ResumeAtlas

Resume writing · original research

How to quantify your resume accomplishments

“Add numbers to your bullet points” is the most repeated resume advice there is, and the least followed. We looked at the bullets on 733 real resumes to see how much quantifying people actually do — then worked out how to fix it, including for jobs that don't come with a dashboard.

Last updated: September 7, 2026

Quick answer

To quantify an accomplishment, name what changed because of your work and attach a number to it: a count (people, clients, tickets, dollars), a rate (per week, per month), or a before/after (time, cost, error rate). If your job has no obvious metrics, multiply a small number by how often you do it, count the things around the work, or use a before/after with no percentage at all. Target at least one quantified bullet per role and two or three in your most recent job — not a number on every line. A rough, labelled estimate (“~40 a week”) is fine; a specific figure you can't explain in an interview is not.

What we found

ResumeAtlas checks a resume against a specific job description. Since the June 2026 version of the analyzer, each run also stores a bullet-level breakdown: how many bullet points the resume has, and how many of them contain a measurable result. We aggregated that breakdown across 733 resumes (2026-06-08 to 2026-09-15).

36

median bullet points on a resume

4.5%

median share of those bullets carrying a number (≈ 2 of 36)

27.7%

of resumes had no quantified bullet anywhere

Share of bullets that were quantifiedResumes% of sample
0–10%47865%
10–20%11816%
20–30%578%
30–40%649%
40% or more162%

The median resume in the sample had 36 bullet points and a number on about 2 of them. Most resumes aren't missing quantification by a little — they're missing it almost entirely. That is good news if you're writing yours now: a handful of real numbers puts you ahead of the field.

Method & limitations

  • Sample: 733 resumes uploaded to ResumeAtlas and checked against a real job description, 2026-06-08 to 2026-09-15. Bullet-level data exists only for the v8+ analyzer, so this window is shorter than our full 1,700+ analyses.
  • “Quantified” means the bullet contains a number tied to a result or scale (%, currency, counts, time, multipliers) — detected by pattern and model, not human-graded. A bare year or a phone-style number is not counted.
  • Medians, not means — a few resume-writers who quantify heavily would drag an average up and hide the typical case.
  • Skew: the sample leans toward software, data and engineering postings, where quantification norms may differ from, say, clinical or creative roles.

First, the shift: from what you did to what changed

A quantified accomplishment isn't a duty with a number bolted on. It answers a different question. A duty says what you were responsible for; an accomplishment says what was different because you were there — and the number is the evidence.

Responsibility

Responsible for managing the team's reporting.

Quantified accomplishment

Owned weekly reporting for a 9-person team; cut turnaround from 3 days to same-day by moving 4 recurring reports into a scheduled dashboard.

Lever: Count what you touch (people, reports) and the before/after on time.

Responsibility

Handled customer support tickets.

Quantified accomplishment

Resolved ~40 support tickets/week at a 96% CSAT, and wrote 12 help-centre articles that deflected the 3 most common questions.

Lever: Volume per week + a quality rate + a second-order effect.

Responsibility

Worked on improving the checkout flow.

Quantified accomplishment

Shipped 5 A/B tests on the checkout flow over 2 quarters; the 2 that won lifted completed checkouts 6% combined.

Lever: Effort you can count (tests shipped) even when only some moved the metric.

Responsibility

Helped onboard new hires.

Quantified accomplishment

Onboarded 7 new analysts in 18 months; rewrote the first-week checklist, and the last 3 hit independent delivery in 2 weeks instead of 4.

Lever: Headcount + a cycle-time change you contributed to.

Notice that none of the “after” versions required a company dashboard. They use counts, frequencies and before/after — things you can reconstruct from memory. For the phrasing around the number, strong openers help: resume action verbs covers the verb; this page is about the number.

