “how does this compare with what I already win?”
Here’s how I’d copy it in Ahrefs or Semrush. Export the keywords your site ranks for, keep only positions 1 to 10, and look at where most of their KD values cluster. If your top 10 rankings mostly sit between KD 5 and KD 20, that’s your current comfort zone. A keyword at KD 45? Not impossible. But it’s a project, not a quick article.
Brand new sites have no rankings to benchmark against, so they should stay near the bottom of whichever scale they use and read every SERP carefully. After a few months of rankings, run the benchmark again. It won’t stay put. As your site earns links and covers more of its topic, the comfort zone climbs with it.
Ahrefs vs Semrush vs Moz Keyword Difficulty Side by Side
| Criterion | Ahrefs KD | Semrush KD% | Moz Difficulty | What it means for you |
|---|---|---|---|---|
| Main input | Referring domains to top 10 pages | Referring domains, follow ratio, Authority Score, SERP features | Page Authority, Domain Authority, estimated CTR | Ahrefs is the easiest to reason about |
| Published link mapping | Yes, KD to referring domains table | No | No | Only Ahrefs gives you a rough link budget |
| Personalized to your site | No | Yes, through PKD | Indirectly, through the benchmark workflow | Semrush PKD is the only built-in personal score |
| Sees content quality | No | No | No | You must inspect the SERP yourself |
| Free option | Free KD checker page | Free account, up to 10 searches a day | Free version of Keyword Explorer | Fine for spot checks, not for full lists |
My Verdict on Which Score to Trust
Ahrefs KD is the better score for planning link work, because it maps directly to referring domains. However, if you’re judging whether your specific site can rank, Semrush’s Personal Keyword Difficulty or a Moz-style benchmark of your own rankings beats any generic KD.
My honest take? The tool matters less than consistency. Pick one tool. Learn what its numbers mean for your site, and never compare one vendor’s score with another’s. When I build a keyword plan, I use KD to sort a long list into rough bands, then I stop trusting the number and open the results page.
My bands are simple. Keywords at or below the difficulty my site already ranks for go into “now”. Keywords one band above go into “next”, and I plan supporting content and a few links for them over the following 6 months. Anything far above goes into “later” or gets cut. In my experience, that three-bucket view wastes far fewer articles than a single KD cutoff.
Why not a fixed rule like “only target KD under 20”? Because a fixed cutoff ignores your site entirely, and the whole point of judging difficulty is to compare the people already ranking with the site you actually have today, not with some average site. A cutoff that’s sensible for a brand new blog is far too timid for an established store with hundreds of linking domains.
What Actually Decides Whether You Can Win a Keyword?
KD squeezes a whole results page into one number. Details get lost. Four checks tell you far more:
- SERP weakness. Open the results and look for thin pages, outdated posts, forum threads, or pages that only half answer the query. Those are gaps a single KD average hides.
- Your authority relative to the page one sites. Don’t compare yourself with the web. Compare yourself with the weakest page that ranks. If a site like yours is in the top 10, the door is open.
- Topical coverage. A site that already covers the surrounding subject has an edge the formula can’t see. Semrush builds topical relevance into PKD for this reason. If you have one lonely post on a subject, expect a harder climb than the score suggests.
- Intent and format match. If Google shows product pages and you plan a blog post, the difficulty is effectively 100 no matter what the tool says.
The fourth check catches the most mistakes. A tool can only see links and authority. It can’t tell that every result on page one is a downloadable template or a product listing while you planned an essay. Google has already decided what format satisfies that search, and fighting the format is harder than fighting the links.
Here’s the SERP check I run on every shortlisted keyword. It takes about 10 minutes:
- Which page types rank: guides, product pages, local results, videos?
- Are there big brands in every slot, or a mix of sizes?
- How old are the ranking pages, and are any clearly stale?
- Does any result miss an obvious part of the question?
- Is there a site of similar size and focus to mine anywhere on page one?
If I get three or more encouraging answers, I’ll target the keyword even with a middling score. If the whole page is major brands with fresh, deep content, I pass. Even when KD looks low.
Topical coverage is the lever most small sites underuse. Building out a subject properly, which I explain in my guide to creating a topical map, can make harder keywords winnable over time without chasing hundreds of links.
When Does the Low KD Answer Flip?
A low score isn’t always an opportunity. A high one isn’t always a wall. The answer flips in these cases:
- If page one is all giant brands (marketplaces, government sites, big publishers), treat low KD with suspicion. Those pages often rank on the strength of the whole domain, not on links to the single page that Ahrefs counts.
- If your site already ranks on page two for the keyword, difficulty matters less. Improving a page Google already tests is usually cheaper, and my post on finding low hanging fruit keywords in Search Console covers that route.
- If the keyword is local, organic KD can mislead. The map pack runs on different signals, so a “hard” local term may be reachable through a strong Google Business Profile.
- If the keyword has almost no volume, don’t read a very low KD as proof it’s easy. Tool data on rare terms is thin, so check the SERP before assuming anything, and I explain how I confirm real demand for a “0” keyword in why zero search volume keywords still get traffic.
Where Keyword Difficulty Fits in Keyword Research
KD comes after you have a list. Not before. First you pick the seed keywords that describe what you sell, expand them into real phrases, then use difficulty to sort that list into now, next and later. Using KD before you have seeds just gives you easy keywords that don’t match your business.
If you’d rather have that sorting done for you, my team handles it as part of our keyword research service, where every keyword is judged against your site’s authority and the live results rather than a single score.
Frequently Asked Questions
Is There a Free Keyword Difficulty Checker?
Yes. Ahrefs offers a free Keyword Difficulty Checker that shows the KD score and an estimate of the referring domains needed to reach the top 10, with a choice of country. Semrush’s free account allows up to 10 searches a day, and Moz offers a free version of Keyword Explorer. That’s fine for spot checks. For a full keyword list, you’ll need a paid plan or a lot of patience.
What Is a Good Keyword Difficulty for a New Website?
There’s no universal number, because each tool uses a different scale. As a starting point, Semrush calls 0 to 29 easy, and an Ahrefs KD of 10 maps to about 10 referring domains. Better still, check what difficulty band your site already ranks in once you have some rankings.
Is Google Keyword Planner Competition the Same as Keyword Difficulty?
No. Google defines Keyword Planner’s Competition column as the number of advertisers that showed on a keyword relative to all keywords across Google. It measures ad competition, not how hard it is to rank organically.
Does Google Use Keyword Difficulty to Rank Pages?
No. KD is an estimate built by tool companies from their own link data. Google doesn’t publish a difficulty score, and ranking depends on relevance, quality and many other signals that no KD formula captures fully.
Last updated: September 2026 by Mizanur Rahman



