The Resume Standard, and the Criteria a Good One Must Pass

This guide gives you a standard for your own resume: a set of criteria stated plainly enough that two people grading the same document would reach the same verdict. It is for anyone writing or rewriting a resume who is tired of vague advice like "make it pop" and wants a definition of done. By the end you will have a checklist you can adopt as personal policy and apply to every version you send.

Most resume advice argues about taste. A standard does not. It names what must be present, what must be true, and what fails the document outright, so that grading is a matter of checking boxes rather than debating. I have built this one around the numbers that actually move a hiring decision and flagged, honestly, where the evidence is thin.

What a resume standard is, and why grade instead of edit

A resume standard is a fixed set of pass-or-fail criteria applied to a finished document, not a style opinion. The point of grading rather than editing is that editing has no stopping condition, while grading tells you when the resume is done.

Two graders who share this standard should reach the same verdict on the same resume. That property is what separates a standard from a preference. When a criterion cannot be graded that reliably - length, for example, where the evidence pulls two ways - I say so and give you the rule to apply anyway.

The other reason to grade is priority. Not every criterion carries equal weight. A typo can auto-reject a resume that is otherwise excellent, while a debatable choice about summary placement almost never does. A good standard sorts criteria by how much they cost you when you fail them, and this one does.

A standard tells you when the resume is done. Editing never does.

The seven-second window, and what it actually gates

The first-scan window is short but it is not the only gate, so design for it without believing it is the whole story. TheLadders 2018 eye-tracking study of 30 recruiters over a ten-week period found an average initial scan of 7.4 seconds, up from six seconds in 2012.

That figure is disputed, and correctly so. Some recruiters argue real screening time varies by role complexity, application volume, and stage. A first-pass screen is not a hiring manager's deep read. So treat 7.4 seconds as the budget for one specific thing: the top third of page one, the zone a recruiter's eye lands on before deciding whether to keep reading.

The study also documented a consistent visual path. Recruiters fixate, in order, on:

• Current job title
• Current company
• Start and end dates of the current role
• Previous title and company
• Education

Your standard follows from that path. If a stranger reading only the top third of your first page cannot name your current role, employer, and single strongest result, the document fails the first-scan criterion. Everything below that zone is for the reader who has already decided you are worth more than seven seconds.

The required sections, in the required order

A resume must contain a header with contact information, work experience, education, and skills, listed in reverse-chronological order, with a summary added if you have real experience. There is strong cross-source agreement on this core, and disagreement only at the edges.

Reverse-chronological means newest first, and it is the standard, recruiter-preferred format in the US. Bullets should lead with action verbs and quantify results. The one genuine ordering choice is where skills and education sit relative to experience.

The rule for the edge case: keep Skills after Experience for experienced candidates, but move Education and Skills higher for new graduates and career-changers, who have more to prove in those sections than in their work history. This is not a style call. It is about placing your strongest evidence inside the first-scan zone.

Grading a bullet: verb, metric, defensible result

A bullet passes when it opens with an action verb, describes a result rather than a duty, and - for your strongest lines - carries a number you can defend in an interview. A bullet that begins "responsible for" fails on sight.

The most useful shape is the one Google's former SVP of People Operations, Laszlo Bock, popularized: accomplished X as measured by Y, by doing Z. University career centers teach the same idea as CAR - Context, Action, Result. Both force the same discipline: name the outcome, attach evidence, then explain the method.

Fail:  Responsible for managing the onboarding process for new users.
Pass:  Cut new-user onboarding time from 12 days to 3 by rebuilding the setup flow and automating account provisioning.

Formula: Accomplished [X] as measured by [Y], by doing [Z].
Test:    (1) opens with an action verb  (2) states a result, not a duty  (3) carries a number you can explain out loud.

Metrics matter because they are rare. If only about 8% of resumes contain any metrics - a figure I flag as unverified at its original source but consistent with what recruiters report - then a quantified bullet differentiates on scarcity alone. That scarcity is the honest explanation for the widely repeated claim that quantified resumes are 40% more likely to result in an interview. That percentage traces to a single vendor, TalentWorks, and is echoed uncritically everywhere, so cite it as single-sourced. The mechanism is sound even where the exact number is not verifiable: numbers are evidence, and evidence is rare on the page.

Sources disagree on how many bullets should carry a number. One recommends at least half of all bullets contain a metric; another suggests three to five bullets per role. Grade to the stricter reachable target: every role's single best bullet must be quantified, and you should be reaching for more.

Setting length by career stage

Length is graded against career stage, not against a universal one-page rule. The main empirical study inverts the folk wisdom: out of 7,712 resumes in a simulated hiring process, 5,375 were two pages, making recruiters 2.3 times as likely to prefer two-page resumes.

That preference is not uniform. It scales with seniority, which is exactly what you would expect if the mechanism is "more relevant evidence earns more attention" rather than "longer is better."

Recruiters in the same study spent 2 minutes 24 seconds on one-page resumes and 4 minutes 5 seconds on two-page resumes. The longer read is attention, not fatigue - but only when the second page holds relevant content. A practitioner rule that still holds as a sanity check: roughly one page for every ten years of experience.

So the grading rule is a hinge, not a hard number. Under about ten years of experience, default to one page. Senior and managerial roles justify two. Use a longer CV format only for academic, research, or international roles; keep the private-sector corporate application tight.