When your job “has no numbers”

This is the most common objection, and it's almost always wrong — the numbers are there, they're just not handed to you in a report. Five ways to find them:

Multiply a small number by frequency

“Reviewed 8 contracts a week” becomes “reviewed ~400 contracts a year.” Rough is fine — recruiters read “~” as honesty, not weakness.

Count the things around the work

Team size, number of stakeholders, systems supported, markets covered, budget you had signing authority over, events run. You almost always know these.

Use a before / after, even without a percentage

“Manual month-end close that took 5 days” → “… now 2 days.” A time or step change is a metric.

Quantify scale, not just outcome

“Migrated 12 years of records (≈ 300k rows) with zero data loss.” The size of what you handled is a number.

Name the comparison

“First person in the role to …”, “one of 3 people trained on …”, “highest volume on a team of 11.” Rank and rarity are quantities.

27.7% of the 733 resumes we checked against a real posting had no quantified bullet point anywhere — and the median resume quantified just 4.5% of its bullets. One real number per role would clear most of that gap. See all 1,793 analyses →

How many bullets should actually have a number?

Enough to be credible, not so many it reads as spin. A workable target:

  • Most recent role: 2–3 quantified bullets out of 4–6.
  • Earlier roles: at least 1 quantified bullet each.
  • The rest: scope and responsibility bullets are fine unquantified — not every line needs a digit, and a resume that is wall-to-wall percentages invites disbelief.

That's roughly 20–30% of your bullets carrying a number — against a median of 4.5% in our sample. The bar to stand out is lower than it sounds.

Estimating vs inventing

The line is whether you can walk an interviewer through where the number came from. “About 40 tickets a week” — you can explain that: roughly 8 a day, five days. “Improved team efficiency by 34%” with no story behind it is the kind of figure that collapses under one follow-up question. Label estimates (“~”, “roughly”), round honestly, and never attach a precise percentage to something you never measured. Tailoring the resume to a posting doesn't change this — see how to customize a resume without lying.

Where a tool helps

The hard part is remembering the raw material — the counts, the before/after, the scale — and turning it into a clean line without staring at a blank bullet. If you give it that raw material, ResumeAtlas's work-experience bullet generator drafts quantified bullets from what you actually did (first project free) — you still supply and check the numbers. If you want to see which of your accomplishments even matter for a specific job first, the resume-vs-job-description check flags which requirements your bullets currently prove and which they don't. Neither one should be inventing a number for you.

For the rest of the work-experience section — ordering, tense, how much detail per role — see the work experience examples guide, or the full resume statistics from 1,700+ analyses.

FAQ

How many bullet points on a resume should have numbers?

Aim for at least one quantified bullet in every role, and two or three in your most recent one. That is already far above what most resumes do — in our sample the median resume quantified about 5% of its bullets. You do not need a number on every line; a resume that is all percentages reads as inflated.

How do I quantify my resume when my job has no metrics?

Almost every job has countable context even when it has no dashboard: how many people, clients, tickets, contracts, events, systems or dollars you handled, how often, and what changed before vs after. Multiply a weekly number by 50 for an annual figure, and use “~” for estimates. See the five techniques above.

What is the difference between a responsibility and an accomplishment?

A responsibility is what you were assigned (“responsible for the team's reporting”). An accomplishment is what changed because you did it (“cut reporting turnaround from 3 days to same-day”). Same facts, but the accomplishment names a result and, ideally, a number.

Is it lying to estimate a number I can't prove?

No — a labelled estimate (“~40 tickets/week”, “roughly 15% faster”) is standard and defensible as long as you can explain how you got there in an interview. What fails is a specific, unhedged figure you invented (“increased revenue 47%”) that you cannot walk through.

Do numbers actually matter if a human only skims the resume for a few seconds?

They matter more in a skim, not less. A digit stops the eye in a wall of text, so a quantified bullet is what gets read in the six-second pass. It also gives the interviewer a concrete thing to ask you about, which is usually a conversation you want to have.