Pricing

The price is on the page.

Per endpoint, per month. No quote, no discovery call before a number, and the volume bands are published too. Buy Essentials or Professional online and your console is live in minutes — or start a free trial on your own machines, with no card at all.

Billed yearly. Roughly 20% below the rolling rate.

Essentials

One organisation, every core capability.

£1.80per endpoint / month

From 10 endpoints

Buy Essentials

Card payment. Your organisation is created automatically.

Internal IT teams managing their own estate.

Bring your own model
  • Unattended remote control and attended support sessions
  • Continuous telemetry: CPU, memory, disk, services, processes
  • Windows Update patch scanning and deployment
  • Software inventory and remote uninstall
  • Script execution across devices, sites or the whole estate
  • Alert rules with email delivery
  • Wake-on-LAN, reboot, and device lifecycle
  • Audit trail of every operator action
  • Hosting, backups and platform upgrades included — no server to run

Not in this tier

  • Multi-tenancy — Essentials manages one organisation
  • Session recording
  • Single sign-on and API access
Most chosen

Professional

The MSP tier. Every client, properly separated.

£2.90per endpoint / month

From 50 endpoints

Buy Professional

Card payment. Your organisation is created automatically.

Managed service providers running client estates.

1,000 Denwen AI actions per month, or bring your own
  • Everything in Essentials
  • Multi-tenancy: unlimited tenants, customers and sites
  • Per-tenant isolation enforced at every route, not by filtering a shared view
  • Session recording with per-tenant retention
  • Single sign-on (OIDC) — your directory, or each client’s own
  • API keys scoped to the organisation that issued them
  • Scheduled automation and a shared script library
  • Webhook and email alert channels with delivery history
  • Per-tenant SMTP so a client’s alerts leave through their own relay

Enterprise

For estates that get audited.

£4.20per endpoint / month

From 250 endpoints

Talk to us

Negotiated terms, a response SLA and named onboarding — so this one is agreed, not checked out.

Large providers and regulated environments.

10,000 Denwen AI actions per month, or bring your own
  • Everything in Professional
  • Editable roles with per-assignment site scope
  • Tenant hierarchy for providers who resell through partners
  • Data-protection register: what is held, for how long, and who can reach it
  • Per-tenant retention policy and the recording notice shown on the endpoint
  • Recordings encrypted at rest under platform-managed keys; signing keys held in custody
  • A named hosting region, stated in the contract rather than assumed
  • AI governance: which tenants, which model, what may be in a prompt
  • Priority support with a response SLA and named onboarding
Estimate

What it costs for your estate.

Move the slider. Volume discounts apply automatically at 500, 2,000 and 5,000 endpoints.

250 endpoints
Essentials
£450
per month
£1,350 less per year than rolling
Professional
£725
per month
£2,100 less per year than rolling
Enterprise
£1,050
per month
£3,150 less per year than rolling

Figures exclude VAT. Above 10,000 endpoints the bands keep going — talk to us and we will put it in writing rather than in a meeting.

Artificial intelligence

Two ways to run it, and the difference is where the prompt goes.

Not a feature toggle. For an MSP with a client under legal or clinical confidentiality, this is the difference between using AI and switching it off.

Denwen RMM
Reporting across the estateReporting across the estate

Denwen AI

Set up and managed by the global administrator

We host the model and meter the usage. The platform operator enables it, decides which tenants may use it, and sets each one an allowance. Nothing to procure and no keys to hold.

  • Enabled once, for the instance, by whoever administers the platform
  • A monthly allowance per tenant, so one client cannot spend another’s
  • Consumption visible per tenant and per action, not as one number at the end of the month
  • Included allowances on Professional and Enterprise; top-ups when you need them
  • Prompts are processed by a model on our own hardware, never sent to a third-party AI company, under our data-processing terms

Read the Data Processing Agreement

Included allowance, then top-up packs

Your own model

Configured by each tenant administrator

A tenant points the platform at their own provider and pays them directly. Denwen charges nothing for the integration, sees no prompts, and stores no completions.

  • Any OpenAI-compatible endpoint: OpenAI, Anthropic, Azure OpenAI, Google, Mistral
  • Or a model on your own hardware — Ollama, vLLM, anything that speaks the same protocol
  • The tenant holds the key; it is sealed at rest and never returned to any console
  • Prompts go from the platform to your provider. They are not processed by ours
  • Available on every tier, including Essentials — this is not an upsell

No platform charge — you pay your provider

What one AI action buys

The meter counts actions, not tokens, so an estimate is possible before the month starts. Each of these is one action.

Explain an alert storm

Forty alerts across nine devices in six minutes. One probable cause, ranked by evidence, with the telemetry that supports it.

Review a script before it runs

What it changes, what it cannot undo, and which of the 400 endpoints it is about to run on would be affected differently.

Read a patch in plain language

What the KB actually fixes, whether it needs a reboot, and what it has broken elsewhere according to the update history on your own estate.

Find devices by describing them

“Windows 11 laptops at Acme with a pending reboot and no check-in since Friday” — as a sentence, not a filter builder.

Write the session notes

What was done on the endpoint, in the order it happened, ready to paste into a ticket without anybody reconstructing it from memory.

Explain an audit entry

Who did this, what it touched, what changed, and whether anything else happened in the same minute that ought to be read alongside it.

Governance, on Enterprise

  • Choose which tenants may use AI at all
  • Choose the model each tenant is allowed to reach
  • Decide whether device data may be included in a prompt, or only metadata
  • Redact hostnames and usernames before anything is sent
  • Every AI action written to the audit trail: who asked, what was sent, which model answered
Questions

The ones that decide it.

Is there really a free trial?

Seven or fourteen days, on your own endpoints, up to 25 of them, and no card. A trial on a vendor sandbox tells you nothing about your estate — the whole point is to see your own machines reporting.

What happens when the trial ends?

If you have not chosen a plan by then, the organisation is suspended: nobody can sign in and no new endpoint can be enrolled. Nothing is deleted — the agents you installed keep reporting, and everything comes back the moment you take a plan. We say all of this on the signup form before you start, and email you before the date, on the date, and with a link that works without signing in.

What counts as an endpoint?

A machine with the agent installed and enrolled. Uninstalled devices stop counting on the day they are uninstalled, not at the end of the term.

Can I run it on my own servers?

No. Denwen RMM is a hosted service only — self-hosting was withdrawn and is not coming back. If your obligation is that prompts must not reach a third-party model, point AI at your own endpoint on any tier; if it is that no telemetry may leave your network at all, we are not a fit and would rather say so here than in month three.

What happens after I pay?

Your organisation, your administrator account and your signed agent installer are created automatically while the receipt is still on screen. You set a password from the email and start enrolling endpoints — there is no provisioning queue.

What happens if I stop paying?

The console goes read-only rather than dark, so you can export what you need. Agents keep reporting for 30 days before they stop, so nothing is stranded mid-incident.

Do you charge for extra operators?

No. Price is per endpoint; add as many technicians, roles and tenants as you need.

Where is our data held?

In the United Kingdom. Recordings, telemetry and the database sit in one region and Enterprise names it in the contract. Since self-hosting was withdrawn this is the answer to the question it used to answer, so it is stated rather than implied.

Is support extra?

Support is included on every tier. Enterprise adds a response SLA and named onboarding.

Start on ten machines

Install the agent on a handful of your own endpoints and judge it on what comes back. No card at all for the trial, and nothing to install on a server at either end.