Our latest findings suggest that in a market flooded with polished CVs and AI-assisted applications, the most useful signals are the ones candidates cannot easily manufacture. Measures of how people think through problems and handle ambiguity are becoming more decisive.<\/p>\n\n\n\n
![\"What](\"https:\/\/www.clevry.com\/wp-content\/uploads\/What-2.1-Million-assessments-reveal-about-AI-shaped-work-1024x535.png\")
<\/figure>\n\n\n\nWhat the assessment data can tell you, and what it cannot<\/strong><\/h2>\n\n\n\nAssessment data is useful because it behaves differently to CVs and interviews. It is structured, comparable, and repeatable. When you are looking at high volumes, several patterns emerge.<\/p>\n\n\n\n
\n- Capability distribution is wider than hiring teams assume.<\/strong> Two candidates with similar CVs can sit at very different levels of reasoning capability, accuracy, and speed of processing.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Job titles are a weak proxy for job demands.<\/strong> The same title in two companies can require very different cognitive load and behavioural demands.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- \u201cPotential\u201d is easier to spot when you select for traits.<\/strong> Many hiring teams talk about potential, but do not define what they mean. Assessment results help translate potential into measurable indicators.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
What the data does not do is replace human judgement. It does not capture motivation on its own, or culture-fit in a general sense. It also does not remove the need for good job design and line manager clarity. It simply reduces guesswork, and that matters when the cost of a mis-hire is rising.<\/p>\n\n\n\n
<\/div>\n\n\n\nShift 1. AI is widening roles, so hiring for narrow experience is costing you<\/strong><\/h2>\n\n\n\nAI is not only automating tasks. It is also adding new tasks to jobs, often without job descriptions keeping up. People are expected to write better, summarise faster, interpret dashboards, data, handle exceptions, and make judgement calls when tools disagree.<\/p>\n\n\n\n
Our report indicates that organisations relying heavily on prior experience can end up filtering out candidates who could acclimate quickly, simply because their experience is not a perfect match. The introduction of these AI tools means adaptability becomes a miore practical requirement, and not just a nice-to-have.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruitment<\/strong><\/p>\n\n\n\n\n- Rebalance your criteria away from \u201chas done the exact job before\u201d and towards \u201ccan learn the job fast and execute reliably\u201d.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Update role profiles so they reflect what people actually do now, including the cognitive load created by tools, data, and decision speed.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Use assessments early enough to prevent wasted interview cycles with candidates who look right on paper but cannot handle the role\u2019s complexity.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 2. Baseline reasoning capability is more vital than ever<\/strong><\/h2>\n\n\n\nA few years ago, many organisations who came to us generally reserved ability testing for graduate pipelines or specialist roles. The recent data suggests this boundary is fading. Roles that were once considered \u201cprocess-driven\u201d now contain more decision-making. Roles that were once \u201cpeople-focused\u201d now require sharper written communication and judgement with information.<\/p>\n\n\n\n
The report indicates that verbal, numerical, and abstract reasoning are increasingly being seen as useful predictors of performance, especially when roles require candidates to interpret information, prioritise, and avoid errors under time pressure.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruiters and hiring managers<\/strong><\/p>\n\n\n\n\n- You need a clearer definition of what \u201cgood\u201d looks like for each job family. Not an abstract benchmark, but a role-relevant standard.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Cut-offs and scoring banding should be applied with some thought. Overly harsh thresholds can reduce talent pools unnecessarily, while too-low a threshold can have the opposite effect.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 3. Candidate expectations are now part of your selection validity<\/strong><\/h2>\n\n\n\nCandidate experience used to be positioned as employer branding. In 2026, it is also a measurement issue. If candidates disengage, rush, or drop out, your process stops reflecting the talent market you are trying to hire from. When the hiring and\/or assessment processes are slow, unclear, or feel unfair, candidate dropouts and disengagement rise.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for candidates<\/strong><\/p>\n\n\n\n\n- Candidates want speed, clarity, and respect for their time.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- They also want evidence that decisions are fair. Structured assessments, used properly, can support this by reducing subjective bias and giving consistent measurement across applicants.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Practical moves<\/strong><\/p>\n\n\n\n\n- Tell candidates why you are assessing, what you are assessing, and how it relates to the role.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Keep the process lean. Measure what you need, then decide.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Provide meaningful feedback where possible, even if it is brief. It protects your reputation and improves completion rates.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 4. Recruiters roles are changing and need better tools and governance<\/strong><\/h2>\n\n\n\nWhile AI has not removed the recruiter\u2019s role, it most definitely has changed it. Recruiters are now expected to manage larger volumes of applicants, work with increased speed, and navigate AI tools that can introduce risk if used carelessly.<\/p>\n\n\n\n
