An employee is locked out of a business application. The service desk restores access and closes the incident. A few days later, another employee reports the same failure.
The recurring issue offers a useful way to evaluate IT service management software. Can a team connect separate incidents, investigate the pattern, share a workaround, and keep track of a lasting fix?
IT service management, or ITSM, is the practice of managing IT services. ITIL provides guidance organizations can use to shape those practices. Incident management focuses on restoring service; problem management investigates underlying causes and seeks to reduce the risk and impact of future incidents.
Start with a scenario
Give each vendor the same hypothetical situation: several employees experience intermittent sign-in failures. Restarting a component temporarily restores access, but the failures return. The team needs to investigate while continuing to support affected users.
When evaluating an IT service management platform such as Matrix42, ask the vendor to demonstrate that workflow rather than relying on a feature list.
Agree on the expected results before the demonstration. At a minimum, the team should be able to find related incidents, identify the investigation owner, retrieve the current workaround, and see what happened to the proposed fix.
Connect incidents to an investigation
Ask the vendor to create a problem record and link three sign-in incidents to it. Each incident should retain its own details, including when it occurred and whom it affected. The problem record should give the investigator a view of the pattern.
Then add a fourth incident after the investigation has started. Can a service desk agent find the existing problem and link the new report without starting a duplicate investigation? Can the investigator move from the problem record back to the individual incidents?
Test ownership as well. Assign an investigator, record an initial hypothesis, and hand a task to another team. Look for a clear record of who remains responsible, what work is outstanding, and where new findings belong.
The record should distinguish a suspected cause from a confirmed finding. That distinction helps prevent an early guess from becoming the accepted explanation as work passes between teams.
Make known errors and workarounds usable
Suppose the investigation identifies a condition associated with the sign-in failures, but a permanent correction will take time. The team documents a known error and a temporary workaround: restart the affected component after appropriate checks.
Ask the vendor to show how an agent handling a new incident would find that information. Can the agent see who maintains the workaround, when it was reviewed, and any restrictions on using it? A restart might interrupt other users, so the instructions need enough context for the agent to decide whether it is appropriate.
The workaround can restore access while the underlying problem remains. Check whether the platform can keep the investigation open, record changes to the workaround, and make current guidance available to the people who need it.
Access controls matter too. Investigation notes can contain technical details that should be restricted rather than displayed to every employee who submits a ticket.
Follow the fix through review
Continue the demonstration with a proposed correction to the application component. Ask the vendor to link the problem to the relevant change or implementation record and show how the investigator can check its progress.
Once the work is complete, the problem record should accommodate a review: who checked the outcome, what evidence they considered, and whether sign-in failures returned during the agreed observation period. Implementation and resolution are separate events.
If the issue recurs, can the team reopen the investigation or link the new incident to the existing problem? If the fix is deferred, can the record capture that decision and the workaround that remains in use? The answers show how well the platform preserves the history behind a status.
Check deployment and data handling
Problem records can contain employee details, screenshots, system logs, and descriptions of internal infrastructure. For a European organization, assess the proposed deployment alongside the workflow.
Ask where live records and backups will be stored, which organizations provide the hosting, and where vendor support staff can access the data. Check the arrangement against the organization’s policies and applicable obligations. A European data center address alone leaves questions about access and data processing unanswered.
Confirm that the required deployment model is available for the product and contract being considered. Ask whether problem-management features require additional configuration, modules, or integrations.
After each demonstration, record what was shown, what depended on customization, and which questions remain unanswered. Compare those observations against the same requirements, giving weight to both the investigation workflow and the way its records will be handled.