Recruiter in
Your Pocket
All research
ATS & Automation/4 min read/Direct sources linked

Where automation enters the hiring process

Hiring automation is not one bot. It is a chain of systems that can shape exposure, eligibility, ranking, and review.

Where automation enters hiring

Automation isn't just "scanning keywords" at the end. It starts before you even see the job ad.1

Before a person reviews the resume

Bias can enter earlier than the human review.

Data, rules, and automated filters can all affect who appears in the final group.

FIG 01
  1. Sourcing
    Who sees the role

    Ad delivery can shape the pool before applications exist.

  2. Parsing
    What gets understood

    Non-standard structures can be read unevenly.

  3. Ranking
    Who rises

    Gaps, schools, titles, and other proxies can inherit old patterns.

  4. Assessment
    Who advances

    Automated scoring can make a narrow definition of fit look objective.

Earlier filters matter. A human reviewer may only see the candidates who made it through every previous step.
Fig. 1 / Before the human reviewAutomation does not remove judgment. It can shape the pool before a person sees it.

One hiring process, several decision points

Most candidate advice focuses on resume parsing. The Upturn report shows a wider system: job advertising, sourcing, screening, assessment, and ranking can each involve automated decisions. If an ad is not delivered to someone, that person cannot enter the applicant pool at all.1

We cannot fix the ad servers. Once you are in the pipeline, clear structure can reduce parser friction and make your qualifications easier to interpret.2

When the system changes who gets seen

Exposure bias happens when the system decides who sees a job or who is surfaced first. Candidates who are not surfaced never get evaluated.1

Common questions

Where does bias enter the hiring funnel?

Bias can enter at ad delivery, eligibility screening, ranking, assessment, and human review. An early decision can change who is available for every later stage.

Can a resume fix algorithmic bias?

No. A resume can only reduce avoidable noise, not systemic bias.

What should candidates do?

Candidates cannot fix systemic bias. They can keep the file readable and make relevant experience explicit so avoidable parsing failures do not become another problem.

How this shows up in your report

01

Reduce system friction

We use conventional structure and explicit evidence so both systems and people have less to decode.

02

Separate fact from folklore

We separate what public research documents from what private platforms do not disclose.

Sources and related research

Sources

  1. 01Help Wanted: An Examination of Hiring Algorithms, Equity, and Bias — Upturn (2018).
  2. 02Unsuccessful resume parse — Greenhouse Support (2026).

Related research

ATS MythsIndustryHuman vs. AlgorithmTrustHiring DiscriminationResearch

Apply it to your resume

See what your resume makes clear—and what it leaves open.

Review my resume