Resume Bullet Points: 16 Before-and-After Rewrites That Survive the Scan

Good resume bullet points describe what changed because you were there, not what the job asked of you. "Responsible for managing the social media calendar" tells a recruiter nothing, because every social media coordinator who ever lived was responsible for that. You wrote your job description. You didn't write your resume. Every bullet needs one strong verb, one specific action, and one result with a number or a stake attached.

The formula: [Action verb] + [what you did, specifically] + [the result, with a number]. Lead with the result when it's your strongest part.

Here's what that looks like across roles. Every row below is an invented illustration, not a real person or a real result: the figures show the shape of a strong bullet, not targets. Swap in your own real numbers.

Role	Before (duty)	After (achievement)
Marketing	Responsible for email newsletters	Rebuilt the onboarding email sequence, lifting open rates from 18% to 31% in two quarters
Sales	Managed a territory of accounts	Grew a 40-account territory 22% year over year by re-engaging 9 lapsed customers
Customer support	Handled customer tickets	Resolved ~60 tickets a week across three products at a 94% satisfaction score
Software engineer	Worked on the checkout service	Cut checkout page load time from 3.1s to 1.4s by moving image processing off the main thread
Data analyst	Created weekly reports	Automated the weekly sales report, saving the team 6 hours a week of manual pulls
Product manager	Led product roadmap discussions	Shipped a self-serve upgrade flow that moved 1,200 users to annual plans in its first month
Project manager	Coordinated cross-team projects	Delivered a 5-team ERP migration two weeks early and 8% under its $300K budget
Finance	Assisted with month-end close	Cut the monthly close from 5 days to 2 by scripting the ledger reconciliation
HR / recruiting	Helped with hiring	Filled 14 engineering roles in one quarter, cutting average time-to-hire from 52 to 37 days
Operations	Oversaw warehouse processes	Redesigned pick routes, reducing order errors 30% across a 20-person shift
Teacher	Taught 10th-grade English	Raised the share of students passing the state writing exam from 61% to 78% in one year
Nurse	Provided patient care on a busy ward	Cared for 6 post-surgical patients per shift and trained 4 new hires on the discharge checklist
Designer	Designed marketing materials	Redesigned the pricing page, lifting trial signups 15% in an A/B test
Retail / service	Worked the register	Trained 5 new cashiers and kept the busiest lane under a 3-minute average wait
Student / intern	Assisted the research team	Cleaned and coded 2,000 survey responses used in the lab's published report
Volunteer	Member of the fundraising committee	Organized a 300-guest gala that raised $45,000 for the food bank

Here's what most people get wrong: they think a bullet's job is to prove they did the work. It isn't. The recruiter already assumes you did the work; that's why you held the title. The bullet's only job is to show how well you did it and at what scale, and a duty list can never carry that. The differentiator is outcome and magnitude, and you can find both even in a job that never handed you a dashboard.

What makes a strong resume bullet point?

A strong bullet is specific enough that the next applicant couldn't copy it. The formula above is close to what university career offices teach: Yale's Office of Career Strategy teaches action plus project plus result, and says to quantify the result as a % improvement or increase. The other popular version comes from Google's former HR chief: Laszlo Bock's formula is "Accomplished [X] as measured by [Y] by doing [Z]", which simply puts the result first.

Both work. Pick whichever puts your best fact in the first five words. Picture a hypothetical email marketer, Dana, rewriting her top line:

Weak: "Responsible for managing the company's email marketing program and sending out newsletters to subscribers."

Strong: "Rebuilt the email onboarding sequence, lifting open rates from 18% to 31% and recovering about $40K in at-risk renewals over two quarters."

The weak version describes a job ten thousand people hold. The strong one describes something only Dana did, with figures the recruiter can repeat to the hiring manager.

The formula also beats the STAR-style narrative that flattens everyone into the same shape because it's compressed for skimming. You don't have room to build a situation and a task. You have the front half of one line.

Bullet type	What it tells the recruiter	What it's missing
Responsibility ("Responsible for X")	You held the title	Did you do it well? At what scale?
Activity ("Managed X, handled Y")	You were busy	Did anything change because of you?
Achievement ("Did X, which produced Y")	You moved a number	Nothing; this is the one that gets read

Which action verbs should start a resume bullet?

Start with a verb that names what you personally did, and match it to the result. Career offices agree on the negative rule: UC San Diego's career center says to open bullets with skills-based action verbs and avoid passive terms like "Responsible for" and "In charge of". Here's a compact list, grouped by the kind of result each verb sets up.

• Grew something: increased, expanded, grew, raised, doubled, accelerated, won
• Cut something: reduced, cut, eliminated, consolidated, shortened, lowered
• Built something: built, launched, designed, created, developed, founded, shipped
• Fixed something: rebuilt, redesigned, streamlined, automated, resolved, overhauled
• Led people: led, trained, mentored, hired, coached, recruited, directed
• Persuaded people: negotiated, secured, pitched, convinced, closed, partnered
• Figured something out: analyzed, diagnosed, forecast, modeled, audited, identified

A verb can't carry a weak bullet; "Spearheaded the newsletter" is still a duty in a nicer coat. And if "led" is accurate three times, use it three times rather than reaching for "orchestrated." Thesaurus work shows, and AI tools have made it worse: when everyone runs their resume through the same AI tool, every bullet starts to sound identical. In a January 2024 survey of 625 hiring managers by Resume Genius (a resume-builder company, so read it as vendor research), AI-generated content was the most-cited resume red flag, at 53%. The number is the one part of the bullet a model can't invent for you.