Our data suggests organisations that perform best treat selection as a system, not a set of disconnected steps. They use structured data to support decisions, and they monitor their hiring process like any other key business process.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruitment teams<\/strong><\/p>\n\n\n\n\n- Your value is shifting towards diagnostic capability. You are translating business needs into measurable hiring criteria.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- You need governance. That means knowing which tools are used, what they measure, how decisions are made, and how to evidence fairness.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
A simple approach that works<\/strong><\/p>\n\n\n\n\n- Define role success outcomes.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Identify measurable predictors.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Use assessments that map to those predictors.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Use structured interviews to validate and explore.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Track outcomes and refine.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 5. The financial case for better selection is becoming unavoidable<\/strong><\/h2>\n\n\n\nMost organisations can quantify cost per hire. Fewer can quantify the cost of a poor hire, high attrition rates, underperformance, and management time lost to remediation. Yet these are usually much bigger costs.<\/p>\n\n\n\n
The data suggests that improving quality of hire by even a small margin has outsized impact because it compounds. A better hire performs sooner, creates fewer errors, strengthens team performance, and reduces replacement hiring.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhere the savings come from<\/strong><\/p>\n\n\n\n\n- Shorter time-to-hire through better early screening.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Fewer late-stage interviews wasted on weak-fit candidates.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Reduced early attrition and probation failures.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
\n- Higher productivity through stronger capability alignment.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
This is not theoretical. It is operational. If you are hiring at any scale, selection quality becomes a measurable business lever.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat to do next<\/strong><\/h2>\n\n\n\nIf you want your hiring process to keep pace with the new AI-shaped world of work, focus on four actions.<\/p>\n\n\n\n
1. Re-define what \u201cgood\u201d looks like for each role<\/strong>
<\/strong>Translate desired performance into measurable requirements. Include cognitive load, error tolerance, and learning speed.<\/p>\n\n\n\n2. Build a selection funnel that removes noise early
<\/strong>Use a good psychometric test platform<\/a> to help you narrow the field before you commit hours of time to interviews or having to rehire for the same role.<\/p>\n\n\n\n3. Treat candidate experience as a data quality issue<\/strong>
<\/strong>Make the process clear, relevant, and efficient. You will get better engagement and more reliable signals.<\/p>\n\n\n\n4. Track outcomes and refine<\/strong>
<\/strong>Link selection data to performance and retention where you can. Continuous refinement here is where organisations can gain an edge.<\/p>\n\n\n\n<\/div>\n\n\n\nThe bottom line<\/strong><\/h2>\n\n\n\nIf you want the detail behind these patterns, read the latest Clevry Hiring Intelligence Report<\/a>. The report data indicates where the strongest signals sit across large-scale assessment results, and how those signals are shifting as roles evolve.<\/p>\n\n\n\n
- \n
- Job titles are a weak proxy for job demands.<\/strong> The same title in two companies can require very different cognitive load and behavioural demands.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- \u201cPotential\u201d is easier to spot when you select for traits.<\/strong> Many hiring teams talk about potential, but do not define what they mean. Assessment results help translate potential into measurable indicators.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
What the data does not do is replace human judgement. It does not capture motivation on its own, or culture-fit in a general sense. It also does not remove the need for good job design and line manager clarity. It simply reduces guesswork, and that matters when the cost of a mis-hire is rising.<\/p>\n\n\n\n
<\/div>\n\n\n\nShift 1. AI is widening roles, so hiring for narrow experience is costing you<\/strong><\/h2>\n\n\n\n
