5 Steps to Start Improving Your IT Organization, for the Non-Unicorns Amongst Us

Some highly visible IT organizations have adopted ideas from DevOps, Agile, Lean and other frameworks to help them deliver great customer experience, with lower cost and higher quality. But what about the rest of us? How can we start to improve?

We Can’t All Be Unicorns

When I go to IT service management (ITSM) conferences, or read ITSM blogs, it’s easy to form an impression that every organization has completely changed how they manage IT. I hear about unicorn organizations that have adopted DevOps, Agile, Lean and other frameworks to enable them to deliver much more value, faster, with lower overheads, and greater customer satisfaction.

This newer style of an IT organization typically has enormous competitive advantages over their legacy competitors. Their IT services are flexible, and able to respond very rapidly to changing business needs. They don’t talk about IT-business alignment because it’s a natural part of how they work. They have lower costs with higher quality and they have more satisfied customers as well as more satisfied IT staff.

Then I go out to visit real IT organizations that have hired me to help them, and more often than not I find a totally different story.

Here’s a table summarising the major differences between the old ways of working that I see in many of the real IT organizations I’ve worked with, and the new ways of working that everyone talks about.

OLD	NEW
Command and control culture with well-defined processes that should be followed	Collaborative culture with well-understood outcomes that should be achieved
Focus on a single ITSM framework, usually ITIL	Ideas taken from a wide variety of frameworks, typically DevOps, Agile, Lean, Theory of Constraints (AND possibly ITIL)
Many internal metrics and reports showing SLA achievements	Mainly business and customer facing metrics, far less focus on SLAs
Systems designed to resist failure, with a focus on reducing how often they fail (MTBF)	Systems designed to recover fast after failure, with a focus on how quickly they recover (MTTR)
Customer experience measured in annual surveys and feedback when incidents are closed	Customer experience actively planned, managed, and measured as part of every activity
Some automation used, when this is clearly cost-effective	Automation used wherever possible, to ensure reliable and repeatable service delivery
Low rate of change, typically batched into weekly, monthly, or quarterly releases	High rate of change, typically deployed as small single changes many times per day
Change advisory board (CAB) manages changes by reviewing requests for change (RFCs)	Change management largely delegated to the people responsible for the change
Configuration management based on collecting information from the IT components to support other processes	Configuration management based on deploying infrastructure and code from a managed repository

I could probably add another 20 rows to this table, but this should be sufficient to illustrate my point. There is an enormous difference in attitudes, behaviour, and culture between the organizations I hear and read about, and the ones that exist elsewhere, in real life.

Why Doesn’t Every Organization Adopt these New Ways of Working?

My customers are not stupid. Most of them recognize the potential value in new ways of working. But it can be very difficult to see how to move from where they are now to where they might like to be – when they can also see many risks, especially if they are in a regulated environment or one with high security needs.

There is a common myth that one big benefit of the old way of working, with its rigidly applied formal controls, is better security; and that moving to newer ways of working involves more risks. In practice, the opposite seems to be true. The greater levels of automation and peer review in newer ways of working are very effective at reducing risk, and at providing the audit trails that are so important in a secure environment.

Many organizations try to adopt ideas from DevOps, but because they work within a culture that emphasises technical issues, the changes they introduce are limited to automation of the tools used for testing and deployment. While this is an important part of the transition to new ways of working, it is only one part, and if that is all an organization decides to change, then it won’t see any significant improvements in customer experience or service quality.

5 Steps You Can Take to Start Adopting New Ways of Working

If you want to adopt any of these new ideas but haven’t yet started, here are some steps you can take. Each step is based on making small changes to take you nearer to your goal, rather than on major transformation. But if you do decide to try them, then they will start to transform your culture, and this will make a great start…

1. Review Your Metrics and Reporting

Eli Goldratt, the man who developed the Theory of Constraints, once wrote “Tell me how you measure me, and I will tell you how I will behave.” If you measure and report lots of internal metrics about service desk responses, numbers of incidents closed, and so on then you can guarantee that your staff will focus on what is needed to improve these numbers. On the other hand, if you measure and report the things that are important to your customers, then you will encourage staff to focus on the things that make customers happy.

When did you last sit down with your customers to ask them what is important to them, and how you can measure this? It’s a great thing to do, and you might even learn something you didn’t know about what your customers value.

2. Encourage Continual Improvement

If you don’t continually improve everything you do then you don’t stay still, you gradually fall behind as everyone around you moves on. But this does not mean you should introduce continual improvement as a process, with a continual improvement manager to oversee it. This is the “command and control” way of working. Instead you need to make it clear to everyone in the organization that one of their responsibilities is to identify ways that they could work better, and to do something about it.

Some improvements may be very simple, and can just be implemented by the people doing the work, but others may need funding, or input from many people. These will need to be shared and managed. I have written a number of blogs about continual improvement, so I won’t add much more here, but please do read them and consider what you can do.

• Continual Service Improvement (CSI) – The Most Important Service Management Process
• Managing a continual service improvement register
• The Help You Need to Adopt Continual Service Improvement
• 5 Tips to Help Prioritize Your CSI Improvements
• How to Continually Improve Your Service Portfolio – Even If You Don’t Have One!

3. Start Automating

If you automate one task that you carry out fairly often, then this will free up resources to do more improvement work. Automation has many benefits:

• It makes activities more reliable
• It reduces the variation that happens when different people carry out the task
• It provides a structure for making improvements, because updating the automation affects every future instance
• It provides data for metrics and reporting about task frequency, duration, and errors
• It enables you to carry out complex tasks more frequently, for example you could carry out failover testing every day, instead of once a month

Don’t start a big project to try and automate everything you do, just find one thing that you could automate using tools and skills that you already have. When this is working well you will have built up some experience, and you’ll be ready to automate the next thing. Eventually you may need to invest in more tools, but that’s not where you start.

4. Encourage Everyone to Focus on Customer Experience

This is an important culture change, which is easy to describe but difficult to do. Some of my customers require all IT staff to spend 1 day working in a customer-facing business unit at least once a year. This gives them direct experience of both the business unit and the customer view of the organization, AND it helps them to build empathy with their customers.

Another tactic you can use is to modify how you measure people, placing more weight on customer feedback than on raw numbers from ITSM tools – but do be careful because I have seen this go wrong with charismatic IT staff encouraging their customers to submit complimentary feedback.

5. Delegate the Authority to Make Changes

If you have weekly CAB meetings, where many managers sit round a table reviewing and approving large numbers of IT changes, think about how you can work more effectively. Things that have helped other organizations include:

• Defining standard changes. These are pre-approved, low risk, changes that follow a well-defined procedure. They don’t need any other authorization
• Categorizing changes and assigning different change authorities to different categories. This way the CAB can continue to review changes where its input adds value, but other changes can be approved by different people, who are often in a much better position to judge the risks and benefits.

I have written more extensively on these and other ideas to help improve change management in my blog: >How to make IT change management work for everyone.

Conclusion

We can’t all be unicorns, but that doesn’t mean that we have to stagnate with old-fashioned processes, legacy systems, and a command-and-control culture. There are lots of things you can do to improve, and many of them are both cheap to implement and pretty straightforward, i.e. easy to do.

You can make life better for yourself, your customers, and your IT colleagues, by starting to move towards modern ways of working.