A flexible multi-CRM workspace for teams that need focused pipelines, shared customer context and accountable execution.
Thunder Max CRM gives growing teams a structured workspace for leads, contacts, deals, follow-ups and customer operations. Separate CRM workspaces can stay focused while owners retain the visibility, permissions and subscription controls they need.
Thunder AI CRM is an offline-first business operations platform combining customer acquisition, sales, service, marketing, finance, procurement, maintenance, workforce workflows, analytics and governance in one recoverable application.
Use this chapter to establish the intended users, business outcomes, implementation sequence and explicit non-uses before introducing the system into a department, plant, service organization or multi-team operation.
Thunder Max is a single-host SQLite application suitable for controlled departmental and small-to-medium deployments with testing. It is not a machine-control, safety, accounting ledger, MES, QMS, SCADA or legally validated signature system.
Navigate the complete module catalogue: core CRM, revenue and finance, growth, operations, industrial execution, workforce, governance, reports, custom apps and the AI assistant. Each area is protected by route and permission checks.
The product uses FastAPI, server-rendered Jinja templates, browser JavaScript, SQLite WAL, file-backed uploads, background jobs, migrations, security middleware and a responsive PWA shell. Understand the request, database and job boundaries before integrating.
General, healthcare, real estate, education, insurance, banking, manufacturing, retail, hospitality, logistics, automotive, e-commerce and recruitment packs reuse the same engine with tailored terminology, stages and fields.
Company scope, active CRM workspace, role, module access and action permissions work together. Use least privilege, unique accounts, controlled administrator membership and joiner/mover/leaver reviews for production governance.
Follow repeatable runbooks for shift start, lead-to-customer conversion, service resolution, quotation-to-collection, marketing execution and approval-controlled work. These procedures turn features into accountable operating practice.
Use Thunder as an operational coordination layer across enquiry, quotation, order, procurement, shipment, inspection, maintenance, warranty and service work while keeping safety-critical and legally authoritative records in their required systems.
Deterministic scoring, win probability, sentiment, churn radar, next-best-action, forecasts, briefings and the assistant are explainable. Automation rules connect events to bounded actions; human approval remains required for consequential decisions.
Configure the public URL, allowed hosts, data directory, administrator bootstrap, currency, industry, SMTP, model adapters, reverse proxy, persistent storage and production process manager. Windows, Linux, Docker and AWS paths are documented.
The baseline includes PBKDF2 password hashing, secure sessions, throttling, origin checks, security headers, parameterised SQL, path validation, RBAC, audit trails and controlled public endpoints. Add organizational retention and review controls before regulated use.
Use online SQLite snapshots, JSON record backups, upload-directory backups, off-host copies, retention, restore validation and recovery drills. Define RPO and RTO for the actual single-host architecture rather than assuming high availability.
Extend the product through route modules, migrations, validation helpers, permissions, templates, tests and the custom App Builder. Stable integrations must use an explicit versioned interface with authentication, scoping, idempotency and auditability.
A professional release requires dependency checks, compilation, linting, route and security tests, migration tests, backup/restore tests, permission denial tests, representative workflows, health checks and reproducible deployment artifacts.
Monitor 5xx/503 rates, readiness, storage, WAL growth, backups, login throttles, attachments, stale work, SLA breaches, low stock, risks, restarts and proxy errors. Use the symptom-to-cause runbook before escalating.
Before go-live, obtain named owners, approved data classification, tested access reviews, configured SLAs and approvals, HTTPS, persistent storage, externalized secrets, backup recovery evidence, capacity measurements and business sign-off.
Understand the relationships among leads, contacts, deals, activities, tickets, tasks, quotations, products, approvals, attachments, work orders, inspections, maintenance, risks and audit entries before designing reports or imports.
Keep executable behavior and tests authoritative, review this guide with each release, record operational changes, and require documentation updates when routes, schemas, permissions, workflows, deployment or recovery behavior changes.
This guide is maintained from the Thunder AI CRM professional documentation and the application’s current routes, security controls and deployment notes.
Open Thunder Max CRM →