The Resume Gap Standard, Seven Criteria Graded Against Your Dates of Record

This is a standard for putting an employment gap on a resume so it holds up with all four readers who will check it: the screener, the applicant tracking system (ATS, the software that parses and ranks applications), the interviewer, and the background check. It is for anyone with a gap of a few months to a few years, whether from a layoff, caregiving, illness, a sabbatical, or a change of field. It gives you seven pass-or-fail criteria, the evidence behind each one, a procedure that starts from your records rather than your memory, and a checklist to run before the resume goes out.

What counts as a passing resume gap?

A gap passes when it meets all seven criteria below. One failed criterion fails the gap, and one failed gap fails the resume. Each criterion settles a point where the popular advice contradicts itself, and each is written so that two reviewers grading the same entry give it the same grade.

Two terms come up throughout:

• Dates of record are the start and end dates your former employers' payroll and HR systems hold. They are what a background check sees, whatever your resume says.
• Employment verification is the check that confirms your job title, dates of employment, and employment status against those records.

The seven criteria:

1. Length threshold. Pass: each gap is measured in months from your records, and every gap of 8 months or more has a label entry. Fail: a long gap is left blank, or its length was estimated.
2. Date format. Pass: every entry uses the same format, either Mon YYYY or a year count. Fail: any year range such as "2021-2023", or any mix of formats.
3. Label. Pass: the entry names a recognised category, such as Career Break or Sabbatical. Fail: a placeholder like "N/A", "-", or a joke.
4. Disclosure. Pass: at most one line, naming the category only. Fail: a diagnosis, or a paragraph of justification.
5. Filler. Pass: every activity listed inside a gap is dated, verifiable, and real. Fail: vague "consulting" with no client or dates.
6. Cross-surface match. Pass: the months are identical on the resume, the public profile, and every application form. Fail: any month differs, or a finished break still shows as current.
7. Record match. Pass: every start and end month was copied from a record your former employer holds or reports. Fail: any date entered from memory.

Grading is binary on purpose. A "mostly fine" gap is the kind that gets flagged at the background check after you have resigned from something else or turned down another offer.

How long can a gap be before it needs explaining?

Eight months. A gap under 8 months can stand on its dates alone. A gap of 8 months or more needs a label entry, because the largest published US callback experiment found that most of the gap penalty arrives within the first eight months.

The study behind this is Kroft, Lange and Notowidigdo (NBER working paper w18387). They sent about 12,000 fictitious resumes to about 3,000 postings in sales, customer service, administrative support, and clerical roles across 100 US cities. The overall interview callback rate was about 4.7 percent. Callbacks, in the authors' words, "significantly decreases with the length of a worker's unemployment spell, with the majority of this decline occurring during the first eight months."

That shape settles the dispute over cutoffs. Because the penalty is front-loaded, a 14-month gap costs little more than a 9-month one. An 18-month cutoff fires long after the damage is done, and a 3-month cutoff has no curve behind it. A Swedish study (Eriksson and Rooth) found large callback drops after 9 months of unemployment for medium and low skill jobs, which points to the same region.

Here is how the main experiments compare:

How to measure a gap so two graders agree

Count the full calendar months between your last month of record at one employer and your first month of record at the next. A role that ended in March followed by one that started in December is a gap of 8 months, April through November. For a gap that is still open, count from the month after your last end month up to and including the current month.

Limits of the threshold

• Kroft measured current unemployment spells. A past gap that was followed by a job was not what they tested, so treat the threshold as the conservative choice for past gaps rather than a proven one.
• The roles tested were mostly in sales, service, and clerical work. The penalty for senior or specialist roles is not established by this study.
• Duration dependence was stronger when jobs were relatively plentiful. In a tight market for candidates, a long gap stands out more.
• The UK trial found no significant difference in callbacks between CVs with and without gaps, but it tested one 2.5-year profile. It does not license passing every gap.

Which date format holds up: month-year, years only, or year counts?

Use month and year (Mon YYYY) on every entry. The only gap-hiding format with callback evidence is a year count such as "HR Advisor - 5 years". Year ranges such as "2021-2023", which most guides recommend, were never tested and break some parsers.

The evidence comes from a preregistered UK field experiment (Kristal et al., with the Behavioural Insights Team) covering 9,022 applications. The control resume had an unexplained 2.5-year gap since the last job. One treatment explained the gap as childcare. Another removed dates altogether, so that "HR Advisor - 2013-2018" became "HR Advisor - 5 years." The year-count version raised positive callbacks by 4.8 percentage points (14.6%) over the gap resume and by about 8% over a resume with no gap at all. It did best for high-skill and full-time roles. Follow-up work suggested the count made total experience more noticeable, which is a different effect from hiding the gap.

