# Evaluating CVs: Criteria and Evaluation Matrix > Evaluate CVs with clear criteria: job specification, evaluation matrix with example, AGG pitfalls and documentation. Read the guide now. URL: https://techport.ai/en/hr-software/bewerberauswahl --- 1. [HR Software](/en/hr-software)/ 2. Candidate Selection techport HR Suite # Evaluating CVs: Criteria, Evaluation Matrix and AGG Pitfalls Dozens of applications arrive for a single role, three people review them, and in the end each has a different favourite. That is rarely down to the applications. Usually, nobody defined the evaluation criteria beforehand. This guide shows how to evaluate CVs against clear criteria, what an evaluation matrix looks like, which pitfalls the Allgemeine Gleichbehandlungsgesetz (AGG, General Equal Treatment Act) presents, and what you should document. ## Does this sound familiar? * Each person on the selection panel reads the CV differently: one looks at qualifications, another at well-known employers. * A beautifully designed CV makes a stronger impression than a plain one, even if the content is weaker. * After a rejection, an applicant asks for the reasons, and nobody can substantiate them. * The works council wants to know what criteria were used for the selection. ## First the job specification, then the CVs Before you open the first CV, establish with the hiring department what the position requires. Base this on the tasks, not on the person who previously held the role. * **Must-have criteria** are prerequisites without which the task cannot be performed, such as a formal qualification or a professional licence. Keep this list short and permit equivalent evidence. * **Nice-to-have criteria** make the difference between suitable applications. They are weighted. * **Five to seven criteria** are usually sufficient. Every additional criterion takes time and dilutes the weighting. Formulate criteria in a way that can be evidenced in the CV. 'Team player' cannot be deduced from a CV; 'coordinated projects across multiple departments' can. Traits such as resilience or persuasiveness are better assessed in an interview or via a work sample. Methods for this are described in our consulting guide on [aptitude diagnostics](/en/hr-beratung/recruiting/eignungsdiagnostik). ## Evaluating CVs: what matters ### Professional qualification Degrees, vocational training and certificates necessary for the task. Pay attention to equivalence: a foreign qualification or years of practical experience can replace a formal qualification. ### Relevant experience The activities are what matters, not the job titles. 'Team lead' in a business with ten employees means something different than in a corporate group. Read what the individual was actually responsible for. ### Results and responsibility Specific details regarding budgets, projects, customer groups or introduced processes say more than a list of duties. ### Development and further training Has the person taken on new responsibilities, pursued further education or changed industries? This demonstrates how they handle change. ### What you should not evaluate Photo, age, marital status, origin, religion and health are not criteria. The design of the CV also says little about suitability, except for roles where design is part of the job. Gaps in the CV are not grounds for exclusion. They often conceal parental leave, caring for relatives or illness. Anyone who downrates gaps across the board risks indirect discrimination, such as on the grounds of sex or disability (§ 3 para. 2 AGG). Enquire during the interview if a gap genuinely matters for the position. ## Evaluation matrix: how to build one An evaluation matrix, often also called an applicant matrix, makes ratings comparable. Here is how to proceed: 1. **Enter criteria.** Must-have criteria as a pass/fail check ('fulfilled' or 'not fulfilled'), nice-to-have criteria with a weighting. 2. **Determine weights.** The weights of the nice-to-have criteria add up to 100 per cent. 3. **Scale with anchors.** A scale from 1 to 5 is customary. Describe for each criterion what 1, 3 and 5 mean. Without these anchors, each person assigns their own subjective '3'. 4. **Evaluate independently.** Two people evaluate separately and reconcile afterwards. Large discrepancies indicate that a criterion is unclearly formulated. 5. **Provide justifications.** Every score requires a sentence of evidence from the CV. ### Example matrix The following example is fictional and serves purely for illustration. Role: Procurement Officer. Anchors for the criterion 'Experience in operational procurement': 1 = no procurement experience, 3 = order processing and schedule tracking, 5 = dedicated product category with supplier responsibility. Criterion Type Weighting Applicant A Applicant B Commercial vocational training, degree or equivalent experience Must-have N/A fulfilled fulfilled Experience in operational procurement Nice-to-have 30% 4 5 Experience working with an ERP system Nice-to-have 20% 3 5 Negotiations with suppliers, proven results Nice-to-have 20% 5 2 English for supplier correspondence, at least B2 Nice-to-have 15% 4 3 Tender comparisons and RFQs Nice-to-have 15% 3 4 **Weighted sum** **100%** **3.85** **3.95** Calculation for Applicant A: 4 × 0.30 + 3 × 0.20 + 5 × 0.20 + 4 × 0.15 + 3 × 0.15 = 3.85. What the matrix reveals: Both are close to each other, with distinct strengths. Applicant B brings more system experience, Applicant A has proven negotiation successes. A difference of 0.1 points is no reason for a rejection. The matrix indicates who to invite and what to clarify during the interview, such as Applicant B's negotiation experience. The decision rests with the selection panel. ## AGG pitfalls in candidate selection The AGG also protects job applicants (§ 6 para. 1 sentence 2 AGG) against discrimination on grounds of race or ethnic origin, gender, religion or belief, disability, age or sexual identity (§ 1 AGG). Two rules make this delicate for employers: * **Prima facie evidence suffices.