ResumeAdapter
The rubric
Updated 2026-08-09

Two reviewers,
one resume, the same reading.

What this is for

Feedback stops being useful the moment two advisors say different things about the same document. A rubric fixes that by writing down what each score actually means before anyone starts scoring. Five dimensions, four bands each, weighted, printable, free to reuse with attribution.

Print with Ctrl or Cmd + PReuse with attribution
Resume evaluation rubric

Five dimensions, four bands, one weighting.

Score each dimension on its own before looking at the total. Every band description is checkable against the document and the posting, so a reviewer can pick a band without knowing anything about the candidate.

Resume evaluation rubric: five weighted dimensions scored across four bands from Strong to Not evident
Dimension4 · Strong3 · Adequate2 · Developing1 · Not evident
01 · Weight 15
Parse integrity

Does the software that reads this file recover the name, contact details, section headings, employers, titles, and dates without losing any of them?

Extracted as plain text, every element survives: name, contact details, all section headings, every employer, title, and date range. Single column, no tables or text boxes, contact details in the body.

Everything material survives, with cosmetic loss only, such as an icon or a horizontal rule. Reading order stays correct.

One element is lost or scrambled: dates detach from employers, a heading merges into body text, or contact details sit in a header that does not extract.

A whole section is missing or unreadable after extraction, or the reading order interleaves two columns into nonsense.

02 · Weight 35
Required criteria

For every requirement the posting states as required, is there evidence on the resume, in language a reader searching for that requirement would find?

Every required criterion has evidence on the page, stated in language a reader searching for that requirement would find.

Every required criterion is met, but one or more are described in an internal project name, a former employer's jargon, or a synonym the posting does not use.

One required criterion has no evidence behind it. The rest are met, whether or not the wording matches.

Two or more required criteria have no evidence behind them.

03 · Weight 15
Preferred qualifications

Which preferred qualifications are met, and are the met ones visible rather than buried?

Most preferred qualifications that are genuinely met are stated, and they sit where a reader will see them rather than in the last line of the oldest role.

Several met preferences are stated, others are met but never mentioned. Nothing is claimed that is not held.

Preferences are largely unaddressed even where the underlying experience exists elsewhere on the resume.

Preferred qualifications are claimed without evidence, or contradicted by the experience section.

04 · Weight 20
Evidence quality

Does each claim carry a scope, an outcome, or a number, rather than asserting a responsibility?

Most bullets carry a scope, an outcome, or a number. Where a figure is genuinely unavailable, scope stands in: team size, system size, client count.

A mix. The strongest roles carry evidence, older or less relevant roles assert responsibilities.

Most bullets describe duties rather than results. The reader learns what the person was responsible for, not what changed.

Bullets are a list of tools and adjectives with no scope, outcome, or number anywhere.

05 · Weight 15
Human readability

In the first fifteen seconds, can a human reader tell what this person does, at what level, and why they fit this specific role?

Within fifteen seconds a reader can state the person's function, their level, and why they fit this posting. The top third does that work.

Function and level are clear; fit to this specific posting takes a second read to establish.

The reader has to reconstruct the story from the roles, because the top of the page is a generic summary that would suit any posting.

Function or level is ambiguous after a full read, or the document argues for a different role than the one being applied to.

Resume evaluation rubric · ResumeAdapter · resumeadapter.com/resume-review/scoring-rubric · Free to print and adapt with attribution

The quick answer

How do you score a resume with a rubric?

Score each of five dimensions on a 1 to 4 band using its written band descriptions, then weight and total. The dimensions and weights are parse integrity at 15, required criteria at 35, preferred qualifications at 15, evidence quality at 20, and human readability at 15. To convert a band to points, divide the band by 4 and multiply by the weight, then add the five results for a score out of 100. Required criteria carries the largest weight because it is the only dimension that can end an application by itself, and a band of 2 or lower there is worth treating as a stop sign whatever the total says. Score the dimensions independently and in order, before looking at the sum, because the point of a rubric is to find the weakest dimension rather than to produce a number. .

Primary source

Rubric-based review is standard practice in university career services, and the first pass is increasingly automated alongside it. In a NACE quick poll of 448 colleges and universities, fielded 12 August to 6 September 2025, 76 percent of career centers reported using AI as an assistive tool when working with individual students, with helping students create resumes leading the list of uses.

NACE, Quick Poll: Use of AI by Career Centers to Help Students Is Growing (November 2025) (accessed 2026-08-09)

Scoring

Band first, total last.

Divide each band by 4, multiply by the dimension's weight, and add the five results. The arithmetic is deliberately simple, because the judgment lives in choosing the band, not in the formula.

Worked total

Parse 4 → 15.00
Required 3 → 26.25
Preferred 3 → 11.25
Evidence 2 → 10.00
Readability 3 → 11.25

Total 73.75, so 74 of 100.

Read the worked example the way a reviewer should read a real one. The total of 74 is the least interesting number in it. What matters is that evidence quality scored 2: the bullets describe duties rather than results. That single band is the next edit, and it is worth ten points on its own if it moves to a 4.