On parsing, the public record is thin. One hands-on test of 8 ATS platforms found that year-only ranges made 4 of them flag the position as current or undated, and that one system calculated employment gaps from the dates it parsed. How parsers handle year counts with no dates at all is not established.

The decision rule:

• Applying through an ATS: Mon YYYY on every entry. The format is unambiguous to a parser, and the application form will ask for months anyway.
• Human-first channel (a referral, or a resume emailed straight to a hiring manager): year counts are allowed, but only if every entry uses them, and each count has to round correctly from your record months.
• Never: year ranges, or year-only dates next to the gap with month-year dates everywhere else. That mix confuses parsers, and the inconsistency itself draws a reader's eye to the gap.

So the answer to whether years-only dates hide a gap is no. The form that hides it was never a year range, and month-level dates still show up on the application form and in the background check.

How should a career break entry be labelled, and how much should it say?

Put a recognised category in the title or employer field, such as Career Break or Sabbatical. Add at most one line naming the reason category, and never name a diagnosis.

Label norms differ by country

In Refolk's index of professional profiles, the choice of label depends on the market:

In 25-profile samples from the same index, spelling variants of "Career Break" made up 21 of the 25 top titles in both the US and the UK. Treat the US sample with some caution, because 9 of its 25 profiles are in the San Francisco or Bay Area regions.

The employer field is where people improvise, and it shows:

These samples have only 25 profiles each, so read them as a sign that no standard exists rather than as market shares. The standard here is to put the break label in the employer field too. "N/A" and "-" read like a parsing accident, and "Self-employed" is filler that has to pass criterion 5.

For the reason category, borrow LinkedIn's career break vocabulary so your resume and profile use the same words:

• Bereavement
• Career transition
• Caregiving
• Full-time parenting
• Gap year
• Layoff/position eliminated
• Health and well-being
• Personal goal pursuit
• Professional development

How much to disclose

Whether a reason helps depends on what it signals about your ability. In a US audit of 3,771 applications (Namingit, Blankenau and Schwab), applicants whose long gap was explained by cancer got fewer callbacks than the newly unemployed, but significantly more than applicants with an unexplained gap. A secondary source reports the rates as 27.4%, 25.6%, and 23.3%. I have not confirmed those figures against the paper, so check them before you quote them. In the UK trial, a childcare explanation made no difference compared with no explanation. HR professionals there said they preferred an explanation at interview, but that preference did not show up in how they actually recruited.

So a one-line category costs little and sometimes helps. A detailed justification buys nothing the evidence can see.

Health is the one category where the detail does harm. Under US EEOC guidance on pre-employment disability-related questions, issued in 1995, employers cannot ask disability-related questions or require medical exams until the applicant has a conditional job offer. Before that point they may ask about your ability to do specific job functions and about non-medical qualifications such as work history. If you volunteer a disability, the employer may ask limited questions about reasonable accommodation. Naming a diagnosis on a resume opens a door the law otherwise keeps shut. That guidance does not cover caregiving, so a caregiving label carries no equivalent risk.

Career Break | [Mon YYYY] - [Mon YYYY]
Full-time caregiving for a family member. Available for full-time work from [Mon YYYY].

Career Break - Health and well-being | [Mon YYYY] - [Mon YYYY]
Available for full-time work from [Mon YYYY].

Career Transition | [Mon YYYY] - [Mon YYYY]
[Course or certificate name], [issuing body], completed [Mon YYYY].

Sabbatical | [Mon YYYY] - [Mon YYYY]
[One dated, verifiable activity, or leave this line out.]

Year-count variant (human-first channels only, every entry):
[Job title], [Employer] - [N] years

What activity counts as legitimate gap filler?

An activity counts only if it is dated to the month, someone other than you can confirm it, and it happened at the scale the line implies. Anything else is padding that fails verification and invites probing at interview.

Grade each filler entry on three tests:

• Dated. It has a start month and an end month, or a completion month, in the same format as the rest of the resume. What it proves: the activity fits inside the gap. What it looks like when it lies: dates that conveniently cover the whole gap to the month.
• Verifiable. A client, institution, certificate, or public artifact exists that a reference or verifier could check. What it proves: that it happened. What it looks like when it lies: "Independent consultant" with no client, no deliverable, and no way to find either.
• Real at the scale stated. One short project is not "consulting practice", and one online module is not "retraining". What it proves: you will not be caught out by the first follow-up question. What it looks like when it lies: a line that sounds like a job but falls apart after two questions at interview.

The bottom-right quadrant is the one people get wrong. A long gap with nothing verifiable inside it passes cleanly with an honest label. It fails the moment someone invents a consultancy to fill it.

Do your gap dates match what your former employers hold?

They have to match to the month. The only way to know they do is to pull the records the background check will read, because an honest date from memory can still be a payroll month off.

