Product Architecture and Main Functionality

Product positioning

Capablio should be positioned as an adaptive longitudinal feedback operating system. The differentiator is not merely a survey builder or a 360 report. It is the orchestration layer that decides which small set of questions to ask, to whom, when, why, and when not to ask again.

Primary personas

Capability map

Question and framework studio

• Construct taxonomy and competency framework editor
• Question-family and cadence-variant editor
• Answer-scale library
• Question quality checklist and linting
• Translation workflow
• Version comparison and migration
• Global, vendor, tenant, and imported source labels
• Approval and publishing workflow
• Deprecation and replacement mapping

Pack studio

• Create packs from questions and bundles
• Define anchors, rotation pools, mandatory groups, and optional groups
• Define respondent routing and role applicability
• Define quotas by dimension, relationship, and question type
• Simulate coverage and survey length before publishing
• Fork or overlay global packs without losing lineage

Scheduler and delivery

• Daily, weekly, bi-weekly, monthly, quarterly, annual, anniversary, and event-driven schedules
• Tenant timezone and fiscal calendar
• Quiet days, holidays, vacations, and blackout periods
• Email, Slack, Microsoft Teams, in-app, and link delivery
• Reminder and expiration policies
• Multiple active programs with a shared exposure ledger

Adaptive assembly

• Cooldown eligibility
• Anchor reservation
• PDP and event-trigger reservation
• Constrained weighted rotation
• Atomic and splittable bundles
• Coverage-cycle tracking
• Feasibility simulation
• Deterministic replay for audit and debugging

Feedback and 360

• Self, manager, peer, direct report/team, customer, partner, and custom relationships
• 90/180/270/360 configurations
• Internal and external raters
• Rater nomination and approval
• Minimum anonymity thresholds
• Relationship-specific questionnaires
• Qualitative prompts and evidence examples

PDP and coaching

• Goal creation from a report, import, manual selection, or coaching session
• Baseline, action, evidence, support, and outcome check-ins
• Cooldown overrides for linked questions
• Milestones, target dates, owners, and supporters
• Coaching notes with separate permissions
• Closure criteria and follow-up period

Analytics

• Individual trend, self-versus-others, and relationship views
• Construct and question-family drill-down
• Coverage, fatigue, response, and Not Observed dashboards
• Dispersion and confidence indicators
• PDP progress using multiple evidence types
• Team and organization aggregates with suppression rules
• Item-quality analytics for framework owners

Administration

• Organizational hierarchy and matrix relationships
• Role and attribute-based assignments
• CSV import and directory sync
• SSO, SCIM, API, and webhooks by entitlement
• Branding, localization, and communications
• Audit logs and data lifecycle controls
• Plan entitlements and usage metering

Main end-to-end workflows

Recurring pulse

1. Admin assigns a pack to a population and cadence.
2. Scheduler creates a survey window.
3. Assembly engine filters candidates against applicability, cooldown, exposure, and policy.
4. Mandatory, anchor, PDP, and rotating items fill the quota.
5. Survey is delivered through the preferred channel.
6. Responses are aggregated subject to anonymity thresholds.
7. Coverage and exposure ledgers are updated.
8. Manager and individual receive appropriate insights and actions.

Quarterly 360 pulse

1. Subject or admin chooses raters by relationship.
2. The platform validates anonymity and exposure sufficiency.
3. Relationship-specific variants are selected from the same construct families.
4. Results are shown by relationship only when the minimum group size is met.
5. Development themes may be proposed, but the individual confirms what becomes a PDP objective.

Annual deep review

1. A larger anchor set is used to preserve year-over-year comparability.
2. Rotating modules cover strategic priorities and tenant-specific capabilities.
3. Qualitative prompts ask for continue/start/stop/risk/evidence.
4. Report includes trends from recurring pulses, but does not average incompatible scales or observation horizons.

External assessment import

1. Admin or individual uploads a supported file or structured export.
2. Adapter stores the original source and extracts competencies, items, scores, relationships, dates, and comments.
3. Mapping is exact, semantic, manual, or unmapped.
4. Imported evidence appears as a separate baseline layer.
5. A user may select imported themes for a PDP; no automatic blending is performed.

Logical architecture

Administrative surfaces

1. Library — constructs, families, variants, scales, translations, status.
2. Packs — composition, quotas, relationships, cadence defaults, coverage simulation.
3. Programs — assignments, schedules, channels, populations, reminders, quiet periods.
4. People — subjects, managers, teams, matrix relationships, external raters.
5. PDP — templates, trigger policies, active goals, overrides, milestones.
6. Analytics — program health, coverage, fatigue, response quality, trends.
7. Imports — Thomas and generic CSV imports, mapping review, lineage.
8. Governance — approvals, audit, anonymity, retention, deletion, export.
9. Billing — subject seats, plan entitlements, usage limits, add-ons.