Candidates have no reliable way to know if a company's interview process is well-run before investing 10-20 hours in it. Current review sites bury process quality under general company reviews.
Candidates rate interview processes on specific dimensions (communication, level-appropriateness, timeliness, preparation materials). Companies get a public score and can pay to access improvement insights.
Free for candidates, paid for companies wanting analytics dashboards, benchmarking against peers, and candidate sentiment reports
The pain is real but episodic — candidates feel it intensely during job searches (2-4x per career for senior engineers) but forget about it between searches. The Reddit signals are genuine: people remember great and terrible processes for years. However, it's not a daily pain, which makes organic engagement harder. The pain is sharper for senior+ candidates who value their time at $100-200+/hr equivalent and are evaluating 3-5 companies simultaneously.
TAM is constrained. Candidate-side is free, so revenue = employer-side. ~200K US companies actively hiring tech talent, but realistic B2B targets are mid-to-large companies with recruiting teams (maybe 15-20K). At $5K-$20K/yr, serviceable market is $75M-$400M. Not a billion-dollar TAM. Could expand beyond tech but interview process pain is most acute in tech's multi-round gauntlets. Glassdoor's interview vertical alone likely represents $50M+ in employer spend.
Candidates won't pay — must be free. Employer willingness to pay depends entirely on having enough review volume to make the data meaningful. Chicken-and-egg problem: companies won't pay for analytics until you have critical mass of reviews, and candidates won't review until there's a community. Companies already pay Glassdoor, Starred, and LinkedIn for overlapping employer branding — another subscription is a hard sell unless your data is uniquely valuable and can't be gotten elsewhere.
Very buildable as MVP. Core: review submission form with structured dimensions, company profiles with aggregated scores, basic search/browse. No ML required initially. Standard web app (Next.js + Postgres would do it). Auth, reviews, aggregation, company pages — a solo dev can ship this in 4-6 weeks. The hard part isn't tech, it's content and distribution.
Genuine gap exists. Nobody does structured, multi-dimensional interview process scoring with public cross-company benchmarking. Glassdoor's interview reviews are basic (sentiment + difficulty + free text). Starred/Survale only serve opted-in employers. No platform lets a candidate say 'show me which companies have the fastest feedback loops and most respectful processes for Staff engineers.' However, Glassdoor could ship this feature in a quarter if they prioritized it.
B2B subscription model works in theory (monthly analytics dashboards, quarterly benchmarks, real-time alerts). But value delivery depends on continuous review flow. If review volume stalls, employer dashboards become stale and churn spikes. Unlike Glassdoor (which has salary + company reviews + jobs driving daily visits), a single-purpose interview review site has thin recurring value unless you expand scope. Annual contracts tied to recruiting cycles might work better than monthly.
- +Genuine gap: no one does structured, multi-dimensional interview process scoring with public benchmarking
- +Clear two-sided value prop: candidates get transparency, employers get actionable improvement data
- +Pain signals are authentic and emotionally resonant — people remember bad interview processes for years
- +Technically simple MVP — the challenge is community, not engineering
- +Senior engineers (the target) are high-influence: their reviews carry weight and companies care most about their experience
- +Regulatory/cultural tailwinds: pay transparency laws are normalizing employer accountability, interview process transparency is a natural next step
- !CRITICAL: Cold-start / chicken-and-egg problem — need review volume before employer value exists, need employer profiles before candidates visit. This is the #1 killer for review platforms.
- !Glassdoor could add structured interview scoring as a feature update, instantly leveraging their existing millions of reviews and employer relationships
- !Review frequency is low — most people interview 2-4x in a career, vs. reviewing restaurants monthly. Organic growth will be slow without viral mechanics
- !Verification is hard — how do you confirm someone actually interviewed at a company without employer cooperation? Fake reviews will be a major trust issue
- !Employer sales cycle is long and enterprise-y — hard for a solo founder to close B2B deals while also building community
- !Risk of adversarial dynamics: companies may try to game scores, submit fake positive reviews, or threaten legal action over negative ones
Largest platform with a dedicated 'Interviews' tab where candidates rate overall experience
Anonymous professional network verified by work email. Interview experiences are shared as forum posts in company-specific channels. Extremely candid due to anonymity + verification.
Primarily compensation data platform with an 'Interviews' section where users share anecdotal interview experiences and some community-contributed interview guides.
Employer-facing candidate experience survey platforms. Starred sends automated NPS-style surveys at hiring funnel stages and provides analytics and benchmarking for participating employers. Talent Board runs annual CandE benchmark research.
Workplace culture and compensation comparison platform with some interview-related content embedded in general company culture ratings. Known for demographic-segmented data and 'Best places to work' awards.
Launch as a single-page site targeting ONE job function (senior+ software engineers) at the top 100 tech companies. Pre-seed with 200-300 reviews sourced manually from Reddit threads, Blind posts, and personal network (with permission/attribution). Review form: 5 structured dimensions scored 1-5 (communication speed, interviewer preparedness, process transparency, feedback quality, respect for time) plus optional free text. Company pages show radar charts and percentile rankings. No employer features yet — just the public scorecard. Gate full access behind 'give-to-get' (submit a review to see all reviews). Launch on Hacker News and r/ExperiencedDevs.
Phase 1 (months 0-6): Free, build review volume to 5K+ reviews covering top 200 companies. Phase 2 (months 6-12): Launch free employer claim-your-profile (like Yelp for Business) to get employer emails and engagement. Phase 3 (months 12-18): Introduce paid employer tier ($299-$999/mo) with analytics dashboard, dimension-level benchmarking vs peers, candidate sentiment trends, and the ability to respond to reviews. Phase 4 (18+ months): Enterprise tier ($2K-$10K/mo) with ATS integration, real-time candidate feedback collection, custom benchmarking cohorts, and API access. Long-term optionality: recruiting marketplace (companies with top scores get preferential candidate access).
9-15 months. First 6-9 months are pure community building with zero revenue. Employer revenue requires critical mass (~5,000+ reviews, ~200+ companies with 10+ reviews each). A solo founder could potentially compress this to 6-9 months with aggressive manual seeding and a viral launch, but review platforms historically take 12-18 months to first meaningful revenue.
- “It is always a huge green flag when a company has got their interview process down to a tee”
- “they were so efficient that their name has stuck with me ever since”
- “Could you please share the name of the company so that we can avoid applying to this place?”
- “If that is their interview process, I dread to think what their day to day looks like”