← Back to Blog INTEGRATIONS

How to wire MSP Process into ConnectWise without breaking your dispatcher's workflow

A pragmatic deployment guide for MSPs running ConnectWise PSA — what to enable on day one, what to roll out in week two, and the change-management pitfalls.

The number-one risk in any service-desk platform rollout is breaking the dispatcher's workflow on day one. We've onboarded 1000+ MSPs. The successful rollouts share a pattern: phased enablement, dispatcher buy-in early, no big-bang cutover. Here's the playbook for ConnectWise PSA specifically.

Day 0 — Pre-rollout

Before you turn anything on, do these four things:

  • Map the dispatcher's current workflow. What inbox do they sort first? What's their first action on a new ticket? Where do they go to look up contract SLAs? You're not going to redesign this. You're going to slot in.
  • Pick one client tenant for the first 30 days. A real one — not a sandbox. Ideally one where you have a strong relationship and can call them with questions.
  • Get the API key. Service desk admin in ConnectWise creates an API member with the appropriate scope. We provide the exact scope list.
  • Tell your dispatcher. Not "we're rolling out an AI thing." Tell them: "Starting Monday, the AI will pre-classify inbound tickets. You'll see its classification at the top of every new ticket. Override it if it's wrong. We're tracking accuracy together."

Week 1 — Just intake and triage

Don't turn on AI Voice yet. Don't turn on identity verification on every channel yet. Start with one thing: AI Triage on inbound tickets.

What this looks like to the dispatcher: they open ConnectWise, see new tickets in their normal queue. Each ticket has a small panel at the top showing AI's classification — category, priority, contract SLA, suggested tech, dedup against existing tickets. The dispatcher confirms or overrides. The override gets fed back to the model.

You'll get 90%+ accuracy on day one. By the end of week one you'll be at 95%+. The dispatcher will tell you which categories the model gets wrong (there will be 1-2). We tune those.

"The successful rollouts don't try to change everything on day one. They prove value on one workflow first. The dispatcher becomes the advocate, not the obstacle."

Week 2 — Add SMS and Teams as verified intake

Now turn on SMS and Microsoft Teams as ingestion channels. Both create ConnectWise tickets directly. Both verify the requester through Entra ID before creation. The dispatcher sees these in the same queue with a "verified" tag.

This is where you start to see capacity recovered — clients stop texting techs' personal cell phones, Teams DMs become tickets, the shadow ticket queue collapses.

Week 3 — Identity verification on privileged actions

Password reset, MFA reset, account unlock — turn on the verification gate. Dispatcher gets a one-click button on the ticket: "Send verification challenge." Authenticator push goes to the registered device. Once approved, the privileged action runs and writes an audit entry to the ticket.

This is the highest-value control you'll ship. It's also where dispatchers occasionally push back — "but I've known this client for 5 years." The honest answer: voice cloning is consumer-grade now. Trust is a vulnerability. The verification step takes 4 seconds and saves you the breach.

Week 4 — AI Voice on after-hours

Now bring AI Voice live, but only for after-hours initially. The reason is psychological, not technical. After-hours is the mode where the on-call tech is most relieved to have help. Daytime voice can come later.

What this looks like: after-hours calls go to your dedicated MSP Process number. AI Voice answers, verifies, opens the ConnectWise ticket, triages, and either resolves with a known-issue response or pages on-call with full context. Calls that previously woke the on-call tech — but were actually P3s — get queued silently.

Months 2-3 — Roll out to the rest of your tenants

By now your dispatcher is the advocate. Roll the platform out to the other client tenants on a tenant-per-week cadence. Each rollout takes about 2 hours. The dispatcher's workflow doesn't change between tenants — the same queue, the same panel, the same verification gate.

Pitfalls we've seen

  • Big-bang cutover. Don't. Phased rollouts succeed. Big bangs fail.
  • Skipping the dispatcher conversation. If your dispatcher doesn't know what's coming, they will resist. Spend 30 minutes with them in advance.
  • Turning on AI Voice during business hours first. Daytime voice is high-stakes. Start with after-hours. Build the pattern.
  • Not connecting it to your contract SLA fields in ConnectWise. The AI's accuracy depends on contract data. Spend 20 minutes mapping it.
  • Treating it as a tool rollout instead of a workflow change. You're not just installing software. You're changing how tickets enter the desk. Communicate it that way.

The 30-day test

30 days in, look at four numbers: AI Triage accuracy (target ≥95%), tickets caught from shadow channels (Teams + SMS), verification challenges run (target every privileged action), after-hours pages to humans (target -75% vs baseline). If all four are trending the right way, you're ready to scale.

Most MSPs we onboard hit those four numbers in 30 days. The ones that don't usually skipped the dispatcher conversation. Take the 30 minutes.

Ship a verified service desk in 30 days.

Book a 30-minute call with a solutions engineer who came out of an MSP service desk. Bring your stack. We'll model the impact with your numbers.