Which skills to name, and why scarcity beats polish

Name the skills that are scarce in your target market, because a keyword only differentiates when the underlying pool is thin. How you phrase a skill matters far less than whether naming it sets you apart from the field.

This is where Refolk's index makes the mechanism concrete. Consider two skills in one market.

In Refolk's index, a US Software Engineer who lists Rust belongs to a pool of 619, roughly 90.5 times rarer than the 56,045 who list Python. Listing Python proves competence; it does not distinguish you. Listing a genuinely scarce, in-demand skill you actually have does both. The lesson for grading: audit which skills you name against how common they are, not just whether they are true.

Market depth also sets the bar you are graded against. In Refolk's index there are 354,065 Software Engineers in the United States versus 44,302 in the United Kingdom.

The US pool is about eight times deeper. A US applicant is graded against a far larger field in the same short scan, which raises the practical bar for how much a resume must differentiate. The deeper your market, the harder your scarce skills and defensible metrics have to work.

If you want to see the field you are actually competing against before you decide which skills to feature, Refolk can search live professional profiles by title, skill, and location. Seeing how thin or crowded a skill pool is tells you which lines on your resume are doing real work.

The grading procedure, step by step

Grade in a fixed order so nothing is skipped and each pass has a clear done condition. The sequence below moves from structure to content to format to targeting, which is the order in which failures compound.

The error pass deserves its position near the end and its fifteen-minute budget. It catches the single most common auto-reject, and reading aloud finds errors your eye skips silently.

The format and ATS pass, without the mythology

Format to parse cleanly and to avoid auto-rejects, but do not build your resume around a rejection rate that does not exist. Applicant tracking systems are near-universal - Jobscan detected an ATS on 97.8% of Fortune 500 career pages in 2025 - but their job is to parse and rank, not to guarantee rejection.

The most widely deployed systems in 2025 were Workday at a 39.2% usage rate and SuccessFactors at 13.2%, together covering a combined 52.4% of Fortune 500 companies. The often-repeated claim that 75% of resumes are rejected by ATS has no study behind it. Do not game keyword density against a filter that may not auto-reject at all.

What the format pass should actually enforce:

• Contact details in the document body, not a header or footer some parsers drop
• A clean PDF, or the Word format the posting requests
• Standard fonts and one-inch margins
• No text boxes, tables-as-layout, or multi-column designs that scramble parsing
• No graphics placed where they obstruct the reading order
• Exact employment dates, since one CareerBuilder survey found 27% of employers auto-dismiss resumes without them

A reliable test: copy the whole file and paste it into a plain-text editor. If your name, contacts, and section order survive intact, the machine can read it.

How this goes wrong: failure modes and false positives

The most valuable part of a standard is knowing where it lies to you. Each criterion below can produce a false positive - a resume that looks like it passed but did not - and each has a check that catches the lie.

The 7.4-second rule taken as universal. Designing only for a seven-second glance ignores that the real gate is often an ATS keyword match or a hiring manager's deep read. Check which stage you are optimizing for; the first-pass screen and the deep review are different jobs.

"75% rejected by ATS" cited as fact. This drives people to stuff keywords against a filter that may not reject them. The claim is unsupported by any study. Treat ATS as parsing and ranking, and format to be readable rather than to beat an imaginary quota.

"40% more interviews from metrics" treated as hard science. It traces to one vendor and is repeated uncritically. Cite it as single-sourced. Quantify your bullets anyway, because the mechanism - numbers as evidence, rare enough to stand out - holds even if the exact figure does not.

The two-page license used to pad. A padded second page reads worse than a tight single page. Check every line on page two against relevance; if it is an outdated job or a minor task, cut it.

Metrics that cannot be defended. Inventing numbers fails the moment an interviewer probes. Check that you can explain every figure aloud; hedge with "approximately" or a range when you are unsure.

Contact info in the header or footer. It looks fine on screen and vanishes in some parsers. Check that name, phone, and email sit in the document body.

The typo stat cited as a floating figure. The CareerBuilder number appears as 77%, 61%, 60%, and 58% across different years and write-ups. Attribute it to a specific survey rather than a generic "CareerBuilder says," and do not let the version drift distract from the underlying point: it is a controllable, high-cost failure.

That last point deserves emphasis. With 77% of employers auto-screening typos while 58% of resumes contain them, the majority of applicants eliminate themselves on a fully controllable criterion. Proofreading is the highest-return item on this entire standard, which is why the error pass is a graded step and not an afterthought.

The pass-or-fail checklist

Run this before you send. Every item is a checkable statement, not a topic, and the resume is done only when all of them pass.

Keeping the standard current

A resume standard drifts because the market moves, not because the criteria do. Re-run the tailoring and skills-scarcity steps for each posting, and re-check the load-bearing statistics against their sources rather than trusting a cached number.

Two things genuinely change over time. First, which skills are scarce in your market: the Rust-versus-Python gap is real today, but the differentiating skills in any field shift, so re-check pool depth when you retarget. Second, the ATS landscape: usage rates and dominant vendors are published periodically, so if you are optimizing format, confirm the current figures rather than relying on the ones here. The mechanism - parse cleanly, name scarce skills, quantify defensibly, and never fail the typo gate - does not expire. When you want to see how crowded your target field actually is before you decide what to feature, Refolk will show you the live pool by title, skill, and location, so your grading reflects the competition you are actually facing.