Blog / Insights

Leaving ConnectWise Automate: migrating agents, scripts and monitors without losing an endpoint

Insights By the Helios team · 19 October 2026 · 7 min read

A ConnectWise Automate migration fails on dates more often than on code. Plan it backwards from your renewal: serve cancellation notice inside the contract window, run the new agent alongside Automate for a fortnight or so, rebuild only the scripts and monitors you still need, then retire the Automate server once every endpoint has checked in elsewhere. The technical work typically takes weeks. Missing the notice date can cost you a full extra year.

That order matters, so here is the plan, step by step, whether the machines belong to clients or to your own company.

Start with the contract, not the console

The received wisdom is to build the new platform first and sort the paperwork once you are confident. That gets the risk exactly backwards. The paperwork has a hard deadline. The build does not.

According to Flamingo, ConnectWise agreements commonly auto-renew with a 60 to 90 day written cancellation window before the renewal date, and missing it locks the MSP in for another year. The format of the notice matters too. The ConnectWise 3rd Party Product Additional Terms and Conditions require a minimum of 60 days' notice for downgrades and cancellations (30 days may be permitted depending on the offering), and that notice must be an electronically signed agreement documenting the change. An email to your account manager is not the same thing.

The 120-day test: find your renewal date today, subtract 120 days, and put that in the diary as the decision date. That leaves a month of slack before the earliest plausible notice window opens.

Read your own agreement rather than relying on anyone's summary, including this one. Terms differ by bundle, by reseller and by when you signed. Check separately for Automate, your PSA and ScreenConnect, because they may renew on different dates.

Be honest about the hours, in both directions

The second thing that burns MSPs is underestimating effort. Automate was expensive to learn, and that sunk cost makes people assume leaving will be just as expensive. CheckThat.ai reports Automate proficiency training at 80 to 120 hours for an experienced professional and 240 to 480 hours for new or junior staff, with total mid-market implementation timelines of 6 to 18 months.

You are not repeating that. Most of those hours went into learning Automate's internals: its database, its monitor types, its script engine. A modern RMM asks for less of that. What does take time is deciding what your Automate estate was actually doing, because years of accretion leave scripts nobody remembers writing and monitors nobody reads. Budget the hours for that audit, not for a full relearning.

Audit and export before you touch an agent

Pull everything out while you still have full access: scripts, monitors, groups, searches, patch policies, and any credentials or notes stored against clients. Then sort each item into one of three piles:

The retire pile is usually larger than people expect. That is good news.

Plan the agent rollout as an overlap

Do not rip Automate out and push the new agent on the same night. Run both for a defined period, which we cover in detail in our guide to a safe two-week overlap with two RMM agents. The overlap gives you a reconciliation step: every device in Automate should appear in the new platform before anything is removed.

  1. Deploy through Automate itself. Use Automate's own scripting to push the new agent installer. It is the one time the old tool works hardest for the new one.
  2. Catch the stragglers. Offline laptops, kiosks and machines with broken Automate agents will not take the push. Use Group Policy, Intune or a login script for those.
  3. Reconcile by name and serial. Export both device lists and compare. Every mismatch is either a stale Automate record or a device you are about to lose.
  4. Remove Automate cleanly. ProValTech recommends uninstalling the agent on the endpoint, then retiring the agent entry in the Automate console. Do the same at exit so records do not linger and confuse the reconciliation.
An endpoint is not migrated until it reports into the new platform. Being uninstalled from the old one does not count.

Rebuild scripts and monitors by outcome, not line by line

Automate scripts are written in Automate's own step-based format, so very little ports directly. Treat that as a feature. Rewrite each one as PowerShell that states its outcome: clear the print spooler, confirm the backup agent is running, remove a specific unwanted application. PowerShell travels between platforms, and Automate scripts do not.

Monitors deserve the same treatment. Automate estates accumulate internal monitors the way lofts accumulate boxes, and many generate tickets that are closed without action. For each one you keep, write down three things: what condition triggers it, who acts on it, and what they do. If you cannot fill in the second and third, it is noise.

