How to Judge Keyword Difficulty Without Trusting a Single Tool Score
Keyword difficulty (KD) is a 0 to 100 score that SEO tools use to estimate how hard it is to reach Google’s top 10 for a keyword. Each tool calculates it differently, mostly from the links pointing to ranking pages, so scores disagree. Use KD as a starting point, then judge live results: intent, content quality, authority and SERP features.
What is keyword difficulty?
Keyword difficulty is a tool’s estimate of how hard it will be to rank on the first page of Google for a keyword. It comes as a score from 0 or 1 up to 100, and a higher score means a harder fight. Most tools build it mainly from the links pointing to the pages that already rank.
KD is not a Google metric. Google publishes no difficulty score. Each vendor looks at the pages that rank today, measures their strength in its own way, and turns that into one number.
So a KD score answers a narrow question: how strong, in link terms, are the pages that rank today? It does not answer the question you actually care about, which is whether your site can beat them.
If you are new to how Google reads a query before it ranks anything, start with our guide on how search engines understand search queries, intent and meaning. Difficulty only makes sense once you know what the searcher wants.
How Ahrefs, Semrush and Moz calculate keyword difficulty
Ahrefs counts the websites linking to the top 10 pages. Semrush blends median referring domains, the dofollow to nofollow ratio, the Authority Score of ranking domains and the SERP features on the page. Moz scores the top 10 organic links using Page Authority and Domain Authority, so three recipes give three numbers.
Here is how each tool describes its own score, taken from its documentation.
|
Tool |
What goes into the score |
Scale |
How to read it |
|
Ahrefs KD |
Number of websites linking to each of the top 10 ranking pages |
0 to 100 |
Higher means you need more referring domains to reach the top 10 |
|
Semrush KD% |
Median referring domains to ranking URLs, median dofollow to nofollow ratio, median Authority Score of ranking domains, SERP qualities of the keyword |
0 to 100 |
0 to 14 very easy, 15 to 29 easy, 30 to 49 possible, 50 to 69 difficult, 70 to 84 hard, 85 to 100 very hard |
|
Moz Difficulty |
Page Authority and Domain Authority of the top 10 organic blue links |
1 to 100 |
Moz calls anything over 40 an uphill climb |

Ahrefs: a pure links count
Ahrefs explains that its KD looks at the top 10 ranking pages and counts how many websites link to each one. Its own example is “best womens hiking boots”, with a KD of 6. Ahrefs estimates you would need about 7 referring domains to reach the top 10 for it. The score is a translation of a link count, nothing more.
Semrush: links, authority and SERP features
The Semrush Knowledge Base lists four inputs. These are median referring domains, the median dofollow to nofollow ratio, the median Authority Score of ranking domains and the SERP qualities of the keyword. Because it adds authority and SERP features to the link data, its score will not always match a pure link count.
Moz: page and domain authority
Moz’s help page says its Difficulty score measures the top 10 organic blue links using Page Authority and Domain Authority. Moz is clear that a score over 40 is an uphill climb, which is a useful line to remember.
Why the same keyword gets different scores in different tools
Scores differ because each tool uses its own formula and its own data. A keyword can look easy in one tool and hard in another. They are measuring different things in different ways.
In one Moz Q&A thread, a user saw the same keyword scored 80 in Semrush and 20 in Moz. They had no idea which number to trust. A similar question on Quora asks why KWFinder, Semrush and Ahrefs disagree. Si Quan Ong of Ahrefs answers that each tool calculates it differently. He adds that Ahrefs estimates the top 10, not the #1 spot, so one very strong #1 page can make the score look wrong.
Five reasons the numbers drift apart
- Different inputs. Ahrefs counts linking sites. Semrush mixes in authority and SERP features. Moz uses its own authority scores.
- Different link data. Each tool works from its own link database, so the counts behind the scores need not match.
- Different country data. A UAE database and a US database can show different top 10s for the same phrase.
- Different timing. Results change, and a tool’s data may be older than today’s results page.
