LSEG
250+ candidates screened for five specialist roles, with manual review time cut by up to 60%.
My role: Designed the psychometric screening and matching logic used on the Gyfted platform.
The Challenge
London Stock Exchange Group's shared service centre in Gdynia (1,300+ staff) needed analysts for narrow, specialist roles: Data Analyst, Senior Gas Market Analyst, Coal Market Research, LNG Analyst and Power Market Research Analyst. Each role combined Python, SQL and Excel skills with analytical judgment and client-facing English. Reviewing hundreds of applicants by hand for that combination is slow, and CVs say little about the judgment part.
The Approach
Candidates went through Gyfted's screening: validated measures of cognitive ability and work-relevant traits, combined with a matching layer that ranks each candidate against the profile of the specific role. Recruiters received a short, ranked list with interpretable profiles, so interviews were spent on the most promising people.
Results
- 250+ candidates screened across five specialist roles
- 47 invited to final interviews; a first batch of five candidates presented within a week
- Up to 60% less time spent on manual candidate review, so one recruiter could run the whole flow
- Typical recruitment cycle of about one month
LSEG is the plainest example of what screening is for. The roles were narrow, the applicant pool was large, and the expensive step was human review. Good measurement moves effort from reading CVs to talking with the right people.
The part I care about is the matching layer. A ranked list is only useful if a recruiter can see why someone ranks high, so each candidate’s result was reported as a readable profile against the role, with no opaque score behind it.
Delivered through Gyfted. The engagement is also described in Gyfted’s LSEG case study.