The usual advice for tailoring a resume is to copy the job post's keywords into it. That helps with search filters, but it misses what a recruiter is actually doing in the few seconds they spend on your resume: looking for proof that you can do what the job asks.
A better method is to treat the job post as a checklist and ask, for every line, "where on my resume does a reader see evidence of this?"
Step 1: Turn the job post into a list of requirements
Most job posts have three kinds of content:
- Requirements or qualifications: what they expect you to already have
- Nice-to-haves: preferred, bonus or "plus" items
- Responsibilities: what you'll do in the job
Start with the requirements. Responsibilities matter, but they describe the job rather than the candidate, and most of them are covered if you prove the requirements.
Write each requirement on its own line. Break compound ones apart: "SQL and Python for data cleaning" is two checks, not one.
Step 2: Find the best evidence for each one
For every requirement, find the single line on your resume that best supports it, then rate it honestly:
| Rating |
What it looks like |
Example |
| Proven |
A bullet that names the skill and shows what you did or achieved with it |
"Ran 15 A/B tests on checkout pages, lifting conversion 3.2%" |
| Mentioned |
The skill appears, but with nothing to back it up |
"Skills: Python, pandas, SQL" |
| Missing |
Nowhere on the page |
(nothing about e-commerce) |
The middle category is the one most people underestimate. A skills list passes a keyword filter, but it doesn't answer the recruiter's real question. Someone who learned Python from one weekend tutorial can write "Python" just as easily as someone who has used it daily for five years.
Step 3: Turn "mentioned" into "proven"
For each mentioned skill, find a real piece of work where you used it, and write a bullet in this shape:
action verb + what you did + the skill or tool + the result
Some before-and-after examples:
| Before |
After |
| Skills: Excel |
Built an Excel model of 400 products' margins that the buying team now uses for every reorder |
| Responsible for customer emails |
Answered about 60 customer emails a day in Zendesk, keeping first-reply time under 4 hours |
| Python, pandas |
Cleaned 2 years of sales data in Python (pandas), removing 8% duplicate orders before the annual forecast |
| Team player, good communicator |
Presented monthly results to the executive team and turned their questions into the next quarter's tests |
Numbers aren't compulsory, but they make a claim concrete. When you don't have an exact figure, a scale ("about 60 a day", "a 12-person team", "every reorder") still helps.
Step 4: Decide what to do about "missing"
Missing requirements need an honest decision:
- You have the experience but left it out. Add it, with evidence. This is the most common case: people trim old roles or side projects that would cover the gap.
- You have something close. Describe the related experience in the job's terms if it is accurate. A checkout-page testing project is e-commerce experience; say so.
- You don't have it. Leave it off. Mention a related strength or how you'd close the gap in your cover letter instead. Don't add skills you don't have: interviews and reference checks find them, and one false claim can undo the rest of an application.
Prioritise required items over nice-to-haves. Most posts list more nice-to-haves than anyone fully meets.
Step 5: Check years-of-experience lines
"4+ years of experience in data analysis" is usually judged from the dates on your resume. Make sure:
- Every role has start and end dates (month and year is ideal)
- Your job titles or bullets make the field obvious, so a "Junior Analyst" role counts towards "data analysis"
- Overlapping roles aren't double-counted in your own head; recruiters won't count them twice
Step 6: Use the job's own words where they're true
Now the keyword advice comes in, as a final pass. If the post says "stakeholder management" and your resume says "worked with other departments", use their phrase in a bullet that describes the real work. Applicant tracking systems and recruiters both search for the post's wording, and matching it costs nothing when the claim is accurate.
A quick way to run the check
Doing this by hand takes 20 to 30 minutes per application. The resume vs job description checker does the first pass in your browser: it splits the job post into requirements, finds your best evidence line for each, and labels it proven, mentioned or missing. Neither document is uploaded.
Use it to find gaps, then make the judgment calls yourself: the tool can see which words match, but only you know what you have actually done.