When Joakim and I wrote the book we had a chapter on measurement in it, chapter 11 - “Using metrics to guide improvements”. It was intended to show a few ways that metrics can be used in a flow-based process that uses kanban for improvements.

When we wrote it I happened to show it to Torbjörn Gyllebring since he’s very sincere in his criticism. His first words:

You can't write a word about measurements if you haven't read "How to measure anything"

When you have not read that book and writing a lot of words on measurements… hearing that has a bit of a “DOH!”-effect on you and your writing. But Joakim had and that made me feel a little better. I was largely satisfied with the chapter too.

But now I have read it and I wanted to share some of the main points that I’ve picked...

Just a short post about data and a common objection. At my current client we have a lot of data about the customers (patients at a hospital) that we serve each day. We have measured the same way for about 4 months now so it’s pretty accurate.

Lately I started to see a trend about how the patients is spread. Here’s a typical month. See how the Sundays is really bad (yeah, that’s the super low points). But there’s another trend here. The weeks keeps falling - I thought at least. The Mondays are always best and the number of served patients gets lower and lower.

What can we learn from that?

Verification

First I verified my hypothesis by a proper analysis. I created an excel formula that went through all our data and calculated the average per weekday.

There’s a Excel formula...

I’ve just started to use Jekyll as my blogging engine. It’s mostly nice but I’m getting used to a new tool. And maybe actually the lack of tools since it’s just markdown and git.

One of the things that I found early to be a stumbling block was to create a new post. Since I’m still fresh to the structure of the YAML front-matter I found myself copying and pasting. Missing and missunderstanding.

So I looked for a post generator and found this gist that is used, at the command line, to scaffold up the structure of a new blog post.

Let me show you how I tweaked it and a problem that I ran into, being a newbie.

Tweaking the script

I’m very much a newbie in bash script files so bare with me and please enlighten me if this can be better.

First...

The “Stop starting - start finishing”-slogan has been the call to action for kanban practitioners for a long time. In Stockholm there’s even a conference each year named just that. And we used it as our first picture in the book.

It’s a great saying and teaches us a lot, and lately I’ve got a new practical experience of the implications of “stop starting - start finishing”. The meaning of “Stop starting - start finishing”, to me, like a guiding star and policy that we agree on in the team, or in the company: here we try to complete things before we start new things.

I think it was Karl Scotland said it like this:

It's not the more we start the more we finish - it's the more we finish the more we finish.

The reason for us doing whatever we...

At my current client we are gathering the most important data (number of customers) for the company well-being and showing it to all the staff, every morning. This is great and have proven very useful to get the attention and interest for everyone. We have spontaneous applauds when we are doing great - we have discussions (also spontaneous) on days with bad data.

After the morning meeting with the entire staff (quite literally the Morning Prayer, being a Salvation Army hospital :)) we hold a morning briefing with the extended management of the hospital.

Lately I’ve stared those meetings a bit different. Quite simply I just point to the diagram with the data up to yesterday and ask:

Kenapa?

The result was a bit surprising and also rewarding Of course (:)) Kenapa means “Why?”. Why did that data happen yesterday? Is there something special that happened?

I noticed...

I don’t consider myself a Excel expert user but recently I’ve started to use it more and more and both come to like it and start doing some pretty advanced stuff with it. As always this kind of knowledge cannot be had in faked, training environment - for me it has to be something real to stick.

We have quite a lot of data for one of our hospitals that we now can get some pretty good trends from. But when I wanted to show only part of the trend line on a diagram showing part of the data … I ran into problem with the default, tooling suggested, ways of doing things.

I had do it myself a little bit, and try to extract some Math-skills from way back when. Luckily I had good help around me…

In this post I’ll show you what we did to get part...