Pymetrics
Updated 1 July 2026What is a good pymetrics score?
Candidates preparing for investment banking summer-analyst programs in New York or management consulting graduate schemes in London often approach the Pymetrics assessment with a traditional exam mindset. They look for a target percentage, a passing threshold, or a specific percentile cut-off. However, this search for a universal good score is based on a fundamental misunderstanding of the platform. Because the assessment is built on customized, role-specific predictive modeling, a profile that triggers a rejection for a quantitative trading desk might be exactly what secures a final-round interview for a relationship management position. Understanding this mechanism is vital to passing.
12
Behavioral games typically completed in a standard assessment session
Approximate
9
Core cognitive and emotional trait categories analyzed by the system
Industry standard
12 Months
Validity period of your game-play data before a retake is allowed
Standard policy
0
Global numerical passing scores or universal percentage cut-offs
Platform design
Quick answer
A good Pymetrics score is not a numerical percentage, but a high-statistical match to a specific role profile. Pymetrics, now owned by Harver, evaluates your unique cognitive and emotional traits against a bespoke benchmark created by analyzing the employer's top-performing employees in that exact business function.
Key points
- There is no global passing mark or numerical score; success is defined entirely by your alignment with a specific role model.
- The assessment measures approximately nine core behavioral categories across roughly 12 separate behavioral games.
- Your raw performance data is processed into a trait profile that remains valid for multiple employer submissions over a rolling 12-month period.
- Attempting to game the test by picking what you think is the right answer often introduces inconsistencies that reduce your overall role fit.
- Employers use your results to sort applicants into distinct compatibility bands, such as highly recommended, recommended, or unique fit.
The Reality of the Role-Specific Match Model
The core mechanism behind Pymetrics, which operates under the Harver brand, rejects the concept of a standardized test score. When a candidate completes the assessment for a corporate graduate scheme or summer-analyst program, their performance does not generate an overall percentage or percentile rank against all global test-takers. Instead, the platform compares the candidate's unique behavioral data points to a custom baseline model built specifically for the role they applied to. This baseline is established by running the assessment on the employer's existing top-performing employees within that specific division.
Because every business function requires different cognitive and emotional strengths, an ideal profile varies dramatically between departments. For example, a global investment bank may want a high risk tolerance and rapid attention deployment for its sales and trading desks, but require high attention to detail and a more deliberate, risk-averse approach for its compliance and operations divisions. Therefore, a trait profile that represents an excellent match for one career path could lead to an automatic rejection from another, even within the same company.
What the Candidate and Employer See
Upon completing the games, candidates are typically provided with a personalized trait report rather than a pass or fail notification. This report is framed entirely around personal attributes and workplace traits, outlining your natural tendencies in categories like attention, risk tolerance, fairness, and decision-making. The report highlights unique strengths and workplace preferences, ensuring that candidates do not receive a negative scorecard. This feedback is designed to be developmental, showing how you manage focus, process information, and respond to rewards or frustrations.
Behind the scenes, the employer receives a significantly different dashboard. Rather than reviewing individual game metrics, recruiters see a consolidated match rating or fit category for each applicant. This is often displayed as a tiered ranking, such as highly recommended, recommended, or unique fit. The software algorithmically calculates how closely your multidimensional trait footprint overlaps with the target role model. Recruiters use these bands as an initial screening filter to determine which candidates move forward to an assessment centre or superday interview, effectively bypassing standard CV or resume screening.
The Core Traits and Games Structure
The standard evaluation is built around roughly 12 short, interactive games that take between two and three minutes each to complete, leading to a total testing time of approximately 25 to 30 minutes. These exercises are designed to collect thousands of micro-behavioral data points by observing how you react to shifting instructions, financial risk, monetary rewards, and facial expressions. The platform uses this data to map your profile across approximately nine core categories: attention, effort, emotion, fairness, focus, generosity, learning, risk, and trust.
Because the software records every reaction time and adjustment, the process prioritizes how you reach an outcome over the outcome itself. For example, in the balloon game, the system does not just count how much digital currency you accumulate; it tracks whether you become more cautious or more reckless immediately after a balloon pops. This level of granularity makes it incredibly difficult to figure out what a good score looks like while you are playing.
Balloon Pumping
Measures risk tolerance and reward sensitivity. You inflate a digital balloon to earn money before it bursts, and the system tracks whether you grow more cautious or more reckless after each pop.
Money Exchange
Measures trust and fairness. You share an allocation of funds with an AI partner under varying rules, revealing how you cooperate and reciprocate.
Keypress Speed
Measures effort and motivation. You tap a specific keyboard key as fast as possible within a tight window.
Digit Memory
Measures attention and working memory. You recall increasingly long sequences of numbers displayed on screen.
Why Gaming the Profile Disastrously Backfires
Many undergraduate applicants attempt to reverse-engineer the platform by researching what they assume to be the perfect corporate profile. A candidate might decide that a top-tier consulting firm only wants highly competitive, fast-moving risk-takers, and will subsequently click through every game as quickly as possible while maximizing every financial risk. This approach almost always backfires. The underlying algorithm is trained to look for internal consistency across multiple games that test similar overlapping traits.
When a candidate forces artificial choices, they create conflicting data points across the 12 games. For instance, if a candidate acts highly altruistic in a trust game but reveals completely selfish, high-risk tendencies in a reward game, the model flags these contradictory behaviors. This erratic data profile reduces your statistical match score against the target role model. Furthermore, attempting to game the system can trigger anti-cheat or data-inconsistency flags, which can ruin your chances before a recruiter ever reviews your CV or resume.
