The one top tip for IT operations and support: Checklists

We have the world of information at our finger-tips. Our operational tools get better every day. Yet things still go wrong on a regular basis. Then we take the opportunity to prove that there is no problem so bad that you can't make it worse. Why do we screw up? And why do we then screw up trying to fix it? My favourite magic-pixie-dust quick-win cure-whatever-ails-you fix (no seriously this time) is checklists.

According to Atul Gawande, author of the Checklist Manifesto, the magnitude of today’s knowledge has exceeded our ability to deliver it safely, consistently and correctly. A simple checklist can bring about striking improvements in almost any field (think aircraft pilots). Checklists are an essential tool to minimise errors and maximise efficiency when the heat is on and you need to act. My one top tip for IT operations and support: have the checklist ready when you need it.

Start by reading the Checklist Manifesto. Come back when you are done.

Already? OK, moving on:

Ever watched one of those aircrash TV programmes? What do the pilots do when three engines fail? The pilot says "Quick, get the checklist for Three Engines Failed" and the co-pilot reaches for a chunky ring-binder which seems to have a checklist for every conceivable scenario.

Pilots understand the importance of a cool head and correct action in a Situation. They understand that a cool head is kept when the thinking is done in advance.

They also understand that Situations are caused in the first place through basic procedural errors and checklists are just as important to prevent them.

To take another similar industry, I was lucky enough to hear Chris Hadfield, singing astronaut, speak at the recent Pink Elephant ITSM Conference. He told us how, when NASA decide a space walk is required they plan it for a week before anybody opens the door. Even when it is an emergency (the space station is leaking) they plan it FOR A WHOLE DAY. (He also gave me the "no problem so bad..." line I used above). What's the best way to capture all that pre-thinking? A checklist.

Checklists don't insult our intelligence: they don't tell us HOW to do something. They are not a script or work instruction. They just remind us that it should be done, and sometimes they also remind us in what order.

Hadfield gave us another lesson: visualise - and plan for - failure more than success. So we have one checklist that describes the normal successful actions. Then we have several more checklists for each of the negative scenarios we can anticipate.

Finally checklists are a learning tool. We all practice service improvement, right? So when an incident or problem or change review captures that an error was made, we fix the cehcklist. And when a Situation was created that we had never foreseen, we create a checklist for how to deal with it next time. (For more on this cycle, see my book Plus! The Standard+Case Approach.)

So what do you need to do to take advantage of this one simple top tip?

First read that Checklist Manifesto book (liar! read it). Then:

• Create a process to capture knowledge for checklists
• Store checklists in your knowledgebase (got one of those, right?)
• establish accountability for each checklist - this is the hard part *
• Train people in what is there, how to find them, and how to use a checklist.
• review and enforce use
• Watch the improvement. Feed back results.
• Capture learnings to improve, including improving checklists

I provide a bunch of free service management checklists here
and some basic principles here.
I did a webinar on Checklists here

* thanks to Joshua Biggley (@jbiggley) for pointing this out