What is ReQuest DESK?

ReQuest DESK is an AI-first service management platform. You describe what your organisation does — or point it at a document — and it builds your service catalog, classifies every incoming request across email, chat, WhatsApp and voice, and routes each one to the right owner with a committed resolution time.

How is it different from a traditional service desk or an AI chatbot?

Traditional platforms make you configure the catalog by hand over months. AI-augmented tools bolt a chatbot onto an existing ticket form. ReQuest is AI-first: the catalog generates itself from your own documentation, and AI classification is the core of the request flow — not an optional add-on.

How long does it take to go live?

Days, not months. The AI Builder session, catalog review and portal configuration can typically be completed in a single working day, and the platform delivers value from the first day of operation.

Do I need an IT team or technical specialists?

No. ReQuest removes the specialist configuration burden entirely — the AI does the work that consultants and administrators have always had to do by hand.

Which channels does it support?

Web portal, embedded chat widget, WhatsApp and other messaging apps, and voice (transcribed). Every channel produces text, which the same AI classifier processes identically.

How does it avoid wrong answers or hallucination?

Every request is matched against your own service catalog and knowledge base and confirmed with a confidence score, so answers come from what you actually offer. The platform routes anything outside known patterns to a human with context.

What is the AI allowed to do on its own?

Only what you grant. Autonomy is set per request type and per resolution step — the AI can own a step, hand it to a human, or wait behind an approval gate. High-stakes actions like refunds, policy exceptions or clinical questions are typically gated: the AI prepares everything, a human approves. Full autonomy is a switch, never a default.

Does it integrate with the systems we already use?

Yes — through connectors and skills. Connectors let the desk act inside your existing tools (calendars, CRMs, billing, ticketing) so your systems stay the source of truth. Skills package how to operate one of your systems or procedures — teach it once, and every request type that needs it can use it.

Can it handle bookings and scheduling?

Bookings are built in. Appointments, resources and time slots are first-class: the desk offers times, books them, and absorbs reschedules.

No. It works for salons, clinics, facilities teams, professional services, trades and local government just as well as IT — any organisation that delivers defined services.

Does it work for MSPs managing multiple clients?

Yes. A single deployment supports multiple scoped catalogs through the service registry — one platform, N clients, each with its own branded portal.

Priced for access, not for enterprises — no six-figure license and no implementation tax. Start with a pilot and go live in days, then scale as your service layer grows.

MSPs & IT TEAMS

One AI service desk, every client: ReQuest DESK for MSPs

“Onboarding a new client used to mean two weeks of catalog workshops. Now it's one AI Builder session — with a branded portal live by the afternoon.”

The situation

An MSP's margin lives in onboarding and routine tickets. Every new client historically meant workshops to map services, weeks of catalog configuration, and then a steady stream of password resets, how-to questions and status checks that consume engineer hours without needing engineer judgment.

Multiply by ten or thirty clients and the picture is familiar: the tools that promise to fix this were built for one enterprise, not for N small ones.

Going live

ReQuest's registry supports multiple scoped catalogs in a single deployment — one per client, each with its own services, request types, SLTs and branded portal. Onboarding a client becomes a step, not a project: feed the AI Builder the client's service agreement or a plain-language description, review the generated catalog, publish.

The classifier is a feature you can demo in the sales call: a request arrives in natural language, is matched to the client's catalog with a confidence score, assigned an owner, and committed to a resolution time — before an engineer is involved.

A day at the desk

● WhatsAppresolved by AI

“Locked out — my MFA app was on the phone I traded in yesterday. Need access before my 10:00.”

→ Identity verified per policy, MFA reset, all sessions cleared — resolved before the meeting.

● Emailresolved by AI

“I deleted the whole Q3 planning folder from the shared drive an hour ago. Can you get it back?”

→ Restored from last night's backup plus version history, verified with the requester — resolved end-to-end.

● Webrouted to a human

“Give our new external agency edit access to the Brand Assets site — they start Monday.”

→ External access is a gated request type: the AI prepared a guest-access plan and routed it to the data owner for written approval.

What stays human

Access and security changes never auto-execute. New accounts, permission changes, offboarding — each is a gated request type where the AI does the preparation and a human gives the approval, exactly as the client's security policy says. P1 outages page the on-call engineer immediately; the AI opens the incident timeline while they pick up.

What changed

Client onboarding compressed from weeks of workshops to a working session
Routine tickets resolved without engineer time — resets, restores, how-tos
One deployment, N clients: no per-client platform configuration
A demonstrable AI capability competitors can't show in a sales call

Sounds like your desk? See it on your services.

Describe what you do — a working catalog and portal, live in days. No enterprise contract.

Join the waitlist Request a demo