Migrating to Atomicwork from an ITSM/ESM platform

This document compiles best practices and lessons for organizations using Atomicwork to run AI-powered IT and enterprise service management. Rather than theory, this is a practical field guide: what works, what trips teams up, and how to set Atom up to deliver real results from day one.

The transition process is carefully structured to address all aspects of the migration, from initial planning to final implementation, with a focus on maintaining business continuity and minimizing service disruptions.

The goal is to seamlessly transition all service management workflows from your current service desk to Atomicwork, achieving full adoption across all relevant teams. In our experience, it usually takes 60-90 days for a team to go live with Atomicwork.

Implementing Atomicwork is not just a software deployment. It's a change management project. The technology part is honestly the easier half.

At Atomicwork, we recommend leveraging agile methodology to manage your service desk migration. This way, you can effectively manage changes, track progress in real-time, and ensure that each iteration aligns with your strategic objectives. This approach allows for a more rapid response to evolving needs, minimizes risks, and ensures that quality and compliance are consistently upheld throughout the development process.

In our experience, it takes 60–90 days from kickoff to go-live for a first workspace. That number is heavily influenced by one thing: how quickly your team can make decisions about process, not how fast we can configure the product.

A phased approach reduces risk and builds internal confidence before full deployment.

Phase 1 — Pilot: IT only. Stand up KB, connect core integrations, tune Atom responses.

Phase 2 — Expand: Add HR and Finance. Build department-specific knowledge workspaces.

Phase 3 — Launch: Full employee rollout with Slack/Teams integration and self-service portal live.

Phase 4 — Optimize: Review KPIs monthly. Retire deflected ticket categories. Identify automation candidates.

Identify one executive who will own the 'why' for this project and communicate it to the rest of the organization. They don't need to attend every meeting — but they need to be able to articulate the benefit clearly enough that others repeat it.

Atom is not a one-time deployment. Assigning clear roles from the start ensures the platform stays healthy and continuously improving.

Before any configuration starts, you need to decide which workspaces you're building first and what's in and out of scope. Everything downstream — timeline, testing, go-live dates — depends on this decision.

This is the most valuable thing you can do before your first onboarding call. Pull 3–6 months of ticket data from your current service desk. This data drives everything: your knowledge base priorities, your test questions, your deflection targets, and your golden evaluation set.

What makes teams go faster

What slows teams down

For Atom to deliver precise answers, your knowledge must be structured so it’s clear, self-contained, and easy to interpret. This improves accuracy, reduces routine requests, boosts productivity, and frees service teams for higher-value work.

We've put together a guide to explain how Atom processes your content, how to organise and format it so Atom can find what’s relevant, understand the context, and compile complete responses, and how to keep it accurate and useful over time.

But here's some 50,000 ft. insights from other Atomicwork customers:

Run a structured pass before rolling Atom out using our test suite. Use real employee language, not IT terminology.

Atom compounds in value. Organizations that treat KB management as an ongoing practice see steadily improving deflection rates month-over-month.

Develop marketing material and messaging for the Agent rollout plan for employees. 

Setup a live training session for all of the agents in the launch workspaces. Identify process managers and front-line users who will be able to lead peer-to-peer training sessions and answer questions. The training will focus on how the new processes work in Atomicwork and how to work effectively with them. At the end of the training sessions, agents can take the Atomicwork certification quiz and feel confident of their knowledge.

Enable the Agent for a beta group of testers who were not a part of the first two testing phases. These would ideally be champions who can promote quick wins and influence adoption amongst their peers. Setup a review session with the Atomicwork team one week after use

Inform employees of the new software and processes in place through the collaboration tool and email.

Assign a dedicated support team, along with Atomicwork personnel, to handle any issues that arise during the go-live period and be in hypercare mode.

Deploying Atom in the flow of work through the browser, Slack or Teams is the highest-impact adoption move. Employees get answers where they already work; no portal to navigate, no workflow disruption.

Establish a baseline before go-live. Demonstrate ROI by comparing against it monthly.

When the committee decides on a cutover date, the date should be validated by the committee based on:

Implement a system freeze on the existing ITSM to prevent changes the week of Atomicwork’s first workspace rollout. At this point, the old service desk will become read-only. Schedule the cutover during a low-activity period (the weekend) to minimize disruption.

Communicate the freeze to all users, indicating the time it will begin.
All tickets created in the old service desk before this time period will be handled in the existing ITSM tool but no new tickets.

Ready to go deeper?

Reach out to your Atomicwork customer success team. We’re here to help you get the most out of your AI-powered service management investment — from day one through long-term optimization.