See MatchCV's required and preferred keyword weights, a worked 70-point example, empty-category behavior, and the limits of a resume match score.
Published by MatchCV. Editorial scope and corrections.
MatchCV's resume match score measures whether terms extracted from one job description appear in your resume text. It does not verify qualifications, reproduce an employer's ATS, or predict an interview. Different products can calculate completely different scores from the same document.
You can use the job description matcher without signing in, within the guest usage allowance. Review the required and preferred keyword lists alongside the total.
The job matcher first asks AI to separate terms from the posting into must-have and nice-to-have lists. It then checks the resume text for those terms.
These are MatchCV's current weights, not a description of how Workday, Greenhouse, or another platform ranks candidates. AI extraction can miss or misclassify terms; inspect the lists against the posting.
This is a fictional example to explain the arithmetic, not a customer result or a benchmark of hiring success.
| Group | Extracted keywords | Found in the example resume | Coverage |
|---|---|---|---|
| Required | SQL, Python, dashboards, stakeholder communication | SQL, dashboards, stakeholder communication | 3 / 4 = 75% |
| Preferred | Tableau, dbt | Tableau | 1 / 2 = 50% |
The calculation is 75 × 0.8 + 50 × 0.2 = 70.
If the applicant has never used Python, adding it to reach a higher score would be misleading. If they have used it and can explain where, they can add that evidence. The report cannot decide which case applies.
An empty group currently contributes zero. If there are no preferred keywords and every required keyword is found, the match score is 80. If there are no required keywords and every preferred keyword is found, it is 20. If both groups are empty, the score is zero.
That behavior is another reason to read the lists instead of treating 100 as a universal target. An extraction that returns no useful terms is not evidence that you are unqualified.
The matcher normalizes case and whitespace. It checks term presence and can also find words from a multi-word phrase separately. This means it may miss synonyms or count a term without understanding its context. For example, merely mentioning a tool in a training goal does not mean you used it professionally.
Check negations, credentials, years of experience, and the context of each claim yourself. Matching a phrase does not prove a certification is valid or that a job requirement is met.
The matcher also displays a separate ATS checklist. With a JD, that checklist weights keyword coverage at 40%, basic text-format checks at 30%, and writing checks at 30%. Without a JD, the general review uses resume fundamentals at 30%, basic text-format checks at 30%, and writing checks at 40%.
These are simple text-based diagnostics. They do not fully inspect visual layout or run your document through an employer's parser. They should not be interpreted as the same metric as the 80/20 job match score.
For a step-by-step example of factual editing, see how to tailor a resume. For role-specific examples, see software engineer keywords and project manager keywords.
No. A score of 70 does not mean a 70% chance of an interview. It summarizes the keyword coverage described above.
Yes. They may extract different terms, use different matching rules, or measure writing rather than coverage. Compare the explanations and your actual application, not the number alone.
No. Stop when your relevant experience is clear and accurate. Unsupported additions, misleading title changes, and keyword repetition can make the document worse even if they raise a text-based score.
Use the resume job description matcher to inspect one posting, then review the evidence before editing.
See how well your resume matches a job description
Get my match score →or sign in free to tailor your resume with AI