- Different scales. A 40 in Moz and a 40 in Semrush do not mean the same thing.
How to run the 3-tool comparison yourself
Use the gap as a signal rather than picking a winner.
- Pick three to five keywords you care about, including at least one English and one Arabic phrase.
- Check each one in Ahrefs, Semrush and Moz on the same day. Set the country to the UAE wherever the tool allows it.
- Record the score and the date in a simple sheet, and note each tool’s band, not just the number.
- Where the tools agree, you can trust the general direction.
- Where they disagree by a wide margin, open Google and check the results yourself with the scorecard below.
A wide gap can point to one strong page, a crowded results page or thin data for the market. You can check each of those on the results page itself.
Keyword difficulty vs Keyword Planner competition: not the same metric
Google Keyword Planner’s “competition” column is about ads, not organic rankings. Google defines it as the number of advertisers showing on a keyword relative to all keywords. It is a paid search measure, so it does not tell you how hard the organic results are.
That definition comes from Google Ads Help. Because it counts advertisers, it says nothing about the links, content or authority of the pages in the organic results. Treat it as an ads signal and judge organic difficulty separately.
What a KD score cannot see
A KD score mostly sees links. It misses intent, content quality and age, and the space taken by AI Overviews, local packs, video and aggregators. Those factors often decide whether you can rank and whether ranking is worth it.
Google’s own guide to its ranking systems lists link analysis and PageRank alongside freshness, passage ranking, BERT and other systems. Links are one input, not the whole picture. Google’s Gary Illyes has gone further. At Pubcon in 2023 he said links are not a top three ranking factor, though they still matter, as Search Engine Land reported. In 2024 he said Google needs very few links to rank pages, according to Search Engine Roundtable.
Here is what a link-based score misses:
- Intent. If Google mostly shows tool pages and you plan a blog post, a low score will not fix the format mismatch.
- Content quality and age. A score does not show whether the ranking pages are thin or out of date.
- AI Overviews. On 12 November 2026, Google UAE showed an AI Overview at the top for “keyword difficulty”. It pushes every organic result further down the screen.
- Local packs and maps. When a map appears above the blue links, it competes for the same clicks.
- Video and aggregators. Marketplaces, directories and video results take places a business site might want.
How to judge keyword difficulty yourself: the 6-check SERP scorecard
Open Google in the country you target, search the keyword, and score the top 10 on six checks. Give each check 0, 1 or 2 points, where 2 means the results are open to you. Our rule of thumb: 9 to 12 suggests a fair chance, 5 to 8 a long project, and 0 to 4 a fight to skip for now.
These bands are our own, meant to force a decision rather than predict a ranking.
|
Check |
0 points |
1 point |
2 points |
|
Links to ranking pages |
Most pages have far more referring domains than yours |
A mix of strong and lightly linked pages |
Several pages have few or no links |
|
Authority vs your site |
Big brands and strong domains fill the top 10 |
Some sites are close to yours in strength |
Many sites are weaker than yours |
|
Intent fit |
Google shows a page type you cannot make |
Mixed page types |
Google shows the page type you plan |
|
Content quality and age |
Fresh, deep, expert pages |
Some pages are thin or dated |
Many pages are thin, old or off-topic |
|
SERP features and local packs |
AI Overview, map, video or ads push results far down |
One or two features above the results |
Mostly plain organic results |
|
Weak spots |
No forums, directories or social posts in the top 10 |
One weak result |
Several weak results |

How to score each check
For links and authority, you can still use your tools. Look at referring domains and domain strength for each ranking page, not the single KD number. Compare them with your own site’s figures from the same tool, so you compare like with like.
For intent fit, look at what type of page ranks. Our guide to analysing a SERP before you write walks through page types, formats and angles in detail.
For content quality, open the top five pages. Check the date, the depth and whether a real expert wrote them. For SERP features, note anything that sits above or between the organic results.
Finding cracks in the SERP: forums, thin pages, old pages and weak domains
Cracks are results that should not be hard to beat. Forum threads, social posts, directory listings, thin pages, outdated articles and low-authority sites in the top 10 all tell you Google lacks a strong answer. Each one is a place your page could take.
