MECEmate

LOGO
Case Library
Community
Pricing
Investor Info

Inspiration

Peer practice often yields vague feedback; expert coaching is effective but expensive, scarce, and hard to schedule. Different firms prize different behaviors (McK interviewer-led vs. BCG/Bain candidate-led), yet most tools are one-size-fits-all. We wanted a lovable way to make coaching accessible, firm-specific, measurable, and fast—so every rep converts to progress.

What it does

Firm-specific rubrics: McKinsey/BCG/Bain-calibrated weights across Structure, Math, Exhibits, Brainstorming, Communication, Synthesis. Clear metrics: 1–5 anchored scores per component + concrete examples (“what 3 vs 4 looks like”). Progress & benchmarking: Trendlines vs. your history and anonymous peer percentiles. Tailored plan after every rep: Instant, actionable next-steps and drills; flags persistent gaps and highlights trend deltas. Session capture: Guided scorecard for solo or peer mocks; optional notes/audio timestamping. Pricing: One-off detailed eval ($5–$10) or monthly sprint ($50/mo). Turnaround: Eval → scores → dashboard → improvement plan in <60 seconds.

How we built it

Based on the user pain points we identified, we translated them into four core user journeys: user registration, mock interview, response evaluation, and trend analysis. We then created a detailed design document outlining the overall system architecture, clearly defining the responsibilities, functions, and resource dependencies of each component, and mapping out the end-to-end workflow. Next, we converted each component’s description into structured prompts that were easy for the system to interpret and used Lovable to generate the initial prototype. Once the prototype was built, we conducted manual testing, gathered feedback, and iterated to refine the product and improve user experience.

Challenges we ran into

Calibration: Making firm-specific anchors precise enough to feel “insider” without overfitting. Actionability vs. brevity: Ensuring feedback is sharp, not generic, yet still scannable. Fair benchmarking: Normalizing peer percentiles across case types and experience levels. Speed vs. depth: Hitting sub-minute turnaround without dumbing down insights. Behavior change: Getting users to replace unstructured peer notes with structured scorecards.

Accomplishments that we're proud of

Converted a manual evaluation matrix into production-ready, firm-specific rubrics with behavioral anchors. Shipped a working scores → dashboard → plan loop that’s actually motivating (trendlines, deltas, percentiles). Defined clear MLP acceptance criteria and success metrics (activation, D30, +1.0 average score lift over 6 sessions). Built a clean brand system (MECEmate wordmark + AI “brain” icon) consistent with a premium consulting aesthetic. Laid a pragmatic tech path: rules-engine first, LLM for narrative, human-in-the-loop calibration when needed.

What we learned

Firm targeting matters: Users respond strongly when feedback names their target firm’s expectations. Anchored examples beat adjectives: Showing “what a 4 looks like” drives faster improvement. Trends motivate: Visible deltas and peer percentiles increase practice frequency. Timing is everything: Immediate feedback (<60s) keeps users in flow; delays break momentum. Clear next steps win: Specific drills (“Do 3 profitability frameworks with X pattern”) outperform generic checklists.

What's next for MECEmate

v1.5: Smart drills library, exportable “coach brief,” richer exhibit handling. v2: AI interviewer (scripted exhibits, speech capture), auto-timed prompts, talk-through analysis. GTM: Pilots with consulting clubs/cohorts, ambassador program, “30-day interview sprint” packs, referrals. Enterprise: Team dashboards and admin for clubs/career centers; campus licensing. Quality & fairness: Ongoing rubric calibration, bias checks, and A/B tests on feedback templates. Security & scale: Harden auth/PII, audit logs, SOC-2 track as traction grows. Expansion: Adapt rubrics to PM/tech case interviews; broaden from consulting prep → interview performance analytics platform.