Why do recruiters skip most bullets?

Because they barely read them. Ladders' 2018 eye-tracking study found recruiters skim resumes for an average of 7.4 seconds on the first pass. The resumes that held attention used F- and E-pattern layouts, with bold job titles supported by bulleted lists of accomplishments, rather than dense paragraphs. That six-second scan is very real, and bullets get skimmed, not read.

In a 2018 survey that the job board CareerBuilder commissioned from The Harris Poll (1,138 hiring and HR managers), 23% spent less than 30 seconds on a resume and 39% less than a minute. A bullet that buries its payload at the end has already lost.

Phrasing matters beyond the scan. In a field experiment covering nearly half a million jobseekers, those given algorithmic writing help on their resumes saw an 8% increase in the probability of getting hired. It measured cleaner writing, not quantified bullets, so don't stretch it, but how a line reads moves real outcomes.

How do you quantify bullets when the job never gave you numbers?

You reconstruct them. Almost every job produces measurable impact; it just doesn't hand you a dashboard. Work the question from four angles:

• Scale. How many tickets, accounts, users, events, dollars, people or locations? "Handled customer support" becomes "Resolved ~60 tickets a week across three product lines."
• Frequency. "Wrote blog posts" becomes "Published 2 articles a week for 18 months without missing a deadline."
• Before and after. Anything you improved had a starting state: the wait time, the error rate, the turnaround, the manual step you automated.
• Rank. "Top seller on a team of 9." "Trained the 4 newest hires." Relative position is a number even when absolute output isn't.

If you truly can't find a result, quantify the stakes: the budget you were trusted with, the headcount you coordinated, the size of the client. "Ran point on the agency's largest account ($1.2M a year)" carries scale without an outcome metric.

This matters most for jobs that sound ordinary. In an incentivized resume-rating study, employers recruiting at the University of Pennsylvania gave students no credit for being a lifeguard, barista or cashier, even though those jobs can build useful skills. That study tested job titles, not bullet wording, so it isn't proof a better bullet fixes the gap. The reasonable inference is that a title alone earns you nothing. If your experience is a service job, the bullet has to show the training, volume and responsibility the title hides.

How many bullet points should each job have?

Three to five for recent, relevant roles, fewer for older ones. That's the standard advice: UC San Diego says to write 3-5 bullets describing your outcomes, tailored to the position, and UW-Madison's career center also says 3-5, starting with what is most relevant to the job you want. The "3 bullet rule" you'll see in search results is usually a version of the same idea: enough lines to prove the pattern, few enough that each gets read.

Order matters more than count. The scan grabs the top of each entry, so your most quantified bullet goes first, not stranded at position five.

Situation	Bullets	What goes first
Current or most relevant role	4-5	Your biggest measurable result
Older but relevant role	2-3	The result closest to the target job
Unrelated older role	1-2, or cut	Scale or responsibility, not duties
Internship or student job	2-3	Anything with a count or deliverable

Bullets also have to match your skills list. A skill named in your resume skills section with no bullet proving it two inches lower is just a word the reader discounts.

What are the red flags in resume bullet points?

Typos top the job board's list. In the same 2018 CareerBuilder survey, typos or bad grammar were an instant deal breaker for 77% of hiring managers, a resume without quantifiable results for 34%, long paragraphs of text for 25%, and generic, non-customized content for 18%. Resume Genius's 2024 vendor survey ranks them differently: it put "no measurable achievements" fourth among red flags, cited by 44%, behind AI-generated content, job-hopping and poor formatting, with typos fifth. Different years, different samples, same lesson: an unmeasured bullet is on both lists.

In bullet terms: an opener of "Responsible for," three-line bullets, the same bullet pasted under every job, and a number that can't survive one follow-up question. For red flags beyond the bullets, see what a good resume review should catch.

When does quantifying every bullet backfire?

When the number is weaker than the truth. A "300% increase in engagement" that started from two likes is technically accurate and instantly transparent to anyone who reads resumes for a living. Suspiciously round, huge or unitless numbers ("improved efficiency by 40%," no baseline, no mechanism) read as inflation, and inflation costs more trust than a missing number would.

Some roles resist clean metrics, and forcing them makes worse bullets. Research, early-career roles, support functions and creative work often deliver value that won't reduce to a percentage without distortion. For those, use scale and scope: what you were trusted with, how much of it, how complex, who relied on it. An honest description of real responsibility beats a faked metric every time.

None of this overrides fit, either. Strong bullets win the rooms you belong in. They don't manufacture rooms you don't.

What to do now

1. Mark every bullet that starts with "Responsible for," "Duties included," "Helped with" or "Worked on." Each one is a rewrite candidate.
2. Find one number per bullet. Ask the four questions: scale, frequency, before-and-after, rank. Reconstruct a count from memory if you have to.
3. Rewrite to verb plus action plus result. Use the table above as a pattern, and lead with the result when it's your strongest fact. If you draft with AI, use ChatGPT as an editor of your real numbers, not the author of new ones.
4. Reorder each job so the best bullet sits first. Keep 3-5 for relevant roles and trim older ones to one or two.
5. Pressure-test every figure. If you can't explain how you measured it, soften it to something specific and true. A defensible 19% beats a fictional 300%.
6. Tailor the top bullets to real postings. Pull three listings from our job search and check that your first bullet under each role answers what they ask for.

Staring at a bullet that still says "Responsible for" and can't find the number hiding in it? Paste it to Praxy on WhatsApp with the job you're targeting. I'll ask the questions that surface the scale, pick the verb that fits the result, and hand back a rewrite you can defend in the interview.