One of the questions on the Quality Ready Assessment (QRA) asks about the interaction between development and testing teams. One of the possible answers is:

The teams do not interact. “New builds are thrown over the wall” when they are completed.

So I decided to ask the question, what percentage of those companies ship on time and on budget? I even gave them some wiggle room since sometimes even the best-laid plans don’t work out. All I was looking for shipping on-time greater than 75% of the time. What did I find?

11.76% could ship on schedule and not over cost.

Ouch.

Wait, I’m sure that this is only like one or two companies that have taken the QRA. Let’s go see how many do this throw over the wall business…

15.86% (!)

Aye carmaba! To put that in perspective that is one out of every six companies.

I spend an obnoxious amount of time in the application lifecycle and quality worlds. I also have been very lucky in that every company I’ve ever worked for has had mixed teams of developers and testers. To me the concept of throwing a build over a wall is as foreign as the game of cricket.

(Have you ever tried to follow cricket? I was working in London once and I desperately tried to figure it out. I even had one of the British guys walk me through it as slowly as humanly possible. I think I lost all higher brain processing ability when he explained that some cricket games can last up to five days.)

Judging by those shipping numbers I’m glad the experience is foreign to me and I really feel for those poor souls that are stuck in that situation. I suppose if anyhing the QRA provides some pretty tangable data that that is not the optimal way of doing things.

Kind of like if you are the batsman trying to French cut a googly into the square leg.

(I mean, if that’s not good. I have no idea what I just said there.)

A few months back we released the Quality Ready Assessment (QRA), a fancy web based tool to help people measure their overall software quality level. We knew we wanted to create something that could provide quantifiable measurements about a subject that oftentimes is looked at as the eating right and exercise of software development.

“Yes I know that I should focus more on software quality but have you ever tasted bacon? It’s freaking awesome.”

Anyhow, did you know that 80% of the people that have taken the assessment so far use some kind of SCM tool? Of course you didn’t I just told you that, but that’s like 4 out of every 5 respondant. Wait, that’s exactly 4 out of every 5. What do you know, 5th grade math did come in useful.

Note to self: Send thank you/apology card to fifth grade teacher.

Anyway, lots of interesting patterns can be found by looking at the data. For example:

“It looks like most shops use an SCM tool now, but a large percentage of the time it is a stand alone system.”

“Companies in the financial services industry sure like integrated toolsets.”

“Companies in the greasy spoon breakfast diner industry don’t seem to care much at all.”

Going back to the first stat though, that means that 20% do NOT use an SCM system. This got me curious about the overall quality practice level of shops that use SCM vs. those that do not. Since I happen to know people who know people I was able to get that comparison report created. The numbers are enlightening:

In the Track metric SCM users scored 62% higher than non-SCM users.

In the Test metric SCM users scored 47% higher than non-SCM users.

In the Automated metric SCM users scored 70% higher than non-SCM users.

In the Change metric SCM users score a whopping 120% higher than non-SCM users.

I don’t need my 5th grade math to tell me that is a huge difference. It also highlights just how fundamental a SCM system is to software development and how its usage indicates a higher level of overall skill. I have to admit though, I never expected the numbers to be so big. Moral of the story, use SCM. It will absolutely help you ship a quality product. Oh and try to cut back on the bacon, it’s all about moderation people.