Ahrefs uses the same idea in its own keyword difficulty guide, listing outdated results, low-DR sites, forums and fluffy content as signs of an open SERP. Look for:
- A forum, Facebook page or Instagram post ranking for a commercial or informational term
- A directory listing or marketplace category page with no real content
- An article with old facts, old prices or old screenshots
- A page that only half answers the query
- A small site with few links ranking on page one
Our rule of thumb: one crack can be luck, while two or three suggest a focused, useful page can break in.
Difficulty for your site, not an average site
KD scores difficulty for an average site. Your real difficulty depends on your own authority, how much you have already published on the topic, and how well your internal links support a new page. The same keyword can be hard for a new site and fair for an established one.
This is why both Semrush and Ahrefs now talk about personal keyword difficulty, a score adjusted for your own site. Before you trust any score, ask three questions:
- Do we already rank near this topic? Pages ranking for related terms make the next one easier.
- Do we have strong pages to link from? Internal links pass strength and context to a new page.
- Is this part of a cluster? A page inside a group of related articles has more support. Our guide to clustering keywords by SERP overlap shows how to build those groups.
If the answer to all three is no, aim lower first. Our post on head terms vs long-tail keywords explains where smaller sites should compete. Then decide what to chase using search volume vs business value, so difficulty is weighed against what a keyword is worth.
Keyword difficulty in the UAE: Arabic vs English results
In the UAE, the English and Arabic versions of the same search can return very different top 10s. Marketplaces, directories and social posts can also take places on page one. Check the live results for each language before you trust a score built for one of them.
We compared both versions of one service search on Google UAE on 20 November 2026. “ac repair dubai” and its Arabic equivalent “تصليح مكيفات دبي” shared only one website in their top 8. The two results pages differ so much that each language needs its own check.
Three local patterns matter:
- Aggregators and marketplaces. Urban Company ranked for the English search, and a Dubizzle community services listing ranked for the Arabic one.
- Social and directory pages. A Facebook page ranked in the English top 8. An Instagram post and a directory listing ranked in the Arabic top 8.
- Local packs. Service searches can also show a map. Winning there depends on your Google Business Profile, which our local SEO services team handles separately from page rankings.
For the full method on Arabic terms, see our guide to Arabic keyword research for UAE and KSA websites.
Worked example: scoring a Dubai service keyword against live Google UAE results
Here is the scorecard applied to “ac repair dubai” and “تصليح مكيفات دبي” using the Google UAE top 8 on 20 November 2026. The English results lean towards company sites and one marketplace, while the Arabic results show more cracks. We had no link or map data for this test, so we score only the checks we could see.
What ranked
For “ac repair dubai”, the top 8 held five local company sites, Urban Company, one blog post and a Facebook page with about 1.2K followers. The People Also Ask box asked, “How much is AC service in Dubai?”
For “تصليح مكيفات دبي”, the top 8 held company sites, one showing prices, a Dubizzle listing page, an Instagram post and a directory listing. People Also Ask again raised a cost question.
How the checks read
|
Check |
“ac repair dubai” |
“تصليح مكيفات دبي” |
|
Links to ranking pages |
Not scored: needs link data from your tool |
Not scored: needs link data from your tool |
|
Authority vs your site |
1: one large marketplace, the rest local companies |
1: one large marketplace, the rest local companies and a directory |
|
Intent fit |
2 for a service page: company sites dominate, one blog post |
2 for a service page: company sites and listings |
|
Content quality and age |
Not scored: open each page and judge depth and pricing |
Not scored: one page shows prices, which sets the bar |
|
SERP features and local packs |
Not scored: check the map and ads yourself |
Not scored: check the map and ads yourself |
|
Weak spots |
1: a Facebook page |
2: an Instagram post and a directory listing |
On the three checks we could score, the English search reaches 4 of 6 and the Arabic search 5 of 6. The other three checks could still change the picture.
