We Analyzed 10,000 Resumes. Here's What We Found.
Working across hiring teams and candidates on the APTI Careers platform gives us a fairly unusual vantage point. APTI Careers is an AI-powered hiring and interview platform built to help recruiting teams evaluate candidates faster and more fairly than a resume alone allows. We see resumes at scale, across industries, alongside how those resumes actually performed once real hiring decisions got made. We pulled together patterns from a large sample of resumes reviewed through our platform to see what actually separated resumes that moved candidates forward from resumes that did not. Some of what we found confirmed common advice. Some of it did not.
Length mattered less than expected, up to a point
One of the most repeated pieces of resume advice is to keep it to one page. Looking at the data, resume length on its own showed almost no meaningful relationship with how far a candidate advanced, as long as the resume stayed under roughly two pages. One page resumes and two page resumes performed similarly overall. What did show a clear pattern was resumes over three pages, which consistently underperformed, almost certainly because length at that point usually reflects a lack of editing rather than a genuinely longer relevant career history.
The takeaway is not that length does not matter at all. It is that the common one page rule is more rigid than the data actually supports, and candidates with legitimately extensive relevant experience should not feel forced into an artificially compressed single page if it means cutting real, relevant detail.
Quantified achievements showed a real, measurable advantage
This is one area where common advice held up clearly. Resumes with specific numbers attached to accomplishments, percentages, dollar figures, counts, timeframes, consistently advanced further in the process than resumes describing similar work in purely qualitative terms. The advantage was not small. Resumes with at least a few quantified achievements moved to the interview stage noticeably more often than resumes describing comparable experience without any numbers at all.
Interestingly, the advantage was not just about the numbers themselves impressing a reader. It also showed up in how these candidates performed in the interviews that followed. Candidates who described their prior work in specific, measurable terms also tended to answer interview questions with more specific, concrete detail rather than vague generalities, suggesting the habit of thinking in specifics carries through to how someone communicates in general, not just how they write a resume.
"Specificity, genuine tailoring, and honest framing of real experience consistently outperform surface-level polish and generic completeness."
Tailored resumes outperformed generic ones significantly
Resumes clearly customized for the specific role, evident from language that closely mirrored the actual responsibilities and requirements of the posting rather than reading as generic, advanced at meaningfully higher rates than resumes that appeared to be sent out broadly without much adjustment. This pattern was consistent across every industry we looked at, though it was especially pronounced for mid level and senior roles, where hiring teams tend to expect a higher degree of intentionality from applicants.
What stood out here is how easy this signal was for both human recruiters and our AI-conducted screening to detect. A generic resume tends to describe skills and experience in broad, transferable terms that could apply to many different roles. A tailored resume tends to lead with the specific parts of a candidate's background that connect most directly to what the posting is asking for. That difference is noticeable almost immediately, even without any deep analysis.
Career gaps hurt far less than candidates assume
A significant number of candidates in the sample had employment gaps of six months or longer, and contrary to what a lot of job seekers fear, an unexplained gap alone showed only a small negative effect on advancement, and a gap accompanied by even a brief explanation showed almost no negative effect at all compared to resumes with no gaps.
This matched what we consistently hear from recruiters directly. Gaps have become common enough, particularly since the disruptions of recent years, that most experienced recruiters no longer treat them as an automatic red flag the way they might have a decade ago. What still causes hesitation is a gap that seems to be actively hidden through vague or missing dates, since that reads as evasive regardless of the actual underlying reason.
Job titles that did not match seniority level created friction
One pattern that stood out involved candidates whose most recent job title suggested a lower level of seniority than the role they were applying for required, without any accompanying explanation of expanded scope or responsibility beyond the title itself. These resumes advanced at a lower rate than resumes where the seniority gap was explicitly addressed, either through a brief note about taking on responsibilities beyond the formal title, or through a summary statement clarifying the intended trajectory.
This suggests that recruiters are reading job titles somewhat literally as a first pass filter, and candidates whose actual scope of work exceeded their formal title benefited noticeably from making that gap explicit rather than assuming a reader would infer it.
AI-polished resumes showed a more complicated pattern
Given how common AI writing assistance has become, we looked specifically at resumes that showed clear signs of heavy AI polish, unusually uniform sentence structure, a certain generic fluency across every bullet point regardless of role. These resumes performed slightly better than average at the initial screening stage, likely because the polish itself reads well in a fast scan. But they showed a noticeably higher drop off rate once candidates reached a live, structured interview, particularly when interview questions asked for deeper specifics beyond what was written on the resume itself.
This gap between resume performance and interview performance is exactly the kind of thing that reinforced our thinking in building AI-conducted interviews into the platform in the first place. A resume alone increasingly measures how well someone, or a tool assisting them, can write. An interview measures whether the underlying substance actually holds up.
Skills sections with too many items diluted attention
Resumes with skills sections listing more than roughly 15 items showed a slightly lower advancement rate than resumes with a shorter, more focused list, even when the longer lists technically included all the relevant skills somewhere within them. This lines up with what we know generally about how quickly resumes get scanned. A long list makes it harder for a reader to quickly confirm the specific skills they are looking for, even if those skills are technically present.
Industry differences were larger than expected
- Technical & Engineering: Specific project details and named tools drove advancement, while executive summaries had minimal impact.
- Sales & Business Development: Quantified achievements and financial metrics carried substantially higher weight than average.
- Creative & Marketing: Including direct links to portfolios and live campaign samples significantly improved shortlist rates.
Referral candidates followed a different pattern entirely
Resumes submitted through an employee referral showed a noticeably different relationship between resume quality and advancement compared to resumes submitted cold. Referred candidates with fairly average, unremarkable resumes still advanced to interview stages at meaningfully higher rates than equally average resumes submitted without a referral. This is not particularly surprising, since referrals carry an implicit vouching effect that a resume alone cannot replicate, but seeing the actual size of that effect in the data was still notable. It reinforces something candidates hear often but do not always act on: investing time in genuine networking can outweigh a significant amount of resume polishing, at least for getting an initial foot in the door.
What surprised us most
If there was one finding that stood out above the rest, it was how much more predictive interview performance was of eventual hiring outcomes compared to any resume characteristic we measured, including the strongest resume signals like quantified achievements and tailoring. This was not a small gap. It reinforced something we already believed strongly enough to build a company around, but it was still striking to see reflected so clearly once we actually looked at the numbers across a large sample rather than relying on intuition alone.
What this means going forward
None of these patterns amount to a rigid formula, and we would caution against treating them that way. But taken together, they point toward a fairly consistent picture: specificity, genuine tailoring, and honest framing of real experience consistently outperform polish, length, and generic completeness. That pattern held up whether we looked at initial resume screening or how candidates performed once they reached a live interview.
For hiring teams, this reinforces the value of looking past surface level polish and toward specific, verifiable detail, something structured interviews are particularly good at surfacing. For candidates, it is genuinely encouraging news. The things that actually help are within reach for almost anyone willing to put in a bit of extra effort, regardless of access to expensive resume services or advanced AI tools.
Evaluate candidates beyond the resume surface.
If you want a closer look at how APTI Careers evaluates candidates beyond the resume itself through structured AI interviews, we would be glad to walk you through it.
Comments
Post a Comment