Greenhouse vs Workable

Greenhouse

Structured interview kits, scorecards and scheduling ship in the Core plan and an audit log arrives in Pro — more than an inbox with a status field, and the structured-hiring pitch is at least specific. We found no public information on hiring-manager access scoped to their own vacancies, on calendar-connected scheduling, on German-language candidate emails, or on a per-candidate history of every stage change and who made it.

Workable

Requisition approvals with multi-level sign-off, offer approval flows, granular access rights configurable at each stage, and interview kits with scorecards are all stated, and audit trails tracking all changes plus logged, overridable agent decisions give a record of who did what. I found no public information on time-to-hire or drop-off reporting per stage, on bulk actions, or on exporting a candidate's full decision history for a discrimination dispute, so this stops short of an auditable fair-process record.

Greenhouse

MyGreenhouse gives candidates real-time status updates, the one concrete candidate-experience fact on the page. We found no public information on multiposting, on the Bundesagentur für Arbeit or any named German job board, on a hosted career page, or on mobile application without an account, so nothing here evidences how a vacancy reaches German candidates at all.

Workable

A branded careers-page builder with a website jobs widget, distribution to 200+ boards, Indeed and LinkedIn named as partner channels, and mobile-optimized applications with auto-fill are stated, and candidates can withdraw online directly. I found no public information naming German boards such as the Bundesagentur für Arbeit or a major German paid board, no Google for Jobs markup, no per-board booking or budget tracking, and no source attribution per applicant.

Greenhouse

Deletion is promised only "when we have no ongoing legitimate business need", plus a ninety-day deletion queue tied to contract termination rather than to a configurable deadline after rejection; a data-subject request portal and a Pro-tier audit log are the concrete artifacts. We found no public information on recorded talent-pool consent, per-country retention rules, or how deletion propagates to backups and synced boards — and the privacy policy states candidate personal information may be used to train AI models.

Workable

The retention statement holds data as long as needed to provide the services or fulfil the original purpose — a purpose, not a configurable deletion deadline — and deletion happens on request; the talent CRM is marketed with no stated consent step, expiry or re-consent mechanism. GDPR rights free of charge and a DPO contact are documented, which covers answering an access request, but I found no public information on automatic deletion after rejection, deletion-run logs, view or export logs on applicant files, or how deletion propagates to backups and boards.

Greenhouse

AI Notetaker, Voice AI, Talent Matching, AI-powered report filters and Real Talent fraud and identity checks are sold as benefits, the master agreement states Greenhouse makes no hiring decisions, and a bias audit statement and AI principles are published. We found no public information on what any of these features evaluate, on a per-feature switch to turn them off, or on the model provider — and the agreement incorporates DORA and EU Data Act addenda while we found no comparable public EU AI Act position.

Workable

An account-level kill switch with data left intact, configuration per account, per job and per candidate of screening settings and scoring weights, reasoning shown with every score, logged reversible actions, low-confidence cases passed to a human, and protected characteristics excluded by design are all stated, alongside fairness-research participation and an EU AI Act alignment claim. The providers behind the third-party large language models are not named on the captured pages, I found no public information that candidates are informed when AI is used or that AI processing can run in the EU, and the privacy policy's statement that Workable does not engage in automated decision-making sits beside marketing that clear no's are routed out automatically — the captured pages give different pictures of automated rejection.

Greenhouse

An API exists with a free export key and contractual usage limits, SSO sits in Core, a Business Intelligence Connector in Plus and a developer sandbox in Pro, next to "hundreds of pre-built integrations" that the captured pages do not name. We found no public information on a handover to any HR system common in DACH, on calendar or video-interview integration, e-signature, SCIM or webhooks, so the hire-to-HR step is unevidenced.

Workable

A documented REST API with webhooks, MCP access, calendar and Zoom/Teams/Meet integration, e-signatures, single sign-on, and automatic export of a defined field set when a candidate reaches the hired stage are stated. I found no public information naming any specific HR system — not Personio, not a DATEV-connected system, not SAP SuccessFactors — since the handover page is a partner-program page, and there is no public information on SCIM, published API limits, or a deprecation policy.

Greenhouse

The contracting entity is Greenhouse Software, Inc. under exclusive New York jurisdiction, with primary offices in the United States, processing that "may be processed in any of these countries", and the EU-U.S. Data Privacy Framework as the stated safeguard. We found no public information on hosting location, on a subprocessor list, or on where CV parsing and the AI features run — and every applicant record here belongs to a third party who never chose this vendor.

Workable

The privacy policy states the services are hosted and operated in the United States through Workable, Inc., with storage on servers in the USA as well as possibly the EEA but no customer-selectable EU region described, and the contracting entity is the London company under English law with transfers relying on SCCs and the Data Privacy Framework. A sub-processor list exists as a link while the hosting provider itself is not named in the captured text, and where the third-party AI models process candidate data is not stated — for applicant records, this is US-default hosting with no EU option described.

Greenhouse

Tier names and feature lists are public and the master agreement fixes the basis — employee headcount including affiliates, due in advance, non-cancelable, auto-renewing in one-year terms with renewal price increases reserved. The captured pricing page shows no price figure for any tier, and API usage fees, professional services and migration rates all defer to the order form, so the annual invoice cannot be computed from public pages.

Workable

Three bundles are priced with add-ons itemised — texting, video interviews, assessments, performance — and AI is priced unusually concretely in credit bundles with per-action costs, one-year expiry, a renewal cap of five percent with 45 days' notice, and a 15-day trial without a credit card. But the captured pricing pages do not say what the monthly price is based on — jobs, employees or recruiter seats — and I found no public information on job-board posting costs or the booking model, the minimum term, or VAT treatment, so a real annual invoice for a given company size cannot be computed from public pages alone.