Invariant eQMS Roadmap

A single place for anyone in the company to submit a request to the quality team — with priority, description, and attachments. QA sees every open request in one queue, prioritizes and assigns it, and the requester can see when it's resolved. Related requests can be grouped — for example, batching document-change requests into the next controlled revision.

Capture and classify product complaints in a controlled quality workflow, where Complaints is enabled for your organization. Implemented scope: Manual complaint entry, including source channel and reported product/lot context Organization-managed type/subtype taxonomy Linked CAPA or deviation follow-up and supported evidence attachments Severity/status tracking and supported signed closure/reopen controls A source-channel field does not mean the system reads your inbox. Email/web intake, complaint-specific AI extraction and classification, automatic regulatory reporting, and a public complaint portal are not available workflows. Start here: https://docs.elevatedsignals.ai/invariant/reference/complaints September 11, 2026 — Choose linked quality records by their readable labels instead of pasting identifiers. Actions now show more specific errors and prevent duplicate submissions while a request is in progress. Existing Complaints availability and permissions still apply. Guide: https://docs.elevatedsignals.ai/invariant/reference/complaints

Connect a configured CAPA retraining action to training assignments and signed, passing completion evidence. Where the CAPA retraining loop is enabled: Configure explicit users and/or supported platform roles as the audience Mark the retraining action complete to issue the assignments; adding the action alone does not issue them Each required assignment must have signed, passing completion evidence before the CAPA can finish, have effectiveness verified, or close This is not a general automatic retraining engine for every SOP change. Audiences are not inferred from job titles. Email/SMS delivery is not part of this workflow. Walkthrough: https://docs.elevatedsignals.ai/invariant/how-to/capa-retraining-loop

Get answers without leaving Invariant. A built-in Help Center provides task-oriented articles — how-to guides, reference, and explanations — searchable by title or heading, right inside the app. What you get today: In-app article library organized for quick answers Search across titles and headings Contextual guidance alongside the module you're working in

Attach supported evidence files to supported quality records where attachments are enabled. The workflow records a SHA-256 file fingerprint and checks the stored bytes on download against the recorded value. This detects a mismatch within the implemented verification path. It is not a promise of WORM storage, provider-enforced retention, or protection against a privileged rewrite of both stored evidence and its verification history. Supported record types, file formats and limits are defined by the current release. See: https://docs.elevatedsignals.ai/invariant/how-to/attach-evidence

Manage user accounts and import users from a CSV where bulk user import is enabled. Administrators can invite and deactivate users and assign supported roles. Implemented scope: User administration within an organization Validated CSV user import with explicit handling of existing accounts Audited provisioning and role changes through supported workflows This is not a general migration service for SOP content, training matrices or historical quality records; supported row imports are described below. It is not self-serve organization signup, SSO, or automatic consultant access across separate client organizations. Plan access and onboarding with Elevated Signals: https://docs.elevatedsignals.ai/invariant/quickstart September 11, 2026 — Where the relevant import capability is enabled, Validate saves an audited validation report with predicted counts. It does not create the predicted users, documents, deviations, invitations or emails. Each validation appends a record and can fail if its audit write fails. Supported document/deviation imports can commit valid rows while reporting invalid rows; malformed files are still refused. Actual import is a separate action, with its own permissions and effects. Invitation withdrawal and deactivation handle outstanding invitations more consistently; the optional invitation-visibility screen remains separately enabled. Guide: https://docs.elevatedsignals.ai/invariant/reference/settings

Track quality work with an owner, a due date, a status, and a link to the quality record it supports. Review work in list or Kanban views and see which tasks are overdue. Implemented scope, where enabled: Assigned quality work items linked to supported records Controlled owner, due-date and status changes with reasons and audit history Content-bound completion signatures where that control is enabled This is a quality task tracker, not a full project-management suite. Sub-task trees, time tracking, task comments, Gantt timelines and email/chat escalation are not included. In-app alerts are a separate capability. Start here: https://docs.elevatedsignals.ai/invariant/reference/quality-tasks September 11, 2026 — Dated quality work items now appear on the quality calendar alongside other supported dated records. This does not add recurring-task generation, a document-routing Gantt or a general project-management suite. Guide: https://docs.elevatedsignals.ai/invariant/reference/quality-calendar

A formal extension-request workflow (reason, approver, new due date, full audit trail) so that a legitimately extended CAPA or deviation — for example, one waiting on a third-party lab result — doesn't simply appear as a breached SLA on the dashboard.

Let each organization choose whether closure-time and SLA math is calculated on calendar days or working days (excluding weekends and holidays), with configurable org calendar, timezone, and holiday policy. Different programs report this differently — tell us how yours does it.

Set different closure-time targets by severity — for example, a materially shorter window for critical deviations and CAPAs than for routine ones — instead of a single org-wide target. We're actively shaping the tier structure and how it's surfaced on the dashboard, and would love your input on what tiers match your program.