AI is not only automating tasks. It is also adding new tasks to jobs, often without job descriptions keeping up. People are expected to write better, summarise faster, interpret dashboards, data, handle exceptions, and make judgement calls when tools disagree.<\/p>\n\n\n\n
Our report indicates that organisations relying heavily on prior experience can end up filtering out candidates who could acclimate quickly, simply because their experience is not a perfect match. The introduction of these AI tools means adaptability becomes a miore practical requirement, and not just a nice-to-have.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruitment<\/strong><\/p>\n\n\n\n
- \n
- Rebalance your criteria away from \u201chas done the exact job before\u201d and towards \u201ccan learn the job fast and execute reliably\u201d.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Update role profiles so they reflect what people actually do now, including the cognitive load created by tools, data, and decision speed.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Use assessments early enough to prevent wasted interview cycles with candidates who look right on paper but cannot handle the role\u2019s complexity.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 2. Baseline reasoning capability is more vital than ever<\/strong><\/h2>\n\n\n\n
A few years ago, many organisations who came to us generally reserved ability testing for graduate pipelines or specialist roles. The recent data suggests this boundary is fading. Roles that were once considered \u201cprocess-driven\u201d now contain more decision-making. Roles that were once \u201cpeople-focused\u201d now require sharper written communication and judgement with information.<\/p>\n\n\n\n
The report indicates that verbal, numerical, and abstract reasoning are increasingly being seen as useful predictors of performance, especially when roles require candidates to interpret information, prioritise, and avoid errors under time pressure.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruiters and hiring managers<\/strong><\/p>\n\n\n\n
- \n
- You need a clearer definition of what \u201cgood\u201d looks like for each job family. Not an abstract benchmark, but a role-relevant standard.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Cut-offs and scoring banding should be applied with some thought. Overly harsh thresholds can reduce talent pools unnecessarily, while too-low a threshold can have the opposite effect.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 3. Candidate expectations are now part of your selection validity<\/strong><\/h2>\n\n\n\n
Candidate experience used to be positioned as employer branding. In 2026, it is also a measurement issue. If candidates disengage, rush, or drop out, your process stops reflecting the talent market you are trying to hire from. When the hiring and\/or assessment processes are slow, unclear, or feel unfair, candidate dropouts and disengagement rise.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for candidates<\/strong><\/p>\n\n\n\n
- \n
- Candidates want speed, clarity, and respect for their time.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- They also want evidence that decisions are fair. Structured assessments, used properly, can support this by reducing subjective bias and giving consistent measurement across applicants.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Practical moves<\/strong><\/p>\n\n\n\n
- \n
- Tell candidates why you are assessing, what you are assessing, and how it relates to the role.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Keep the process lean. Measure what you need, then decide.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Provide meaningful feedback where possible, even if it is brief. It protects your reputation and improves completion rates.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 4. Recruiters roles are changing and need better tools and governance<\/strong><\/h2>\n\n\n\n
While AI has not removed the recruiter\u2019s role, it most definitely has changed it. Recruiters are now expected to manage larger volumes of applicants, work with increased speed, and navigate AI tools that can introduce risk if used carelessly.<\/p>\n\n\n\n
Our data suggests organisations that perform best treat selection as a system, not a set of disconnected steps. They use structured data to support decisions, and they monitor their hiring process like any other key business process.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat this means for recruitment teams<\/strong><\/p>\n\n\n\n
- \n
- Your value is shifting towards diagnostic capability. You are translating business needs into measurable hiring criteria.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- You need governance. That means knowing which tools are used, what they measure, how decisions are made, and how to evidence fairness.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
A simple approach that works<\/strong><\/p>\n\n\n\n
- \n
- Define role success outcomes.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Identify measurable predictors.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Use assessments that map to those predictors.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Use structured interviews to validate and explore.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Track outcomes and refine.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
