After my last blog discussing how analytics can help manage risks while driving successful changes, I received a few emails from colleagues curious about how analytics can drive change velocity and help IT teams move to the DevOps model. Given that DevOps is a hot topic and there is an ongoing debate around how it replaces or complements traditional ITSM, the subject seemed like a natural topic for a second blog.
I also recently had a great discussion with one of our customers on this topic. She is a Director of Enterprise Services and has spent more than two decades in the IT industry, including having chaired the CAB. She made two key points on transitioning to a DevOps model:
1. Modern IT organizations have to operate in multi-speed models, largely driven by the company’s business model. The more agile environment is good for organizations that have shifted from monolithic applications to microservices and automated all release steps. However, for many legacy and enterprise-critical applications that run backend operations, change management is an absolutely critical part of the organizational innovation pipeline. Therefore, change management and agile are complementary and necessary methodologies for most IT organizations.
2. CABs are often blamed for slowing down change velocity. However, in her experience the delays are often either on the business side (such as awaiting risk assessments) or the technical side (perhaps the right developer resources are not available to apply and support the change).
Understanding Change Lead Time
The Director of Enterprise Services urged me to talk about how IT organizations can apply analytics to improve change velocity by showing how the various steps of the process cause delays. Every pause in the change approval process postpones the release and leaves the change trapped in a loop, not unlike an airplane stuck taxiing on an overcrowded runway while awaiting its turn for takeoff.
One such analysis is shown below, which breaks down the change process into individual steps and highlights where the delays happen. The top graph shows trends in change lead time by CAB area (selected via a drop-down box in the top right corner). The dashboard shows that at the CAB level, users can analyze trends by application area, requestor department, or any other relevant dimension. They can then drill into any given month to see how many changes went through and which step took the longest. The circle sizes indicate total time, while the color represents the average elapsed duration for that step. Both metrics are important, as approval for a handful of changes may have taken a long time, but on average changes may have been approved at a rapid clip.
This is just one example of how analyzing the change approval process can help identify the true cause of delays. However, there are many other ways in which change managers can analyze the overall process to drive improvements and increase compliance while increasing change velocity.
Monitoring Change Process Effectiveness
Shown below is another customer use case that analyzes change process effectiveness to drive continual improvement. The data in the screenshot is demo data, but it illustrates how insightful visual analyses can help identify process inefficiencies.
For example, we can quickly see on a monthly basis which assignment groups have complied with the process while managing successful changes. Each bubble represents an assignment group, the size of the bubble shows the number of changes closed, the x-axis indicates the success rate (of RFCs raised), and the y-axis denotes the process adherence rate. Groups in the bottom left corner are the ones that need to improve their process, as they are neither compliant nor successful in pushing through changes. You want to see large bubbles towards the top right of this graph.
An animated version of the graph can take the user through each month and show patterns over time. Thus, groups that are consistently in the wrong corner can be offered evidence on why their changes need to spend more time in risk reviewal.
What sort of challenges do you face in increasing the rate and number of changes being pushed through the system? Do you have specific mechanisms or initiatives to reduce change backlog and accelerate the change approval process?
Find out more about how analytics can help your team drive change velocity in our exclusive change management webinar.
[Photo courtesy of flickr user Bryce Bradford.]