Our published standard
How we check — all twelve points, on every order
Here is exactly what we check, published in full. Every audit and every full optimization runs against the same twelve checkpoints below — no sampling, no "premium" rubric. If a finding appears in your report, it maps to one of these, with evidence from your own document. We publish this because we think you should be able to see the standard before you pay anyone — including us.
Machine-checkable
Run automatically on every order — the structural facts a parser either survives or doesn't.
1. Parse safety and structural hazards
AutomatedWe run your document through parsing analysis to find the structures that commonly break automated resume readers: two-column layouts, tables, text boxes, content in headers or footers, text rendered as images, non-standard fonts and embedded objects.
2. File format and contact block
AutomatedFile type and naming, contact block position and parseability, a real email address and phone number present, link integrity, and photo or date-of-birth presence checked against your region's conventions.
3. Section architecture
AutomatedPresence, naming and ordering of the standard sections; where your skills sit relative to the first-page fold; and any non-standard headings automated parsers will not map to a field.
Analysis with human verification
Automated analysis drafts these; nothing reaches a deliverable until a named human reviewer has verified it against your actual document.
4. Keyword coverage against your target title
AutomatedRequired and preferred terms for your exact target job title from our own screening research, your coverage percentage, and the explicit list of terms your document is missing.
5. Date and chronology integrity
AutomatedDate-format consistency, unexplained gaps over three months, overlapping roles, and tense consistency. Gaps are reported as facts — we never speculate on their cause.
6. Length and density
AutomatedPage count against your region's norm, words per role, bullet-count distribution, white-space ratio, and how much of your strongest material sits on the first page.
7. Achievement quality
AI-assisted · human-verifiedThe ratio of responsibility statements to result statements, the share of bullets carrying a number, and whether scope indicators (team size, budget, volume) are present where your experience supports them.
8. Title and seniority alignment
AI-assisted · human-verifiedYour titles against the conventions of the target title in your industry and region, and whether the framing reads at the right level — neither inflated nor undersold.
9. Language distinctiveness
AI-assisted · human-verifiedHow much of your document reads like the documents recruiters see every day: AI-cadence phrasing density, cliché bullet openers, and template wording. We hold our own writing to the same standard before anything ships.
10. Regional and dialect fit
AI-assisted · human-verifiedYour document against the conventions of your target market — spelling family, document naming, length norm, photo and personal-details rules, referee conventions and date formats.
Human judgement, required
No automation exists for these, deliberately. They are the reason a person reads every order.
11. Target realism
Human-onlyAn experienced reviewer's honest view: is the stated target plausible given this history — and if not, what is the adjacent target that is? Counsel, not extraction.
12. Narrative coherence
Human-onlyDoes the document tell a story a recruiter can follow in eight seconds? Does the career arc read as deliberate or accidental?
How severity is graded
Every finding carries one of three fixed severities — the definitions are part of the standard:
- critical
- Likely to cause a parse failure or an automatic screen-out
- major
- Will not break parsing, but materially reduces how the document ranks or reads
- minor
- Polish; worth fixing, unlikely on its own to change an outcome
Two promises this standard makes
Facts, never speculation. Checkpoint 5 reports employment gaps as facts — where and how long. We never guess at, infer, or comment on the reasons for a gap, or on anyone's health, family, immigration status or personal circumstances. That rule is enforced in our tooling with no override.
Our own writing passes checkpoint 9. We check your document for template phrasing — so our deliverables are held to the same distinctiveness threshold before anything ships.