Shift 5. The financial case for better selection is becoming unavoidable<\/strong><\/h2>\n\n\n\n
Most organisations can quantify cost per hire. Fewer can quantify the cost of a poor hire, high attrition rates, underperformance, and management time lost to remediation. Yet these are usually much bigger costs.<\/p>\n\n\n\n
The data suggests that improving quality of hire by even a small margin has outsized impact because it compounds. A better hire performs sooner, creates fewer errors, strengthens team performance, and reduces replacement hiring.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhere the savings come from<\/strong><\/p>\n\n\n\n
- \n
- Shorter time-to-hire through better early screening.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Fewer late-stage interviews wasted on weak-fit candidates.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Reduced early attrition and probation failures.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
- \n
- Higher productivity through stronger capability alignment.<\/li>\n<\/ul>\n\n\n\n<\/div>\n\n\n\n
This is not theoretical. It is operational. If you are hiring at any scale, selection quality becomes a measurable business lever.<\/p>\n\n\n\n
<\/div>\n\n\n\nWhat to do next<\/strong><\/h2>\n\n\n\n
If you want your hiring process to keep pace with the new AI-shaped world of work, focus on four actions.<\/p>\n\n\n\n
1. Re-define what \u201cgood\u201d looks like for each role<\/strong>
<\/strong>Translate desired performance into measurable requirements. Include cognitive load, error tolerance, and learning speed.<\/p>\n\n\n\n2. Build a selection funnel that removes noise early
<\/strong>Use a good psychometric test platform<\/a> to help you narrow the field before you commit hours of time to interviews or having to rehire for the same role.<\/p>\n\n\n\n3. Treat candidate experience as a data quality issue<\/strong>
<\/strong>Make the process clear, relevant, and efficient. You will get better engagement and more reliable signals.<\/p>\n\n\n\n4. Track outcomes and refine<\/strong>
<\/strong>Link selection data to performance and retention where you can. Continuous refinement here is where organisations can gain an edge.<\/p>\n\n\n\n<\/div>\n\n\n\nThe bottom line<\/strong><\/h2>\n\n\n\n
If you want the detail behind these patterns, read the latest Clevry Hiring Intelligence Report<\/a>. The report data indicates where the strongest signals sit across large-scale assessment results, and how those signals are shifting as roles evolve.<\/p>\n\n\n\n
- Higher productivity through stronger capability alignment.<\/li>\n<\/ul>\n\n\n\n
- Reduced early attrition and probation failures.<\/li>\n<\/ul>\n\n\n\n
- Fewer late-stage interviews wasted on weak-fit candidates.<\/li>\n<\/ul>\n\n\n\n
- Shorter time-to-hire through better early screening.<\/li>\n<\/ul>\n\n\n\n
- Track outcomes and refine.<\/li>\n<\/ul>\n\n\n\n
- Use structured interviews to validate and explore.<\/li>\n<\/ul>\n\n\n\n
- Use assessments that map to those predictors.<\/li>\n<\/ul>\n\n\n\n
- Identify measurable predictors.<\/li>\n<\/ul>\n\n\n\n
- Define role success outcomes.<\/li>\n<\/ul>\n\n\n\n
- You need governance. That means knowing which tools are used, what they measure, how decisions are made, and how to evidence fairness.<\/li>\n<\/ul>\n\n\n\n
- Your value is shifting towards diagnostic capability. You are translating business needs into measurable hiring criteria.<\/li>\n<\/ul>\n\n\n\n
- Provide meaningful feedback where possible, even if it is brief. It protects your reputation and improves completion rates.<\/li>\n<\/ul>\n\n\n\n
- Keep the process lean. Measure what you need, then decide.<\/li>\n<\/ul>\n\n\n\n
- Tell candidates why you are assessing, what you are assessing, and how it relates to the role.<\/li>\n<\/ul>\n\n\n\n
- They also want evidence that decisions are fair. Structured assessments, used properly, can support this by reducing subjective bias and giving consistent measurement across applicants.<\/li>\n<\/ul>\n\n\n\n
- Candidates want speed, clarity, and respect for their time.<\/li>\n<\/ul>\n\n\n\n
- Cut-offs and scoring banding should be applied with some thought. Overly harsh thresholds can reduce talent pools unnecessarily, while too-low a threshold can have the opposite effect.<\/li>\n<\/ul>\n\n\n\n
- You need a clearer definition of what \u201cgood\u201d looks like for each job family. Not an abstract benchmark, but a role-relevant standard.<\/li>\n<\/ul>\n\n\n\n
- Use assessments early enough to prevent wasted interview cycles with candidates who look right on paper but cannot handle the role\u2019s complexity.<\/li>\n<\/ul>\n\n\n\n
- Update role profiles so they reflect what people actually do now, including the cognitive load created by tools, data, and decision speed.<\/li>\n<\/ul>\n\n\n\n
- Rebalance your criteria away from \u201chas done the exact job before\u201d and towards \u201ccan learn the job fast and execute reliably\u201d.<\/li>\n<\/ul>\n\n\n\n
- \u201cPotential\u201d is easier to spot when you select for traits.<\/strong> Many hiring teams talk about potential, but do not define what they mean. Assessment results help translate potential into measurable indicators.<\/li>\n<\/ul>\n\n\n\n