Two rules keep a rubric honest once several people use it. Score the dimensions in order and independently, because a strong first impression contaminates every later judgment once you have formed it. And never adjust a band to make a total feel right; if a total looks wrong, the weights are wrong for your context, and you change them for the whole cohort rather than for one candidate.

Four of the five dimensions are scored against a specific job description, so a score only travels with the posting it was measured against. Recording a resume as a 74 with no posting attached is recording nothing.

For teams
Getting two reviewers to agree
Calibrate first

Score three together before scoring anything apart

Take three resumes against one posting and have every reviewer band them independently, then compare. Disagreements of more than one band almost always trace to a different reading of a descriptor, and that is a conversation worth having once rather than a discrepancy students discover later.

Record the posting

A score without its job description is not a score

Four dimensions move when the posting moves. Store the posting reference next to the score so a later rescore is comparable, and so a second advisor picking up the file knows what the first one was measuring against.

Adapt deliberately

Change weights for a cohort, never for a candidate

Early-career programmes often raise readability and lower preferred qualifications, which is reasonable. What breaks the instrument is reweighting per person: a rubric applied inconsistently is an opinion with arithmetic attached.

FAQ

Resume rubric FAQ

Questions from advisors and writers putting this rubric into a workflow, and from job seekers using it on themselves. Answers are byte-identical to the FAQPage JSON-LD on this page.

What is a resume scoring rubric?

A rubric is a fixed set of evaluation dimensions, each with written descriptions of what every score band looks like, used so that different reviewers reach the same reading on the same document. It is the difference between a review that says this resume feels weak and a review that says required criteria scored 2 of 4 because one stated requirement has no evidence behind it. The written band descriptions are the part that does the work: without them, two advisors will use the same numbers to mean different things, and a student who sees two advisors gets two contradictory reviews.

How do you score a resume out of 100?

Score each of the five dimensions on a 1 to 4 band, then convert: divide the band by 4, multiply by that dimension's weight, and add the five results. The weights in this rubric are parse integrity 15, required criteria 35, preferred qualifications 15, evidence quality 20, and human readability 15. Required criteria carries the largest weight because it is the only dimension that can end an application on its own. A worked example: bands of 4, 3, 3, 2, and 3 give 15 plus 26.25 plus 11.25 plus 10 plus 11.25, which is 73.75, so 74 out of 100.

What is a good resume score?

There is no universal threshold, and publishing one would be misleading, because four of the five dimensions are measured against a specific job description and change when the posting changes. A resume scoring in the seventies against a demanding posting may be stronger than the same resume scoring in the nineties against a loose one. Use the rubric two ways instead: to find the lowest-scoring dimension, which is where the next edit should go, and to compare a revision against its own earlier version on the same posting. A band of 2 or lower on required criteria is the one result worth treating as a stop sign regardless of the total.

Can I use this rubric with my students or clients?

Yes. Print it, adapt the weights to your programme, and use it in appointments or workshops. If you republish it or adapt it in a handout, attribute it to ResumeAdapter with a link to this page. The one adaptation to think about before making it: if you change the weights, change them for everyone you review in that cohort, because the value of a rubric comes from applying the same one consistently, and a rubric reweighted per student is just an opinion with arithmetic attached.

Do career centers use resume rubrics?

Widely, and the practice predates AI tooling by decades: rubrics are how a career center gets consistency across advisors, peer reviewers, and student workers who all see the same resumes. What has changed recently is that the first pass is increasingly automated. In a NACE quick poll of 448 colleges and universities fielded in August and September 2025, 76 percent of career centers reported using AI as an assistive tool when working with individual students, with helping students create resumes leading the list of uses. The rubric does not become redundant in that setup; it becomes the thing the automated pass is measured against.

How is this different from an ATS score?

An ATS-style score is mostly this rubric's first two dimensions, parse integrity and required criteria, expressed as one number. This rubric keeps them separate and adds three the automated version usually cannot judge well: whether met preferences are actually visible, whether claims carry evidence rather than asserting responsibility, and whether a human reader can place the candidate in fifteen seconds. That is deliberate. The mechanical dimensions are the ones worth automating, and the last one is the one a person still has to score.

Continue
The method behind it
Pillar

The five passes of a resume review

Where these five dimensions come from, and the order a review has to run in.

Read
Dimension 02

Scoring required criteria in practice

The heaviest dimension, worked through criterion by criterion on an illustrative posting.

Read
For professionals

Running this across a client load

Consistency across reviewers, and the authorization question that comes with someone else's document.

Read
Score it against a real posting

The rubric works on paper.
Dimensions 01 to 03 also work automatically.

Parse integrity, required criteria, and preferred qualifications are the mechanical dimensions, and they are the ones worth running against a real posting rather than eyeballing. Evidence quality and readability still need you. Free to run, and you sign in first.

Or use the checklist version →