Plum
Plum is a pre-employment talent assessment platform that uses a Five-Factor (Big Five) psychometric Discovery Survey and a Talent Match model to score and rank candidates on behavioral "durable skills" for role fit, hiring, and development.
HireAIScore rates Plum 54 out of 100 (grade F, Substantial gaps) under rubric v1.0, last reviewed June 7, 2026. Plum ranks 12 of 49 vendors rated in Assessment AI, against a category average of 48. That score is drawn from 19 evidence items across 7 rubric criteria for Plum — 15 with a cited source, 4 recording that nothing was located.
§ 01 - Company facts
- Legal name
- Plum.io Inc.
- Founded
- 2012
- Headquarters
- Canada · CA
- Pricing tier
- Mid
- Market side
- Employer-side (AEDT)
- Categories
- AssessmentScreening
- Website
- plum.io
These are company details, not findings. They are editable by a verified representative and carry no weight in the score, which is built only from the cited evidence below.
§ 02 - Score breakdown
§ Score breakdown
Category scoring
Weighted contribution shown to the right of each bar.
- 01
Article 11 Technical Documentation
Weight 20%58
+11.6 · category avg 53
- 02
Bias Audit Transparency
Weight 18%83
+14.9 · category avg 47
- 03
FRIA Support
Weight 15%28
+4.2 · category avg 32
- 04
Data Governance Disclosure
Weight 15%54
+8.1 · category avg 55
- 05
Human Oversight Design
Weight 12%52
+6.2 · category avg 55
- 06
Post-Market Monitoring
Weight 12%36
+4.3 · category avg 39
- 07
Customer Documentation
Weight 8%63
+5.0 · category avg 59
Category avg is the mean raw score on that criterion across the 49 Assessment AI vendors in scope of this rubric, this one included.
§ 03 - Strongest · weakest
Strongest category
Bias Audit Transparency
Raw score 83 · contributes 14.9 to total.
83 against a 47 category average
Weakest category
FRIA Support
Raw score 28 · contributes 4.2 to total.
28 against a 32 category average
§ 04 - Cited evidence
Download diligence record→§ Evidence
Cited per category
Every score is backed by at least one cited piece of evidence.
Evidence ledger
- Items
- 19
- Documentation
- 12
- Audit report
- 3
- Absence
- 4
- With a source URL
- 15 of 19
- Source hostnames
- 2
- Fewest items
- 1
- FRIA Support
These figures measure how thoroughly Plum was reviewed, not how Plum performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
3 items58
- DocumentationCaptured Jun 7, 2026
Plum Science page documents the Five-Factor (Big Five) basis, forced-choice design, and references reliability and validity testing of the Discovery Survey.
- DocumentationCaptured Jun 7, 2026
Trust Center lists a 'Technical Manual for Plum Assessment methodology' and a 'How Plum Uses and Controls Artificial Intelligence' policy, but these documents are gated behind an access-request form.
- DocumentationCaptured Jun 7, 2026
The FairNow audit includes a detailed system description of the Talent Match model (inputs, scoring 30-99, Talents derived from the Five-Factor Model) functioning as partial technical documentation.
https://use.plum.io/hubfs/Resources/Audit/2024-2025/Plum-Talent-Match-bias-audit-nov-2024.pdf
Bias Audit Transparency
3 items83
- Audit reportCaptured Jun 7, 2026
Independent NYC LL144 bias audit of the Plum Platform by Holistic AI, audit date Aug 23 2023 (distributed Nov 7 2023), ~20M post-cleaning records, gender/ethnicity and intersectional impact ratios all at or above 0.8.
https://www.plum.io/hubfs/Resources/Audit/2022-2023/2022-2023_Plum_Audit_Report_Summary.pdf
- Audit reportCaptured Jun 7, 2026
Independent NYC LL144 bias audit of Plum's Talent Match model by FairNow dated Nov 7 2024, covering 528,891 applications (28,881 with demographic data), finding no disparate impact across race, gender, and intersectional categories.
https://use.plum.io/hubfs/Resources/Audit/2024-2025/Plum-Talent-Match-bias-audit-nov-2024.pdf
- DocumentationCaptured Jun 7, 2026
Blog post explaining Plum's bias-audit approach (adverse-impact/outcome testing) and referencing NYC Local Law 144 requirements.
https://www.plum.io/blog/role-of-bias-audits-in-talent-assessments
FRIA Support
1 item28
- AbsenceCaptured Jun 7, 2026
No Article 27 Fundamental Rights Impact Assessment template or EU AI Act deployer guidance was found on the Plum site, Trust Center (plum.io/trust-center), help center, or in either published bias audit, which address NYC Local Law 144 only.
Data Governance Disclosure
3 items54
- DocumentationCaptured Jun 7, 2026
Privacy policy details data collected, retention periods (e.g. 12-24 months), subprocessors (AWS, Google, HubSpot, LinkedIn, ZoomInfo), GDPR rights, US data residency with standard contractual clauses, and a privacy@plum.io contact.
- DocumentationCaptured Jun 7, 2026
Trust Center lists Data Processing Agreements (GDPR and third-country), a current subprocessor list, Security Control Guidelines, and penetration-test archives for 2023, 2024, and 2025, though most are gated behind an access request.
- AbsenceCaptured Jun 7, 2026
No Plum-held SOC 2 Type 2 or ISO 27001/ISO 42001 certificate is publicly visible on the Trust Center; security relies on AWS infrastructure certifications and parent Phenom's ISO 27001, and no detailed training-data exclusion list is published.
