You can write strong resume bullets without inventing metrics by replacing vague duties with truthful evidence: scope, context, frequency, audience, constraints, and outcomes.
The rule is not "add a number to every bullet."
The rule is "make the proof easier to believe."
That matters because metric pressure creates bad resumes. A real number can sharpen a claim. A guessed number can turn a defensible bullet into an interview problem. Tiny CV's view is simple: facts before phrasing. If the number is real, use it. If it is not, prove the work another way.
What makes a resume bullet strong?
A strong resume bullet says what you did, how you did it, and why the work mattered.
The University of Nebraska-Lincoln Career Services frames bullet statements around activities, skills, and results, then asks job seekers to answer "what," "how," and "why."1 That is the right starting point because it separates the moving parts. The verb names the action. The context explains the situation. The result tells the reader why the action mattered.
MIT Career Advising and Professional Development makes the same point from a reader's perspective: describe experience with specificity, use strong action verbs, include technical details where they help, and quantify if you can.2
The quiet phrase is "if you can."
Some work produces obvious numbers: revenue, latency, volume, attendance, budget, uptime, conversion, response time, incidents, tickets, accounts, students, or team size. Other work produces judgment, coordination, trust, quality, clarity, risk reduction, or a better process. Those are still real contributions.
The job of the bullet is to make them visible without pretending they were all percentages.
Metrics are proof, not decoration
Use a metric when it is true, relevant, and traceable to your work.
Do not use one because a resume template told you every bullet needs a number. Harvard's resume bullet template says to begin with an action verb, include details that help the reader understand accomplishments, and quantify where possible.3 UC San Diego's resume guide says to include numbers, frequency, and examples, and to quantify when it is impactful.4
Those sources are not saying "make up a metric."
They are saying numbers help when they clarify the claim.
Use this test:
| If the number is... | Use it? | Why |
|---|---|---|
| Directly measured | Yes | You can explain where it came from. |
| Reasonably counted | Usually | Team size, audience size, ticket count, budget, cadence, or duration can show scope. |
| Estimated from memory | Maybe | Use approximate language only if you can defend the basis. |
| Borrowed from a team dashboard you did not influence | Be careful | It may imply ownership you did not have. |
| Invented to sound impressive | No | It weakens trust and creates interview risk. |
This is where a resume source of truth helps. Keep private notes on where a number came from: dashboard, project doc, invoice, calendar, manager feedback, customer list, support queue, or launch report. Tiny CV's markdown source is useful because the public bullet and private evidence can stay close while you draft, then the private notes stay out of the published version.
If you cannot name the source of a metric, treat the metric as unverified until you can.
What to use when you do not have numbers
When exact metrics are missing, use other forms of evidence that show scale, difficulty, or consequence.
Here are the best substitutes:
| Evidence type | What it proves | Example phrasing |
|---|---|---|
| Scope | How large or complex the work was | "across three onboarding surfaces" |
| Frequency | How often the work happened | "weekly release notes" |
| Audience | Who used or depended on it | "for sales and support teams" |
| Constraint | What made the work hard | "during a migration with no downtime window" |
| Before state | What problem existed | "from recurring setup tickets" |
| After state | What improved | "so support could answer from one guide" |
| Specific tools | What skill was actually used | "in React, TypeScript, and PostgreSQL" |
| Decision role | What judgment you contributed | "prioritized fixes with product and support" |
UC San Diego gives a practical version of this: if you worked in a group, say how big; if you attended meetings, say how often; if exact numbers are not available, frequency and approximate scope can still add credibility.4
NACE's career-readiness competencies are also useful here because not every job skill is numeric. Communication, critical thinking, leadership, professionalism, teamwork, and technology all need evidence, but they do not always arrive as tidy metrics.5
The move is to make the competency visible through the work.
Not this:
- Excellent communicator and cross-functional collaborator.
This:
- Wrote weekly rollout notes for product, support, and sales so customer-facing
teams could explain billing changes from the same source.
There is no fake percentage. The proof is in the cadence, audience, artifact, and purpose.
Turn a duty into a bullet
The easiest way to improve a weak resume bullet is to add the missing context before you polish the sentence.
Start with the raw duty:
- Responsible for onboarding emails.
Ask five questions:
- What did you actually do?
- Who was it for?
- What problem did it solve?
- What tools, process, or judgment did it require?
- What changed afterward, even if you cannot quantify it?
Now the bullet can become:
- Rewrote onboarding emails for new trial users with product and support,
clarifying setup steps that repeatedly caused confused replies.
If you later verify the scale, add it:
- Rewrote 6 onboarding emails for new trial users with product and support,
clarifying setup steps behind recurring confused replies.
If you verify the outcome, add it:
- Rewrote 6 onboarding emails for new trial users with product and support,
reducing confused setup replies in the first week after signup.
