Online test practice
Practise the formats of the screens that come before the interview. Every set is original, runs in your browser, is scored with explanations, and lets you relax or turn off the timer.
Practice in a similar format, not the real test. No firm's test is copied or reproduced.
Game-based problem-solving screens
Candidates report two kinds of module in some firms' game-based screens: a data study that ends in a short written report, and a puzzle where you choose options that meet several conditions at once. The official and reported facts are in the lesson: Gamified problem-solving assessments.
- Data studySort the data you need from the data you do not, answer three calculations, then write a four-sentence report.About 15 minutes. Feeds: Reading exhibits, Case math, Synthesis.Start
- Constraint puzzleChoose a set of options from a pool so that every numeric condition holds at once. Five puzzles, getting harder.About 15 minutes. Feeds: Case math.Start
Aptitude tests
Outside the largest firms, a timed numerical reasoning test is the most common first screen. Providers differ by firm and office. The official and reported facts are in the lesson: Aptitude and psychometric tests.
Chatbot-led online cases
A case delivered as a series of timed questions by a chatbot, often ending with a short recorded recommendation. The official and reported facts are in the lesson: Chatbot-led digital case interviews.
AI-assisted interviews
Some firms have started interviews where you use an AI tool while the interviewer watches how you check its work. The official and reported facts are in the lesson: AI-assisted case interviews.
How these are scored, and what they cannot tell you
- Every calculation is generated with its answer, and a checking script recomputes each answer a second way from the numbers on screen, across a thousand sets of each kind.
- Every puzzle is checked by a solver that tries every possible choice, so each one has at least one right answer.
- Written answers (the data study report and your question in Check the AI) are checked by published rules and key words. They check form, not the quality of your thinking.
- Some real screens are reported to score how you work as well as your answers. Nobody outside the firms knows how, so we score your answers only.
- Results feed your standing for case math, exhibits and synthesis only, as each set lists. Formats change often: we re-check these descriptions at least every 90 days.