The Science of Potential

Selection for a three-year program relied on a personality inventory plus a structured interview - a combination that could reliably screen out only the bottom 10%.

A Glimpse model was generated based on successful alumni and placed at the top of the applicant funnel.

8.5× increase in the likelihood for successful completion of the program with 91% positive candidate experience. The organization gets in 45 minutes what it used to get after a full battery of screenings, testing days and assessment centers - roughly a month of elapsed time, plus the money and staff hours spent on people who may turn out to be the wrong ones.

Candidates were screened out based on proxy signals: CV, degree, years of experience - missing potential that looked different on paper.

Glimpse modeled the organization's best employees. Each candidate was evaluated against this model as part of the screening process.

44% of the candidate pipeline that would have previously been rejected showed high potential.

A large applicant funnel and no reliable way to know, at intake, who would ramp fast and who would stall.

A Glimpse model distributed applicants into three fit groups at intake, then measured real outcomes twelve months later.

18% of the top group reached top-tier advancement speed, against 5% of the middle group and none of the bottom. Dropout was cut by half between the top and bottom groups.

Often, promoting the best technical performer to a management position results in reduced team output.

A cross-validated Glimpse model was generated based on high performing leaders.

Individuals identified by the model were 12× more likely to succeed in leadership roles than their peers.

In Business Process Outsourcing, early dropout is a defining challenge - annual attrition runs 30–50%. High turnover drives up hiring, training and knowledge-management costs, with productivity implications tied directly to bottom-line results.

Glimpse partnered with a large BPO provider to predict the likelihood of early dropout during recruitment. A predictive model flagged the risk of attrition within six months, and 129 new hires across three contact centers were tested in their first week - with results sealed for eight months.

Over those eight months, 58 of the new employees resigned. When the data was unsealed, the outcome was decisive: had hiring followed the predictive model, early dropout would have been cut in half.

Organizations seek to maximize the potential of their workforce, ensuring access to opportunities based on quality and fit over demographic background or pedigree.

Glimpse models are generated based solely on performance-based signals, with zero demographic data.

With bias structurally mitigated, gender representation improved by 40% without compromising on quality. This was achieved not through targeted selection, but as a natural outcome of measuring what actually predicts success.

Glimpse partnered with a leading European e-commerce company to improve hiring for early-career roles, where potential — not track record — is what predicts success.

Drawing on insights from the company's existing high performers, Glimpse built a predictive model that identified candidates with a 4× higher likelihood of succeeding in the role.

The solution enabled streamlining recruitment, potentially saving the People & Culture team valuable time, and allowing hiring managers to focus their interviews on the most promising candidates.