Is this the same person

It is the same invented candidate, Ada Fennimore, who does not exist. The same work is described in both documents, written two different ways.

What were the two numbers

Measured on 26 September: 100 × 30% + 48 × 30% + 46 × 40% = 63 for the weak version, and 100 × 30% + 100 × 30% + 71 × 40% = 88 for the rewritten one. On 13 September, under the rules of that day, the same two documents measured 40 and 97.

What actually changed

The work was moved out of a projects list into a real experience section with roles, dates and achievement lines, and an education section, a summary and a phone number were added.

No. It is a legible, complete, evidenced document. The specimen took four minutes to write, and the points it still lacks are depth and volume: more achievement lines and more words.

Will my document move that far

That depends entirely on where it starts. The mechanism is the same: find what the scorer could not see, and put it where it can.

Blog · Walkthroughs

The same work, written twice: 63 and 88 in the free ATS checker

The distance between a weak score and a strong one is usually not more work. It is the same work, written where the scorer can see it.

ConxfolioPublished 2026-09-13 · updated 2026-09-26

— the career did not change; the document did.

The score card for the weak document: 100 × 30% + 48 × 30% + 46 × 40% = 63 — The before, measured on 26 September. Parseability held up; completeness and content did not.

Re-measured on 26 September 2026. This page was first measured on 13 September, when the two documents scored 40 and 97. The checker then did not read lines under a Projects heading, so the weak version's content was 0; that was a fault in the checker and it has been fixed, and on 25 September the content rules moved to grading in proportion. The numbers and pictures below are from 26 September, and the distance between the two versions is now 25 points rather than 57.

Two documents were measured for this page. They describe the same invented career: a backend engineer, the same settlement pipeline, the same database cost, the same invoicing API. Nothing about the work changed between them.

The before, and its sum

100 × 30% + 48 × 30% + 46 × 40% = 63, from 48 readable words on 10 lines. Parseability is perfect: the page reads cleanly. Everything else is thin.

Why the number was that low

The ATS review report for the weak document, with the extracted text beside the score and the checks marked Passed or Review — The whole report. Each check marked Review says what would earn its points back.

The work was written up as two project names with a few lines under each. The checker now reads project lines as experience evidence, so that half counts, but not as employment, and neither project carries a date of its own. There was no education section, no phone number and no summary. On content, three achievement lines earned 6 of the 25 depth points, two of them carry a figure, and 48 words earned 2 of the 20 volume points.

What was done about it

The same facts were rewritten into a real experience section: two roles with employers and date ranges, four and three achievement lines under them, each beginning with what was done. An education line, a phone number and a two-sentence summary were added. No new work was invented, because there was none to invent.

The after

The score card after the rewrite: 100 × 30% + 100 × 30% + 71 × 40% = 88 — The rewritten document's score card. 162 readable words.

100 × 30% + 100 × 30% + 71 × 40% = 88. Every parseability and completeness check passes. On content, achievement quality and measurable evidence are at or near full marks; what is left is depth, seven achievement lines where the full score asks for twelve, and volume, 162 words where it asks for 500. Those are honest requests for more of the same kind of line, which is the right last thing.

The lesson, which is not about the number

The ATS review's first screen: upload a résumé or paste the text — Run the check, make one change, run it again. Reports are kept under Previous checks so the pair can be compared.

Most of the 25 points came from two structural facts: the work was not filed as employment with dates, and whole sections were missing. Both are writing decisions rather than career ones. That is the general case, and it is why a low score is a document diagnosis rather than a verdict, as what a score cannot tell you argues.

The general pattern behind both numbers

Three dimensions with different failure modes. Parseability fails on layout and is repaired with structure. Completeness fails on missing sections and is repaired by adding them. Content fails on how lines are written and is repaired by rewriting. The weak document failed the second and third at once, and the largest single gain came from sections the rules could not find: a dated role, an education line, a way to reach the candidate. The prose barely changed.

What this page does not claim

That every document has 25 points in it, or that a high score changes an outcome. The candidate is invented, the résumé took four minutes, and nobody was hired because nobody exists. What the pair demonstrates is mechanical: the scorer measures what it can see, and the commonest reason it sees very little is that real work was filed somewhere the rules do not look.

If you want the cheapest version of this experiment on your own document, it is two runs and one change. Run the checker, take the single check marked Review that is worth the most points, do only that, and run it again. Previous checks keeps the earlier run on this device, so the difference is exactly what the change was worth. That habit is described in what a fix was worth, and the rules underneath in evidence ceilings explained.

A final word on what these two numbers are for. Neither is a target. The useful quantity is the distance between them, because it demonstrates that a score measures a document rather than a person, and that a low one is a diagnosis with an address. Anybody looking at 63 on their own report is looking at a document whose evidence did not reach the rules, and the report names which rules and what each would pay. That is a much better position than it feels like at the moment the number appears.

Run it, fix one thing, run it again

Reports are kept under Previous checks, so the second run can be compared with the first. That difference is what the fix was worth.

Open the ATS checker What changed after one fix

Questions

Is this the same person: It is the same invented candidate, Ada Fennimore, who does not exist. The same work is described in both documents, written two different ways.
What were the two numbers: Measured on 26 September: 100 × 30% + 48 × 30% + 46 × 40% = 63 for the weak version, and 100 × 30% + 100 × 30% + 71 × 40% = 88 for the rewritten one. On 13 September, under the rules of that day, the same two documents measured 40 and 97.
What actually changed: The work was moved out of a projects list into a real experience section with roles, dates and achievement lines, and an education section, a summary and a phone number were added.
Is 88 a good career: No. It is a legible, complete, evidenced document. The specimen took four minutes to write, and the points it still lacks are depth and volume: more achievement lines and more words.
Will my document move that far: That depends entirely on where it starts. The mechanism is the same: find what the scorer could not see, and put it where it can.

Conxfolio is a free set of four career tools: a résumé builder with 37 rendered layouts, an ATS résumé checker that prints its own arithmetic, a cover letter builder that traces every proof paragraph back to the line it came from, and a portfolio builder with 22 authored designs. There is no account to create, nothing is held back for a paid plan, and no language model is used anywhere in the product, so the readers, the score and the letter are deterministic code you can check.