When a former employer's records do not match what you reported, the discrepancy is flagged for the hiring employer to review. One staffing firm describes how this happens to honest candidates (it is a single unverified account, but the mechanism is plausible):

How much mismatch gets through is not standardised. One screening firm says employers may allow a two- to three-month discrepancy on employment dates, while diploma dates usually have to be exact. It also says each employer sets its own standard by industry and position, and no industry-wide threshold has been published. You cannot know which employer you are dealing with, so the target is zero.

Where to get your dates of record

The Work Number holds more than 823 million employee records from nearly 5 million employers, and its data updates once per payroll cycle, typically every 1 to 2 weeks. Whether it returns day-level or month-level dates is not established publicly, so read your own report rather than assuming. The IRS and Social Security records only give years, so use them to confirm which employers fall in which years, and use your report or your own offer letters and final pay stubs for the months. If an employer does not appear in your report at all, that employer's HR or payroll team is the record holder. Ask them what dates they would confirm.

Copy your dates from the record your next employer will read, not from the memory you trust.

The same months on every surface

Each inner layer is harder to change than the one outside it, so the record sets the months and every other surface copies them. On LinkedIn, a career break entry is visible to logged-in members, and you can leave its break type blank. The common failure is a break that has ended on the resume but is still open as the current role on the profile. One source claims some recruiter searches on current title exclude Career Break entries. That is unverified, but a mismatch fails criterion 6 whether or not it is true.

Tailoring is where matching dates drift apart. Ten versions of a resume edited by hand will eventually disagree by a month somewhere. Refolk writes each tailored resume from one stored history, so once your record months are in, they carry into every version unchanged.

The grading procedure, from records to parse test

Run these eight steps in order for every resume version you plan to send. Records come first, before any formatting decision, because a payroll-month shift can move a gap across the 8-month threshold.

Most practitioner advice starts with formatting. I put records first because the payroll-date mismatch is a failure that formatting cannot fix. Building the timeline takes under an hour once the records are in hand (that is my estimate). The Social Security statement is the slow record, so if you need it, request it before anything else.

Use this worksheet to hold the timeline while you grade:

EMPLOYERS
Employer | Start (record) | End (record) | Source | Resume | Profile | Form | All match?

GAPS
From (Mon YYYY) | To (Mon YYYY) | Months | 8+? | Label | Disclosure is category only? | Filler dated / verifiable / real? | Pass?

How gap grading goes wrong

Most failures are false passes: a gap gets graded as fine because one piece of evidence was read too broadly, or because a date "felt" right. These eight are the ones to check for, each with its test.

1. Treating year ranges as the tested format. "2021-2023" gets graded as a pass because a guide cited the 15% result. The experiment used counts ("5 years"). Year ranges are untested for callbacks and risk parse errors. Check: the format is either a count or Mon YYYY, never a mix.
2. Mixing formats. Year-only dates next to the gap, month-year everywhere else. This confuses parsers and draws the eye to the gap. Check: every entry follows the same pattern.
3. Dates from memory. The resume, profile, and form all agree with each other, so the gap looks matched, but all three are a payroll month off the record. Check: compare the resume against the report the employer will actually see.
4. Passing every gap on the UK result. The UK trial found no significant gap penalty, but it tested one 2.5-year profile. Check: for US roles, apply the Kroft 8-month curve.
5. Putting a diagnosis on the resume. This volunteers information that pre-offer law otherwise protects. Check: the label names the category only.
6. Vague filler. "Consulting" with no client and no dates fails verification and invites probing. Check: each filler entry can be proven by someone other than you.
7. A finished break left as the current role on the profile. The resume says you returned to work, but the profile still shows you on the break. Check: the end dates on the profile match the resume.
8. Citing ATS gap thresholds as fact. A resume-analyzer vendor blog says Greenhouse "highlights gaps >= 3 months" and Workday "flags gaps > 6 months if rule is active". No vendor documentation confirms either, and another source says most platforms do not flag gaps at all. Check: treat any product-specific threshold as unverified and grade against the parse test, not the claim.

Failures 3 and 7 are the expensive ones, because they surface late. The screener and the interviewer never notice a one-month difference. The verification report does, after the offer, when you have the most to lose.

Keeping the gap entry current before each application

Re-grade whenever an input changes: a new record arrives, you start or finish an activity inside the gap, you edit your profile, or a new application form asks for dates in a different shape. Each of these can break a criterion that passed last week.

An open gap also changes over time. A gap that was 6 months when you wrote the resume reaches 8 months while you are still applying, and from then on it needs a label entry. Recount open gaps at the start of each week you are actively applying. If the one-line explanation would read better in prose, put it in the cover letter rather than the resume. Refolk drafts the cover letter from the same history, so the category and months it uses match the resume.

When every box is ticked, the gap is graded and the resume can go out. Keep the worksheet with your search files. The same record months will be needed again for the employment history form and the post-offer verification.