For Recruiters and Talent Acquisition

20 Practical Ideas for Recruiters to Stay Cognitively Sovereign

LinkedIn Recruiters AI and similar tools can screen out strong candidates before you ever see them. When algorithms mediate all early contact, you lose the instinct that spots potential a CV does not show.

These are suggestions. Take what fits, leave the rest.

Download printable PDF

Screening and Assessment

Review every rejected candidate flagged by AIbeginner
Manually check at least ten percent of profiles your screening tool rejected last month.
Create a hiring bias audit checklistbeginner
List the specific characteristics your AI screener favours. Check if they match your actual job needs.
Test your job description with real peopleintermediate
Ask five candidates outside your target demographic to read your AI-written job spec. Record what confuses them.
Document candidates AI rejected but you hiredintermediate
Track which rejected profiles turned into successful hires to measure your screener's false negative rate.
Require phone screening before assessment toolsbeginner
Talk to candidates yourself before running them through HireVue or similar video assessment systems.
Build a control group of manual assessmentsintermediate
For one week each quarter, screen candidates without AI. Compare the quality of people you find.
Ask candidates about unconventional career pathsbeginner
In your assessment, specifically ask about job gaps or sideways moves your screening tool might penalise.
Set explicit thresholds for AI confidence scoresbeginner
Decide now what Eightfold or similar confidence score triggers human review. Write it down before screening starts.
Compare AI rankings to your own shortlistintermediate
Rank your top five candidates manually, then compare to your tool's top five. Document the differences.
Interview candidates your tool ranked lowestintermediate
Randomly select one candidate from the bottom ten percent of AI rankings. See if they perform well.

Sourcing and Job Design

Write job descriptions without using AI firstbeginner
Draft the role yourself before asking ChatGPT to refine it. Keep your original version to compare.
Source candidates from outside your usual channelsintermediate
Find five strong candidates this month using methods your LinkedIn algorithm will not surface naturally.
Ask your best recent hires how they found youbeginner
Interview five successful employees about their recruitment experience. Identify which touchpoints mattered most to them.
Build sourcing lists based on skill, not keywordsintermediate
Create a candidate pool for your next role that includes people whose titles do not match traditional keywords.
Test your job description for clarity without jargonbeginner
Remove three industry buzzwords from your AI-written job spec. See if candidate applications stay the same.
Track which candidates apply versus AI-sourced matchesintermediate
Compare your inbound applications this month to profiles your sourcing tool suggested. Measure the quality difference.
Source one candidate through direct outreach monthlybeginner
Personally reach out to one person per month who would not appear in algorithmic search results.
Review your job description for hidden requirementsbeginner
Read your posting aloud. Identify any requirements that are nice-to-have, not essential, that might filter out strong people.
Compare job descriptions across your organisationintermediate
Look at similar roles other recruiters posted. Identify which specs attract more qualified candidates who actually stay.
Create a sourcing strategy that avoids algorithmic homogeneityintermediate
Plan how you will find candidates with diverse backgrounds, industries, and education outside algorithmic recommendations.

Five things worth remembering

Related reads

The Book — Out Now

Cognitive Sovereignty: How To Think For Yourself When AI Thinks For You

Read the first chapter free.

No spam. Unsubscribe anytime.