Access Restored and Major SLA Penalty Avoided

Company background

An American technology company and information technology services provider that produces command and control systems and products, tactical radios, avionics and electronic systems, for use in government and commercial sectors. They purchase various different software licenses from vendors which are supported by the vendors technical specialists.

Issue:

Users were not able to log into their requirements management system, preventing them from entering and managing requirements. This is the first critical step in setting up a project for the organisation, ensuring each project can deliver what the stakeholders require.

Solution:

The vendors support team used KT’s Situation Appraisal to understand the client’s concerns and the impact of those concerns. “Users unable to log” in was confirmed to be the most important concern and it was determined that a Problem Analysis was needed to help find the cause.

Initially, the problem statement was “50 users could not log in to the requirements management system”. Through the first partial iteration of problem analysis (during IS/IS NOT data collection), it became clear that one of the application authentication servers was not functioning correctly. The server was restarted which resulted in 40 users being able to log in.

10 users were still unable to log in. The team redefined the problem statement using 5 Whys and found that admin users no longer had access to the recently restarted server. The problem statement was revised to “10 admin users unable to access application authentication server”. The data collection was recommenced for this new problem statement and the team focused in on the third “WHEN” question in problem specification (ie. “When in the objects history or lifecycle was the deviation observed first?”). This question in the IS/IS NOT data collection led the team to map the internal process of the software and identify a possible cause of the problem. While doing this the team asked what was different about the lifecycle steps followed by the users having the problem (compared to users not experiencing the problem) and from there were able to formulate a possible cause that was specific to the problem.

The client had separately come up with their own possible cause. Both causes were tested against the data collected and the team were able to eliminate the client’s possible cause as it did not stack up against the factual information. The vendors teams possible cause aligned with all of the data. The solution was put in place and root cause was confirmed. Further, the clients’ technical notes were updated to ensure the problem did not recur.

Impact:

This issue needed to be resolved by the following morning as there was a government report that needed to be delivered that day. Failure to deliver the report would have attracted US$100,000 in penalties.