** If an applicant presents indications of discrimination, the employer must prove that none occurred (§ 22 AGG). * **Compensation.** Discrimination entails the risk of compensation claims. It is capped at three months' salary if the person would not have been hired even with a discrimination-free selection process (§ 15 para. 2 AGG). The claim must be asserted in writing within two months of receiving the rejection (§ 15 para. 4 AGG). Typical pitfalls: * **Age-related phrasing.** The Federal Labour Court (Bundesarbeitsgericht) viewed the offer to work in a 'young dynamic team' as an indication of age discrimination (judgment of 11/08/2016, 8 AZR 406/14). * **Mother tongue instead of language level.** The Hessian Higher Labour Court (LAG Hessen) considered the requirement 'German as a mother tongue' to be discrimination on grounds of ethnic origin (judgment of 15/06/2015, 16 Sa 1619/14). Require a language proficiency level, such as C1, if the task demands it. * **Years of experience as a cap.** Limiting roles to applicants in their first year of professional experience can indirectly disadvantage older candidates (BAG, decision of 18/08/2009, 1 ABR 47/08). * **Gaps and part-time work.** Across-the-board deductions for career breaks disproportionately affect women and individuals with disabilities. * **Overlooking severe disabilities.** Employers must examine whether vacant positions can be filled by severely disabled persons and involve the representative body for severely disabled employees (§ 164 para. 1 SGB IX). Public employers must invite severely disabled applicants to an interview unless they visibly lack professional suitability (§ 165 SGB IX). * **Job advert not neutral.** Even the job advertisement itself must not infringe the prohibition of discrimination (§ 11 AGG). ## Documentation: what you should record Good documentation makes decisions transparent and helps if a rejection is challenged. Record the following: * **the job specification with date**, before the first application is reviewed, * **the evaluation for each application** with score, justification and the names of the evaluators, * **the reason for invitation or rejection**, objective and referenced to the criteria, * **the involvement of the works council.** Regarding the works council: In companies with generally more than twenty employees entitled to vote, the employer must inform the council before every hire, submit the necessary application documents and seek its consent (§ 99 para. 1 BetrVG \[Works Constitution Act\]). General assessment principles require its consent (§ 94 para. 2 BetrVG), as do selection guidelines (§ 95 BetrVG), including when AI is used in their formulation (§ 95 para. 2a BetrVG). Clarifying early whether your matrix falls under this is advisable. Support is provided by our advisory on [HR software and the works council](/en/hr-beratung/mitbestimmung-und-compliance/betriebsrat) and the guide [Works Council and AI](/en/hr-software/betriebsrat-ki). Regarding retention: You may retain documents of rejected applicants for as long as claims must be anticipated. Arithmetically, this covers two months to assert a claim (§ 15 para. 4 AGG), three months to file a lawsuit (§ 61b para. 1 ArbGG) and a buffer for service of process. In practice, a period of approximately six months after the rejection therefore applies. After that, you delete the data (Art. 17 GDPR), unless the individual has consented to being included in a talent pool. Find out more in our advisory on [applicant data and GDPR](/en/hr-beratung/recruiting/bewerberdaten-dsgvo). ## How the techport HR Suite helps The [techport HR Suite](/en/hr-software) maps this process in two stages. **In recruiting**, you work with evaluation sheets containing criteria and a scale from 1 to 5. The hiring department submits its feedback directly in the system, and you plan interviews in rounds. A retention period with reminders and deletion runs for each candidate, and the audit log records who changed what and when. Learn more on the [Applicant Tracking](/en/hr-software/bewerbermanagement) page. **With AI candidate evaluation**, you analyse CVs based on approved criteria: * The AI suggests criteria derived from the job advert text. A human reviews, edits and approves them. Each approval creates an immutable version. * Criteria concerning age, gender, origin, native language or other protected characteristics under the AGG trigger a warning. Approval is then only possible with explicit confirmation. * Name, contact details, date of birth, marital status, nationality, religion and photograph are separated prior to evaluation. * Every evaluation comes with calculation steps, facts and verbatim citations from the CV. * The AI proposes, a human decides. The system never sends rejections automatically. Overriding is possible for each criterion; a justification is mandatory. * Criteria and logs can be exported. Via the Works Council Assistant, the