Skip to tutorial content
Rate module Not yet rated

Scoping Problem Management: Reactive and Proactive

4 min read
On this page

In this lesson, you will learn:

  • What reactive problem management is
  • What proactive problem management is
  • SNWR’s rules for opening problems
  • How to spot trends in incident data

🔁 Quick look back: Last time: Problem Architecture and Lifecycle in Practice. You learned the problem states from New to Closed, and who owns each step.

1 “We can’t open a problem for everything”

Grace is keen. In week one, she opens 60 problems.

  • Half are one-off incidents that will never happen again.
  • Nobody has time to work them.
  • The real VPN problem gets lost in the pile.

🎯 The problem: Too many problems means nothing gets fixed. Too few means causes are never found.

👉 The fix: agree clear rules for when to open a problem.

2 Reactive and proactive

Key term: Reactive problem management

Opening a problem after something has gone wrong, like a major incident.

Key term: Proactive problem management

Finding problems before they hurt more, by looking for trends in incident data.

Where SNWR's problems come from. Reactive: after every major incident, for example PRB0040020 for the email outage. Proactive: from trends, for example 12 VPN incidents in one month led to PRB0040012. Also from agents who notice a repeat, and from known issues raised by vendors.
Figure 1. Some problems react to pain. Others spot it early.

3 SNWR’s rules

Open a problem when… Type
A major incident is resolved Reactive (always)
The same issue causes 5 or more incidents in a month Proactive
An agent sees the same fix used again and again Proactive
A vendor warns about a known fault Proactive

📝 Note: Grace signed these off as decision DD-019.

  • A monthly report of incidents by category, subcategory and CI.
  • A list of incidents with the same resolution notes.
  • The top 10 CIs by number of incidents.

💡 Good to know: Good categories and CIs make trends easy to see. That’s why you cleaned them up earlier.

⭐ Remember: React to every major incident. Look for trends every month.

5 Quick recap

  • Reactive: open a problem after something goes wrong
  • Proactive: find problems early from trends
  • SNWR opens a problem after every major incident
  • And when one issue causes 5 or more incidents in a month
  • Clean categories and CIs make trends visible

6 Check yourself

Think of your answer first. Then tap to check.

1 A major incident is resolved. Does SNWR open a problem? Show answer

Yes, always. That’s reactive.

2 12 VPN incidents in a month. Which kind of problem is PRB0040012? Show answer

Proactive.

3 Why not open a problem for every incident? Show answer

Too many problems means none get worked on.

4 What helps Grace spot trends? Show answer

Reports by category, subcategory and CI, and repeated resolution notes.

Nice work! You know when SNWR opens a problem. 🎉
Next up: Creating Problems from Incidents and Linking Them

Was this lesson helpful?

Have a question or doubt?

Ask here and get help from instructors and fellow learners.

0 / 300

Only logged-in learners can post.