Human Oversight Design
3 items52
- Audit reportCaptured Jun 7, 2026
Audit states all candidates are shown to the recruiter ranked by match score and that scores 'can be used to influence whether an employer decides to move a job seeker forward,' indicating human-in-the-loop decision support rather than automated rejection.
https://use.plum.io/hubfs/Resources/Audit/2024-2025/Plum-Talent-Match-bias-audit-nov-2024.pdf
- DocumentationCaptured Jun 7, 2026
Product page describes Plum generating Match Scores and auto-generated Structured Interview Guides for recruiters, framing Plum as decision support that reduces 'gut-feel' bias while humans conduct interviews and decide.
- AbsenceCaptured Jun 7, 2026
No public documentation of concrete override controls, per-jurisdiction toggles, audit logs, or in-product candidate-facing explainability beyond the Plum Profile report.
Post-Market Monitoring
3 items36
- DocumentationCaptured Jun 7, 2026
Trust Center publishes penetration-test archives for 2023, 2024, and 2025 and a Business Continuity / Disaster Recovery summary, showing a recurring security-testing cadence.
- DocumentationCaptured Jun 7, 2026
Privacy policy provides a privacy@plum.io contact and physical address, but no breach-notification protocol, public status page, or model-update changelog is detailed.
- AbsenceCaptured Jun 7, 2026
No public system status page, incident-disclosure channel, continuous-monitoring dashboard, model-update changelog, or dedicated security.txt/security contact was found for plum.io.
Customer Documentation
3 items63
- DocumentationCaptured Jun 7, 2026
Detailed public product pages (talent acquisition, high-volume early-career hiring, RPO/finance/logistics variants) describe the assessment, Match Score, Role Model, and interview guides.
- DocumentationCaptured Jun 7, 2026
Trust Center centralizes privacy notices, Master Service Agreement, GDPR and third-country DPAs, an AI control policy, and an audits-and-compliance section covering bias assessment.
- DocumentationCaptured Jun 7, 2026
Public blog and help-center content explain bias audits, NYC Local Law 144, and Plum's assessment science for deployers and candidates.
https://www.plum.io/blog/role-of-bias-audits-in-talent-assessments
§ 05 - Editorial notes
Company overview
Plum (Plum.io Inc.), founded in Kitchener, Ontario in 2012 (formerly Cream.HR) and acquired by Philadelphia-based Phenom in an acquisition announced April 28, 2026, is a talent-assessment platform built on industrial/organizational psychology and the Five-Factor (Big Five) personality model. Candidates complete the ~20-25 minute Plum Discovery Survey measuring personality, problem-solving, and social intelligence; employers define role requirements via a Match Criteria Survey, and Plum's Talent Match model outputs a 30-99 Match Score that ranks candidates by fit. Plum positions itself as predicting job performance 'four times better than a resume' and serves enterprise customers including Scotiabank, Whirlpool, and Intact Financial across 176 countries and 20 languages.
Regulatory exposure
As an assessment that scores and ranks candidates to influence whether they advance, Plum's Talent Match model is squarely an Automated Employment Decision Tool under NYC Local Law 144 and would likely be a high-risk AI system under Annex III of the EU AI Act when deployed in EU hiring. Plum's strongest signal is bias auditing: it has published two consecutive years of independent, downloadable NYC LL144 bias audits (Holistic AI for 2022-2023, FairNow for 2024-2025), both finding no adverse impact across gender, race, and intersectional groups on large samples. Gaps remain on the EU side: no public Article 27 FRIA template or deployer-obligation guidance, no ISO 42001, and no Plum-held SOC 2/ISO 27001 certificate is publicly visible (parent Phenom holds ISO 27001:2013; Plum inherits AWS infrastructure certifications).
Path to a higher score
Plum already clears the highest anchor on bias-audit transparency; to raise the rest it should publish its gated 'How Plum Uses and Controls AI' policy, Technical Manual, and an explicit human-oversight/explainability statement openly rather than behind a Trust Center access form, and add EU AI Act deployer guidance with an Article 27 FRIA template. Obtaining and publicly disclosing a Plum-specific SOC 2 Type 2 / ISO 27001 (or ISO 42001) certificate, and standing up a public status page, model-update changelog, and named security/incident contact, would lift its data-governance, technical-documentation, and post-market-monitoring scores.
§ Regulatory frame
What applies to assessment ai.
Assessment is the most-regulated surface in AI hiring. NYC LL 144, EEOC and OFCCP frameworks, Illinois HB 3773, and EU AI Act Annex III §4 all apply, with overlap on bias-audit obligations.
§ Compare
Build any comparison→Plum against its nearest-scoring peers.
In Assessment AI.
§ Others rated in Assessment AI
All assessment vendors→Ranked 12 of 49 by weighted total under rubric v1.0. The ordering is arithmetic on the rubric and carries no view on which tool suits a given hiring process.
- 09iMocha56F
- 10Pymetrics (Harver Games)56F
- 11HackerRank55F
- 12PlumThis profile54F
- 13Metaview54F
- 14Codility54F
Conflicts of interest
No vendor pays for placement, scoring, or removal. Casework - the consulting firm that operates this directory - provides paid services to some vendors. Any active or recent (within 24 months) commercial relationship is disclosed on the affected vendor profile and the review is reassigned to an independent reviewer. See the full policy on About.
Casework has no commercial relationship with this vendor.