Do not jump straight to:
- Increased onboarding conversion by 37%.
Unless that number is real, sourced, and actually connected to your work, it is not a stronger bullet. It is a liability.
A bullet formula that does not force fake numbers
Use this formula when you need a repeatable way to write bullets:
Action verb + specific work + context or constraint + truthful outcome
Examples:
| Weak bullet | Stronger bullet without fake metrics |
|---|---|
| "Worked on customer support issues." | "Triaged recurring setup issues from support tickets and turned the fixes into a shared onboarding checklist." |
| "Helped with migration." | "Mapped legacy account fields during a billing migration so engineering could preserve customer records across the new workflow." |
| "Managed social media." | "Planned weekly social posts for campus events, coordinating dates, speaker details, and last-minute room changes with student organizers." |
| "Improved documentation." | "Consolidated scattered API notes into one setup guide for support and implementation teammates." |
| "Collaborated with designers." | "Reviewed checkout edge cases with design and engineering before launch, catching missing error states in the final flow." |
The formula is close to University of Florida Career Connections Center's resume guide, which tells students to use action verb plus task plus result and answer what was accomplished, how it was done, and why it mattered.6 The University of Alabama Career Center gives the useful fallback: if you cannot name a specific result, highlight the skill demonstrated.7
The point is not to make every bullet long.
The point is to make every bullet carry evidence.
How AI can help without making things up
AI can help rewrite resume bullets safely when it is constrained to existing facts.
A good agent prompt is not "make this more impressive." That invites inflation.
Use a prompt like this:
Rewrite these resume bullets using only the facts provided.
Do not invent numbers, tools, employers, titles, dates, customers, metrics,
outcomes, or scope.
If a metric would strengthen the bullet, ask me what evidence exists.
Return before/after versions and mark unsupported claims.
Tiny CV's agent-safe workflow is built around that boundary. The agent can improve structure, ask for missing evidence, and suggest clearer phrasing, but the candidate owns the facts.
If the agent adds a number you did not provide, delete it or verify it before it reaches the resume. If it turns "helped" into "owned," check whether that is true. If it turns a team outcome into a solo claim, put the team context back.
For a fuller review protocol, use the safest way to let an AI agent edit your resume and keep your resume source of truth nearby.
The pre-send bullet audit
Audit resume bullets by checking whether each one is true, specific, relevant, and defensible.
Use this pass before sending:
- Does the bullet start with a concrete action?
- Does it include context the reader would not otherwise know?
- Does every number have a source?
- If there is no number, does the bullet still show scope, audience, constraint, or result?
- Is the claim relevant to the role?
- Could you explain the bullet in an interview without walking it back?
- Did AI add any unsupported detail?
- Does the strongest proof appear near the top of the role?
MIT recommends using the position description to decide what to include and selecting experiences that demonstrate the required skills.2 That applies inside each role, not just across the whole resume. Put the strongest role-matched bullets first.
If a bullet is true but not relevant, cut it or move it lower. If it is impressive but unsupported, rewrite it until it becomes defensible. If it is accurate but vague, add context before adding adjectives.
Strong bullets are calmer bullets
Strong resume bullets are not louder. They are clearer.
They do not need every achievement to become a percentage. They need the reader to understand the work, the setting, the constraint, and the result. Sometimes that includes a number. Sometimes it includes a team, an audience, a cadence, a before state, a tool, or a decision.
Use Tiny CV to keep the draft inspectable: write the bullet in markdown, keep evidence notes while drafting, preview the page, and export only the clean version.
The final question is not "Does this bullet sound impressive?"
The better question is "Can I defend every word?"
Footnotes
-
University of Nebraska-Lincoln Career Services, "Resume Bullet Statements," https://careers.unl.edu/job-internship-guide/chapter-1-resumes-references-cover-letters/resume-bullet-statements/ ↩
-
MIT Career Advising & Professional Development, "Resumes," https://capd.mit.edu/resources/resumes/ ↩ ↩2
-
Harvard FAS Mignone Center for Career Success, "Resume Template I (with bullet points)," PDF, https://cdn-careerservices.fas.harvard.edu/wp-content/uploads/sites/161/2024/07/resume-bullets.pdf ↩
-
UC San Diego Career Center, "Resume Guide," https://career.ucsd.edu/resources/guides/resume.html ↩ ↩2
-
National Association of Colleges and Employers, "What is Career Readiness?", https://www.naceweb.org/career-readiness/competencies/career-readiness-defined/ ↩
-
University of Florida Career Connections Center, "Resume Guide," https://careerhub.ufl.edu/resources/resume-guide/ ↩
-
University of Alabama Career Center, "Sample Resumes," https://career.sa.ua.edu/develop/resumes/sample-resumes/ ↩