works council receives read-only access to approved criteria, versions and logs. How CVs are parsed is explained in the [CV Parsing](/en/hr-software/cv-parsing) guide. The legal framework for AI in selection is described in [AI in Recruiting](/en/hr-software/ki-recruiting). ## Frequently asked questions How do you evaluate a CV properly? You evaluate a CV best against criteria established before reviewing applications. Derive must-have and nice-to-have criteria from the responsibilities of the role, weight the nice-to-have criteria, and allocate points on a scale with described benchmarks. Note a piece of evidence from the CV for each score so that the evaluation remains traceable. What criteria exist for candidate selection? Typical criteria for candidate selection include professional qualifications, relevant work experience, proven results, language and system skills, and further professional development. They should be derived from the position's tasks and be demonstrable in the CV. Personal characteristics such as age, origin, marital status or photographs are not permissible criteria. What is an applicant evaluation matrix? An evaluation matrix is a table in which all applications are evaluated against the same weighted criteria. Each criterion has a weighting and a scale, typically from 1 to 5, with described levels. The weighted sum shows who best matches the job specification. The decision nevertheless remains with the selection panel. How long may I retain job application documents? Documents of rejected applicants may generally be retained for approximately six months after rejection. This period stems from the two-month deadline for AGG claims, the three-month deadline for bringing an action under § 61b ArbGG, and a buffer for service of process. Consent is required for longer storage within a talent pool. Does the works council have to approve the evaluation matrix? The works council must give consent if the matrix constitutes a general assessment principle or selection guideline (§§ 94, 95 BetrVG). This can particularly be the case if it is used uniformly across numerous positions. Clarify this early and disclose criteria and weightings transparently. This also applies when AI is involved. ## Sources * Allgemeines Gleichbehandlungsgesetz, §§ 1, 3, 6, 11, 15, 22: [https://www.gesetze-im-internet.de/agg/](https://www.gesetze-im-internet.de/agg/) (abgerufen am 23.10.2026) * § 61b Arbeitsgerichtsgesetz: [https://www.gesetze-im-internet.de/arbgg/\_\_61b.html](https://www.gesetze-im-internet.de/arbgg/__61b.html) (abgerufen am 23.10.2026) * Betriebsverfassungsgesetz, §§ 94, 95, 99: [https://www.gesetze-im-internet.de/betrvg/](https://www.gesetze-im-internet.de/betrvg/) (abgerufen am 23.10.2026) * SGB IX, §§ 164, 165: [https://www.gesetze-im-internet.de/sgb\_9\_2018/\_\_164.html](https://www.gesetze-im-internet.de/sgb_9_2018/__164.html) (abgerufen am 23.10.2026) * Verordnung (EU) 2016/679 (DSGVO), Art. 17: [https://eur-lex.europa.eu/eli/reg/2016/679/oj](https://eur-lex.europa.eu/eli/reg/2016/679/oj) (abgerufen am 23.10.2026) * BAG, Urteil vom 11.8.2016, 8 AZR 406/14: [https://www.bundesarbeitsgericht.de/entscheidung/8-azr-406-14/](https://www.bundesarbeitsgericht.de/entscheidung/8-azr-406-14/) (abgerufen am 23.10.2026) * LAG Hessen, Urteil vom 15.6.2015, 16 Sa 1619/14, Zusammenfassung: [https://www.betriebsrat.de/rechtsprechung/br/urteil/arbeitgeber-darf-gute-deutschkenntnisse-von-bewerber-verlangen](https://www.betriebsrat.de/rechtsprechung/br/urteil/arbeitgeber-darf-gute-deutschkenntnisse-von-bewerber-verlangen) (abgerufen am 23.10.2026) * BAG, Beschluss vom 18.8.2009, 1 ABR 47/08, Zusammenfassung: [https://www.hensche.de/Rechtsanwalt\_Arbeitsrecht\_Urteile\_Diskriminierung\_Betriebsrat\_BAG\_1ABR47-08.html](https://www.hensche.de/Rechtsanwalt_Arbeitsrecht_Urteile_Diskriminierung_Betriebsrat_BAG_1ABR47-08.html) (abgerufen am 23.10.2026) ## Further reading [techport HR Suite HR Software HR software with personnel administration, recruiting, AI candidate evaluation and company knowledge on one platform. Hosted in Germany. Request a demo now. ](/en/hr-software)[techport HR Suite Digital Personnel File Digital personnel file with master data, contracts and documents, role-based permissions and access logging. Part of the techport HR Suite. Request a demo now. ](/en/hr-software/digitale-personalakte)[techport HR Suite Leave Management Leave management with requests, remaining leave, team calendar, and sickness notifications. Managers approve in the browser, HR keeps track. Request a demo. ](/en/hr-software/abwesenheitsverwaltung)[techport HR Suite Onboarding and Offboarding Onboarding software starting right from offer acceptance: personnel files and tasks for HR, IT, and managers are created automatically. Including offboarding. Request demo. ](/en/hr-software/onboarding) More from techport.ai [ Software Custom process software for mid-sized companies. ](/en/loesungen)[ IT consulting Strategy, architecture, operations and security. ](/en/it-beratung)[ IT maturity check Ten minutes to a clear position. ](/en/it-beratung/reifegrad-check)[ HR maturity check 24 statements, a result per field. ](/en/hr-beratung/reifegrad-check)[ Funding BAFA grant plus more than 50 programmes for delivery. ](/en/foerderung)[ Process in practice How workflows become reliable software. ](/en/sop-praxis)[ Data and AI Analysis, forecasts and assistance systems. ](/en/daten-ki)[ About us The people behind techport.ai. ](/en/ueber-uns)