BrightHire

What it is

BrightHire is the interview-intelligence platform for recruiting — the same category shape as Gong in sales, pointed at candidate interviews instead of customer calls. Recording, transcription, and analysis are the spine; four modules hang off it. Interview Planning drafts the job description and the interview plan before the loop opens. AI Interview Notes returns the transcript, the summary, and a filled scorecard. Interview Insights grades interviewer behavior and feeds the coaching loop. AI Interviewer, sold as BrightHire Screen, runs an asynchronous voice or video screen against a recruiter-defined rubric and hands back a scored write-up, triggered by hand or by an ATS stage change. Founded in 2019, BrightHire named the interview-intelligence category and sells into Canva, Duolingo, HubSpot, Ramp, SoFi, Klaviyo, and HCA Healthcare.

BrightHire is no longer an independent vendor. Zoom announced the acquisition on 13 November 2025 and closed it in December 2025; Zoom had been an investor since 2021 and shipped the in-call integration years before buying the company. Terms were not disclosed. BrightHire operates as a standalone brand inside Zoom on the Workvivo model — same leadership, same team, same product name, same domain — and both parties state that the ATS and conferencing integrations stay put, that Microsoft Teams and Google Meet remain supported, and that a Zoom license is not required to run it. Treat that as the stated intent of an acquirer eight months in, not as a term you already hold.

Why it shows up in Recruiting stacks

• Decision evidence with a timestamp. When the hiring committee splits on a candidate, BrightHire pulls the exact interview moment where the contested signal appeared. The debate moves from “it felt strong” to a clip and a transcript line, which is also the artifact you want when a rejected candidate challenges the decision.
• Coaching aimed at the interviewer, not the process. The analytics name who skipped a required question, who talked over the candidate, and who led the witness — per person, per loop, with the recording attached. That is a different product from a scorecard-compliance dashboard, and it is the reason structured interviewing survives contact with a busy engineering panel.
• Screen and loop share one rubric. BrightHire Screen runs first-round screens without scheduling, then the same rubric and the same evidence model carry into the human loop. Buying the AI interviewer and the interview intelligence from one vendor is what keeps the screen score and the onsite scorecard on the same scale.

Pricing reality

BrightHire publishes no numbers. The pricing page names three packages — Recruiters, Teams, Enterprises — plus BrightHire Screen sold standalone or bundled, and routes all four to a demo. Two variables drive the quote: platform seats (recruiters, hiring managers, interviewers) and recorded interview volume per month.

For a band to negotiate against, Vendr publishes anonymized transaction data from 57 purchases, last updated February 2026: a median of $18,000 a year, with $15,000–$35,000 at 5–15 users and under 100 interviews a month, $35,000–$75,000 at 15–50 users and 100–300 interviews, and $75,000–$150,000+ above 50 users and 300 interviews. The median sitting at the floor of the smallest band is the signal worth acting on — most buyers start narrow and expand, so price the first year for the team that will actually run it. Set that against Metaview, which publishes $0 free, $100 per user per month for Pro, and $300 for Max: a 20-seat Metaview Pro deployment is $24,000 a year at list, knowable before a sales call, and it lands inside BrightHire’s own mid band.

Best for

Talent teams at mid-market and enterprise companies running 100+ requisitions a year in judgment-heavy functions — engineering, product, sales leadership — where a bad hire costs more than the whole contract and the hiring committee argues from memory today. The ROI band is best where interview volume is high enough to justify the analytics but the panel is small enough that coaching a named interviewer changes outcomes. Zoom-standardized organizations get the shortest integration path, though not an exclusive one.

Do not buy BrightHire for high-volume hourly hiring. At warehouse or retail throughput the per-interview cost outruns the per-hire value, and a conversational screener like Paradox or an assessment-first platform like HireVue is the correct shape. Do not buy it if you need a price before a sales call — that requirement alone eliminates it.

Versus the alternatives

Metaview is the direct rival and the pick when you want AI notes and scorecard fill at a published, self-serve price — it starts free and tops out at $300 per user per month before Enterprise. Choose BrightHire instead when interviewer coaching and hiring-quality analytics are the reason you are buying, not a bonus. HireVue owns the high-volume, assessment-heavy end with the deepest Workday and SAP SuccessFactors coverage — pick it when structured screening across thousands of candidates is the job. Pillar, now inside Employ, is the cheaper interview-recording option for teams already on Lever or JazzHR; pick it on budget, not on analytics depth. Apriora is the fastest-growing entrant on the AI-interviewer side specifically — run it against BrightHire Screen if the screen is the whole purchase, and skip BrightHire’s platform tiers. Against the status quo of unrecorded interviews and a shared doc, BrightHire wins where the same five interviewers run the same loop every week and nobody can say why the last three hires failed.

Watch-outs

• The platform-agnostic promise is a statement, not a clause. Zoom says Teams and Meet stay supported and that no Zoom license is required. Acquirers say that in month one and reprice in year three. Guard: put continued Microsoft Teams and Google Meet support, and the no-Zoom-license condition, into the renewal terms in writing, with a termination-for-convenience right if either is withdrawn.
• The headline metrics are vendor-reported. The 27% fewer interviews per hire, 28% pipeline-efficiency gain, and 19% drop in candidate fall-off come from BrightHire’s own research across 25,000+ candidates, not an independent audit; the same applies to the roughly 70% time-to-interview reduction claimed for Screen. Guard: run a paid pilot on two live requisitions, measure interviews-per-hire against your own prior four quarters, and gate the contract on your number.
• No first-party MCP server. The only MCP path is Zapier’s generic wrapper, which exposes Zapier’s action surface rather than a BrightHire-maintained endpoint. Guard: if you plan to have an agent read interview evidence, scope the work against the REST API during the pilot and budget the integration rather than waiting for a connector.
• Recording consent is per-jurisdiction and per-candidate. Two-party-consent states and EU works councils both bite here, and an AI screen adds automated-decision disclosure on top — NYC Local Law 144 and the EU AI Act reach the screening step, not the notes step. Guard: have counsel sign off on the consent flow per hiring region before rollout, and keep BrightHire Screen switched off in regions where the disclosure is unresolved.
• Analytics only bind if someone owns them. Interviewer coaching produces a queue of named feedback that nobody reads unless a recruiting leader runs it. Guard: name the owner and a monthly review slot in the implementation plan, or buy the notes tier and skip the analytics premium.