Retire the on-premises server last

If you host Automate yourself, the server is the final thing to go. Keep it running, though not necessarily licensed for active use, until reconciliation is complete and the cancellation has taken effect.

Failure modes, side by side

MistakeWhat it costs
Notice served late or by plain emailAnother year of Automate fees alongside the new platform
Agents swapped without overlapOffline devices silently drop out of management
Every script rebuiltWeeks spent porting work nobody needed
Server left onlineAn exposed, unmaintained admin system

For the wider choice of where to go, our ConnectWise alternatives guide compares the options.

Where this fits with Helios

Most of this plan is discipline, not tooling: the contract date, the audit and the reconciliation are yours to run whatever you move to. On the receiving side, Helios combines monitoring, Windows and third-party patching, scripting, remote access and a helpdesk with SLA clocks in one platform, so much of the replace pile is covered natively. Its built-in AI technician, Helio, investigates alerts and runs diagnostics within plain-English rules you set, which helps with the monitors you keep. Every feature is on every plan, and there are 14 days free with no card, which is enough to run the overlap before you commit.

Common questions about ConnectWise Automate migration

Can I import ConnectWise Automate scripts straight into another RMM?

ConnectWise Automate scripts use Automate's own step-based format, so other RMM platforms generally cannot import them directly. The practical route is to export the scripts for reference, then rewrite the ones worth keeping as PowerShell, which most modern RMM tools run natively. Rewriting also gives you a chance to drop the logic that was only working around Automate's limits.

Do I lose ScreenConnect when I leave ConnectWise Automate?

ScreenConnect is often licensed separately from ConnectWise Automate, so leaving Automate does not automatically end ScreenConnect, and it may have its own renewal date and notice period. Check the ScreenConnect agreement on its own terms. If the new RMM includes remote access, you can cancel ScreenConnect in the same window; if not, you may choose to keep it.

How long should a ConnectWise Automate migration take?

A ConnectWise Automate migration for a small or mid-sized MSP usually takes a few weeks of active technical work: an audit, a two-week agent overlap, script and monitor rebuilds, and server retirement. The contract is the longer constraint, because cancellation notice may be due 60 to 90 days before renewal, so start planning around four months ahead.

Related: Leaving ConnectWise Manage: moving tickets, agreements and billing to a new PSA, Leaving N-central: migrating devices, Automation Manager policies and service templates without losing an endpoint

Why MSPs choose Helios

One platform that does the work, at a price that stays put

Helio fixes, not just flags

When an alert fires, Helio, the AI technician built into Helios, investigates it, writes the fix and runs it once you approve. It keeps what works, so the next one is quicker.

Everything in one product

RMM, a service desk with SLAs, patching including third-party apps, remote access, Microsoft 365 and Defender checks, backup monitoring, a client portal and billing into Xero or QuickBooks.

Priced by fleet, not by people

£4 a device on Launch up to 25 devices, from £20 a month, then £99, £199 or £399 a month by fleet size, with any number of technicians and every feature on every plan.

A price that stays put

Your price is locked for as long as you stay subscribed, and that is written into our terms. Monthly billing, cancel any time.

Bring your clients across

Import your clients and machines from another RMM's CSV export, then roll the Helios agent out at your own pace, with a count of how many have arrived.

14 days free, no card to start

The full platform from the first minute, on your own machines. No sales call. Remote control asks for one card check, and nothing is charged.

See it on your own fleet

Helios is the AI-native platform for MSPs: monitoring, patching, security, Microsoft 365, backup monitoring, remote access, client billing and an AI service desk in one product, at one flat price per MSP. Contracts and logged time become invoices in Xero or QuickBooks without leaving the platform. Every feature is on every plan. 14-day free trial, card-free, set up in minutes, cancel any time.

Start free

Researched and written with Helio SEO, our AI writer for business blogs.