How to solve your problems with problem management

Now, let’s look at how to solve the problems with problem management.

Can AI help?

All the rage today is about how AI capabilities can improve many aspects of  ITSM performance. Can AI specifically improve an organization’s problem management capabilities? Of course!

Here are some ways that AI can improve problem management effectiveness.

• Better and deeper analysis: Most of today’s decisions with problem management are based on a limited amount of data. The use of AI capabilities such as AIOps can be used to expand the data sources that are considered (e.g., event logs, monitoring data and more) during problem control activities.
• Using unstructured data: Using GenAI along with machine learning can unlock potentially valuable insights captured within unstructured data sources, such as email, application logs and sensor data.
• Using answer engines: Answer engines are tools designed to provide direct and concise answers to user queries, often by synthesizing information from various sources, rather than just a list of web pages that may or may not be helpful. Examples of answer engines include Perplexity , ChatGPT and Google Gemini.

But the introduction of AI is not a “magic wand” for addressing underperforming or nonexistent problem management practices. To effectively leverage AI capabilities for problem management, here’s what organizations need to do first:

• Practice good ITSM hygiene: This means implementing and practicing good habits within the ITSM environment. Good notes on incident records, defined change and request models, clearly defined problem criteria, and having relevant and accurate knowledge articles all contribute to good ITSM hygiene.
• Understand how technology underpins business results: This helps organizations understand the impact and associated risk of technology use on business outcomes. Journey mapping, value stream mapping and defining service catalogs are all good techniques for identifying and understanding how technology and business interact.
• Formally practice continual improvement: Just as with any product, service or technology, implementation of AI capabilities cannot be “one and done” exercises. Without formal continual improvement practices already in place, the advantages that AI adoption may produce will become obsolete. The investments in AI (just like investments in problem management) will soon become a waste of time, money and effort.

3 steps for elevating your problem management practice

Here’s three suggestions to begin elevating your problem management practices:

• Align problem management with MVG: Identify how problem management practices support the goals and objectives identified with the organization’s mission, vision and goals (MVG) statement. This will help identify the measures that are important to the organization, as well as the reports that management needs to justify support of and investment in problem management.
• Define the criteria of a “problem”: What makes a problem a “problem”? In my experience, it must be more than just several recurring incidents within some timeframe or a response to a major incident. A business impact analysis (BIA) report is a great source of information regarding the impact of service loss. This same information can be used to further develop problem criteria.
• Analyze incident data: Use incident data to improve the effectiveness of and support for problem management by analyzing the financial impact and risk exposure caused by incidents. How much money did the IT organization spend to resolve an incident? What was the financial impact of incidents on the organization, considering issues like lost productivity, lost sales, or customer satisfaction? How did incidents make the organization more vulnerable to noncompliance with regulations, loss of reputation, or exposure of sensitive data? Looking at incidents through the financial and risk lens is often eye-opening for senior management and further emphasizes the need for effective problem management practices.