Multi-Role Matching and Trait Redistribution
One of the most useful features of the system for candidates is multi-role trait redistribution. Because your raw game-play data is saved to a centralized profile, your results are valid across all employers utilizing the platform for a rolling 12-month period. If you apply to a summer-analyst program at one investment bank and are prompted to complete the assessment, those exact same results will automatically be submitted if you apply to a different bank using the platform later that season. You cannot retake the test within that window simply because you dislike your initial performance.
However, this architecture also allows for internal redistribution. Many large graduate employers configure the platform to automatically check your profile against alternative business paths. If your game-play data shows a weak match for investment banking but an exceptional match for technology consulting or corporate risk management, the system can flag this alignment to the recruitment team. Some employers will proactively reach out to candidates to offer them an alternative pipeline based entirely on this trait compatibility, turning a rejection into an opportunity.
How it works
How Pymetrics scores your assessment
The underlying mechanics of the platform rely on a proprietary machine learning framework that evaluates human behavioral data. As a candidate interacts with the user interface, the system logs behavioral variables down to the millisecond, including response latency, pattern adjustments following errors, and shift choices under uncertainty. These raw variables are aggregated to build a holistic profile across the nine standard trait dimensions. The scoring engine does not use a normative comparison model that ranks you directly against other test-takers in a traditional percentile curve; rather, it uses a classification model to determine profile similarity.
To establish the target profile, the employer conducts an internal benchmarking study. A representative group of successful incumbents in a specific business unit completes the assessment. The platform analyzes their combined data to identify the distinct behavioral patterns that correlate with high performance in that specific environment. This creates a customized algorithmic mold. When you apply, your multi-trait footprint is fed into this mold, and the algorithm calculates a multidimensional distance score to determine your fit classification.
The provider places a significant emphasis on algorithmic fairness and de-biasing, which is a major selling point for corporate HR departments. The machine learning models are designed to be explicitly audited for adverse impact before they are deployed. If a role-specific model shows a statistical bias against any protected demographic group based on gender, race, or ethnicity, the platform adjusts the weighting of the behavioral variables. It removes those elements that drive demographic differences while retaining the traits that predict job performance, ensuring an objective evaluation.
How to prepare
- 01
Familiarize yourself with the game formats
Before your official invitation, review the standard rules for common behavioral games like the balloon inflation or arrow-matching tasks to eliminate any initial interface confusion.
- 02
Set up a distraction-free environment
Complete the 25-minute session in a quiet room with a reliable internet connection, using a mouse if preferred, as sudden external interruptions will distort your natural response times.
- 03
Commit to a consistent behavioral approach
Do not alternate between extreme caution and reckless speed; approach the games naturally so the system can build a cohesive, unconflicted profile of your traits.
- 04
Read every set of instructions carefully
The game mechanics and rules change rapidly between rounds, so take your time between games to read the guidelines before clicking start.
A preparation timeline
The week before
Research the specific core values of the target firm and understand the day-to-day requirements of the business division you are entering.
The day before
Check your computer setup, clear your browser cache, ensure your hardware is responsive, and get adequate sleep to maintain sharp reaction times.
During the test
Focus purely on the instructions of the active game, ignore past mistakes or popped balloons, and maintain a steady, natural pace throughout the session.
How candidates approached it
Anonymised accounts of how recent applicants prepared, what they experienced, and how it turned out.
Management Consulting / UK Graduate Scheme / Advanced to Assessment Centre
Experience. A non-target applicant applied to a major London consulting practice and was required to complete the assessment. They decided not to overthink the games or try to project a false persona, choosing instead to play through the 12 exercises at a natural, steady pace. The applicant received a candidate feedback report highlighting high attention degradation and cautious learning patterns, which they initially feared was a bad result. However, two days later they were moved forward, as the firm's operational consulting arm specifically valued structured, methodical risk management over rapid, impulsive decision-making.
Outcome. Playing naturally at a steady pace let a methodical trait profile match the operational consulting arm, which valued structured risk management over impulsive speed.
Investment Banking / US Summer Analyst / Rejected at Screening Stage
Experience. A finance student from a target university applied for a competitive front-office summer-analyst program in New York. Believing that investment banks only hire hyper-aggressive, high-frequency decision-makers, the applicant consciously forced an extremely fast and high-risk strategy across the balloon and money-exchange games. This created a major statistical conflict with their performance in the attention and memory games, where their forced speed led to a high error rate. The platform flagged the profile as an inconsistent match for the role benchmark, resulting in an automatic rejection before their resume was ever reviewed by a human recruiter.
Outcome. Forcing an artificially fast, high-risk strategy created statistical conflicts across games, flagging an inconsistent match and triggering rejection before any human review.
Questions to practise
A bank of adjacent questions candidates run into. Drill each one in the exact format firms use.
- Why does the balloon game keep popping early in my assessment?
- Can I reset my behavioral traits profile if I fail an assessment?
- How long do employers store my game-play results in their applicant tracking systems?
- What traits are corporate risk management teams looking for in the assessment?
- How can I tell if my game-play data was shared between two different banks?
- Do consulting firms prefer high attention deployment or high risk tolerance?
- What happens if my browser crashes halfway through the money exchange game?
- Is there a way to view the exact match percentage that recruiters see on their dashboard?
- How does the system adjust for differences in typing speeds during the keypress game?
- Can a high match profile override a low GPA or lack of experience on a resume?
This answer is general guidance for orientation, not a guarantee. Test formats, timings and employer cut-offs change, so verify the details on the provider or employer site before you apply. Last updated 1 July 2026.