All of talyzr's capabilities

Connect your existing HRIS (Buk, SAP SuccessFactors, Workday, Talana, Rankmi) or upload a universal CSV/XLSX. Automatic daily sync + hourly incremental with exponential backoff and per-connector health monitoring. API tokens are AES-256 encrypted at rest. Mapped to an internal Common Data Model: your HRIS changes, talyzr doesn't break.

Composite criticality score (0–100) combining Business Impact, Replaceability, Knowledge Concentration, and Strategic Importance. Weights configurable by your organization (must sum to 1.0). Automatic tiers: CRITICAL ≥80, HIGH ≥60, MEDIUM ≥40, LOW <40. Bulk import via CSV/XLSX to classify hundreds of roles in minutes. Every change is audited with who and when.

Explainable formula: TRI = criticality×0.45 + (100−coverage)×0.35 + flight_risk×0.10 + trajectory×0.10. Automatic ranking of the most urgent roles with tiers CRITICAL ≥75, HIGH ≥50, MEDIUM ≥25, LOW <25. Unlike an opaque ML model, TRI is deterministic: CHROs can explain to the board exactly why a role is red.

Succession coverage visualization by org unit with 4 levels: STRONG (≥2 ready-now or 1 ready-now + 1 ready-1-2-years), ADEQUATE (≥1 ready-now or ≥2 ready-1-2-years), WEAK (no candidates ready in <2 years), NONE (no successors). Filters by department, criticality tier, and coverage level. CSV or PDF export for board presentations.

0–100 global data quality score across three dimensions: completeness, freshness, and consistency. Top 10 detected issues with a suggested action for each (e.g.: 'Juan Pérez has no manager assigned', 'critical role without successors'). If your data is bad, your decisions are too. Data Health turns data cleanup into a trackable task instead of frustration.

Hybrid architecture combining explicit rules (always guarantee a response) with Claude Sonnet 4.6 (enriches with qualitative context). 7 action categories: HIRE, DEVELOP, UPSKILL, REDEPLOY, RETAIN, REVIEW, OTHER. Each recommendation includes title, description, urgency, suggested owner, success metric, and learning resources. If the AI fails for any reason, deterministic rules respond — you'll never see a blank screen.

Full Kanban: PENDING → IN_PROGRESS → COMPLETED / DISCARDED. Each action has owner, deadline, description, link to source role and AI recommendation if it originated from one. When marked COMPLETED, the system requires mandatory feedback to close the learning loop. Automatic alerts on upcoming deadlines, overdue actions view, email notifications when new actions are assigned.

6 types of automatic alerts generated daily: vacant critical role, no successor, low readiness, flight risk on successor, stale assessment, overdue evaluation cycle. CRITICAL/HIGH/MEDIUM/LOW severity with auto-resolution when the condition disappears. Dispatch to multiple channels (Email + Slack + Teams + custom Webhook) with automatic dedupe to avoid spam.

Create persistent and shareable scenarios: what happens to coverage if Juan leaves? how much does TRI rise if I let 3 people go? how does it change if I promote María to Director? Each scenario shows TRI delta, most impacted positions, and mitigation recommendations. Can be shared with your team or the board before making the real decision.

Structured Talent Review workflow with 3 stages: talent nomination, manager calibration, final approval with decision freezing (frozenAt). Immutable PRE and POST calibration snapshots for audit. Minute export in PDF or TXT to share with stakeholders. Cycle decisions are automatically materialized into real succession (update tier, add candidates, archive roles).

ESCO taxonomy (European Skills, Competences and Occupations) with ~13,890 skills classified into Skills, Knowledge, and Attitudes. European standard, multilingual, no need to invent your own ontology. For each critical role, AI generates a required skills profile with minimum level (AWARENESS / WORKING / PRACTITIONER / EXPERT). HR validates the generated profile before it's used in gap analysis.

Launch skill assessment campaigns with defined deadlines. Select target persons, choose mode (full role profile or specific skills), set deadline, and launch. Skills and required level are frozen at campaign launch: if you later change the role profile, it doesn't affect the in-progress campaign. In-app notifications to targets and their managers at launch.

Register your company's learning resources (internal courses, LinkedIn Learning subscriptions, custom programs) and link them to ESCO skills. When AI generates UPSKILL recommendations for a role and the skill matches a resource in your company, the internal resource is injected into the prompt and appears in the final recommendation with priority over generics like Coursera or edX.

Add-on activatable on any plan. Unlimited 360° evaluation cycles with all stages: self-assessment, manager review, skip-level approval, peer 360, calibration with PRE/POST snapshots, feedback, AI IDP generation, and closure with employee acknowledge. Custom templates with 6 section types (competency grid, potential scale, free text, checkboxes, rating, comments). Decision freezing at stage close.

Dashboard designed for the CHRO: total critical roles, average TRI, distribution by tier, succession coverage, average skill gaps, AI adoption (acceptance, thumbs up, latency), risk by area, top 10 riskiest roles, recent alerts. Not 'generic dashboards': these are the KPIs the CHRO needs for a 30-minute conversation with the CEO.

Classic 9-box grid (performance × potential, 3×3 quadrants) interactive with drill-down to person. Compare across cycles to see talent movement. Talent health KPIs: % top talent, % flight risk, positive mobility, top stability, cycle coverage. Multi-cycle trends in charts. Useful for board presentations and HR internal analysis.

Auto-generated monthly executive talent report: executive summary (with AI or rules as fallback), KPIs with YoY/QoQ delta, 9-box distribution, emerging risks, action recommendations. Export to 5-slide PPT via pptxgenjs — ready to present to the board without assembling anything. Automatic monthly snapshot for historical comparisons. Configurable auto-send schedule (monthly/quarterly) with recipient list.

AI agent that queries your company's REAL data in real time (not a generic chatbot). Access to 12 tools: executive KPIs, risk ranking, 360 profile of people, skill gaps, alerts, succession, org hierarchy, role search, department view, pending actions. Uses pgvector + Voyage embeddings for semantic retrieval over your data + Claude Sonnet for reasoning. Tool-use loop max 5 rounds. Zero hallucinations: if it can't find it, it says so.