Resume Analysis
Reading every application against the same role, with the same care, in the order they arrive. Straightforward in principle and full of edge cases in practice.
The short answer
- Analysis reads a resume against a specific posting and reports the evidence it found, rather than judging the resume in isolation.
- Parsing is the weak link. A resume built in columns or text boxes can arrive scrambled, and a candidate penalised for that has been penalised for their word processor.
- It reads what is written. Capability a candidate did not describe does not register, which affects career changers most.
- Use the output as a reading aid. The evidence it surfaces is more useful than any single number it produces.
What the analysis does
It reads the resume and the posting together, then reports what it found: relevant experience, capabilities the role called for, requirements met, and where the evidence is thin.
That framing matters. The output is not a verdict on the resume, it is a report on how this resume answers this role. The same document produces a different report against a different posting, which is why comparing scores across roles is meaningless.
The most useful part of the output is usually the specifics rather than the summary number: which requirements it found evidence for and which it did not. That is what makes it a reading aid rather than a filter.
Parsing is where it fails
Before anything can be assessed, the file has to be read as text. Multi-column layouts, text boxes, tables used for layout, contact details inside a document header, and graphics carrying information all cause content to arrive out of order or not at all.
This is worth knowing because the failure is silent and it penalises the wrong thing. A candidate whose resume did not parse is not a weaker candidate; they used a template with columns. If an application looks inexplicably thin against a strong-sounding background, open the original file before drawing a conclusion.
It is also a reason to accept common formats and to avoid demanding a rigid template, which pushes the problem onto candidates who cannot know how their file will be read.
What it will miss
Unstated capability. Anything the candidate did not write down does not register. Human reviewers sometimes infer charitably from surrounding evidence; automated analysis is more literal.
Context behind an unusual path. A career change reads as missing experience unless the resume itself makes the connection, so career changers are systematically underserved by this kind of review.
Trajectory as a person would read it. Rapid growth in scope is visible to an experienced reviewer in a way that is hard to capture in a comparison against requirements.
All three argue for the same practice: read the top of the ranked list properly, and sample below it.
Using it well
Treat the report as the notes a thorough reviewer would have made, then do the judging yourself. Specifically: use the evidence it surfaced to decide what to look for in the resume, rather than accepting the summary as the answer.
Where it flags a requirement as unmet, check whether the requirement was one you actually meant. Analysis is faithful to the posting, so a wish-list item written casually is being enforced strictly, and that is usually a posting problem rather than a candidate one.
Frequently Asked Questions
Employer Learning Center
On GigFinder
From the candidate side
Review with the reading already done
Each application arrives read against your posting, with the evidence surfaced rather than buried.