What we take from it
Both searches want a service page, so a blog post is the wrong format for either. The English results have one clear crack. The Arabic results have two, an Instagram post and a directory listing. On the cracks we can see, the Arabic search looks slightly more open, pending the link and map checks.
Both cost questions in People Also Ask point to the same gap. A page that shows clear price ranges answers a question searchers clearly ask.
The same test explains the tools’ own keyword. On 20 November 2026, six of the top 8 Google UAE results for “keyword difficulty” were checker tool pages, and the other two were vendor blogs. A blog post there needs an angle the vendors do not cover, whatever the KD says.
How long it really takes to rank
Ranking takes longer than most people expect. An Ahrefs study from May 2025 found that only 1.74% of new pages reach the top 10 within a year. The average #1 page is about five years old.
The Ahrefs time-to-rank study by Patrick Stox also found that 13.7% of top-10 pages are under one year old. New pages can break in, but most of page one belongs to older content.

A low score does not make ranking fast. Moz Q&A users asking how long a keyword under 10 takes got a similar answer. Easy keywords still have competition, and topical authority and crawling still matter, as this Moz Q&A discussion shows.
Set expectations in months, not weeks. Track position over time with a method like the one in our guide to rank tracking done properly, and judge progress on the trend.
To put this into a wider process, our keyword research guide for UAE businesses shows how difficulty fits next to intent and revenue. If you want a second pair of eyes, send us 10 keywords. Our SEO services team will score each keyword against the live Google UAE results, in English and Arabic.
FAQ
Keyword difficulty is a score from 0 to 100 that SEO tools use to estimate how hard it is to reach Google’s top 10 for a keyword. Most tools base it mainly on the links pointing to the pages that already rank. Google does not publish its own difficulty score.
Semrush calls its score KD% and builds it from median referring domains, the dofollow to nofollow ratio, the Authority Score of ranking domains and the SERP features on the page. Its bands run from 0 to 14 very easy up to 85 to 100 very hard. Semrush also offers a Personal KD% for your own domain.
There is no single good score because each tool uses a different scale. As a guide, Semrush rates 0 to 29 as very easy or easy, and Moz calls anything over 40 an uphill climb. A good keyword is one where your site can beat the current top 10, so check the live results before you decide.
Ahrefs and SE Ranking offer free keyword difficulty checkers, and other tools offer a limited number of free checks. You can also judge difficulty with no tool at all. Search the keyword on Google and score the top 10 on links, authority, intent, content quality, SERP features and weak spots.
No. Competition in Google Keyword Planner measures how many advertisers bid on a keyword, so it is a paid search metric. Keyword difficulty estimates how hard it is to rank in the organic results. A keyword can be high on one and low on the other.
They are rough estimates, not predictions. Each tool uses its own formula and link data, and one Moz Q&A user reported the same keyword scoring 80 in Semrush and 20 in Moz. Use the score to sort a long list, then confirm the shortlist by checking the actual search results.
A low score usually means the ranking pages have few links, which is a good sign. It does not guarantee a fast ranking, because intent, content quality and your own site’s authority still matter. Low difficulty paired with low business value is rarely worth the effort.
Yes. Tools such as Semrush let you pick a UAE database. Search the keyword on Google UAE in both English and Arabic, because the two languages often return very different results.
Sources
- Ahrefs: Keyword Difficulty, How to Estimate Your Chances to Rank
- Semrush Knowledge Base: What is KD%?
- Moz Help: Keyword Explorer Difficulty
- Google Ads Help: About Keyword Planner forecasts
- Google Search Central: A guide to Google Search ranking systems
- Search Engine Roundtable: Google needs very few links to rank pages
- Search Engine Land: Links are not a top three ranking factor, says Google’s Gary Illyes
- Ahrefs: How long does it take to rank in Google?
- Quora: Why do KWFinder, Semrush and Ahrefs show different scores for the same keyword?
- Moz Q&A: Moz vs Semrush keyword difficulty
- Moz Q&A: How long does it take to rank easy keywords?
