Gunnar Kudrjavets

Paranoia is a virtue

Development:test ratio causing software quality to self-stabilize

Today I attended one excellent talk covering software engineering culture in general, things which currently need improvement, some internal case studies etc. Test to development ratio was also pointed out in this talk and one interesting thought was mentioned during this presentation. Microsoft is different from most of the software companies by development and test organizations having very similar sizes. Hypothetical little feature team may consist of 6 SDE-s, 4 SDE/T-s and 1 STE for example. What would happen if ratio of testers and developers will be 1:10 for example and at the same time you would have a top notch development team who’ll be obsessive about code reviews, unit testing, proper engineering practices etc? Natural reaction is that the product quality will be doomed ;-) However, one of the projected outcomes was that the product quality will be self-correcting and will stabilize itself automatically to the level that would be acceptable for shipping? You know, all these factors like people developing the software understanding that they’re the ones responsible for product quality and there’s nobody else ;-) Major assumption is of course that peer pressure and strong personal accountability for the quality will be applied.

P. S. Everyone who has shipped any products where testing was done only by developers can claim that they’ve participated in this experiment already and if their product made enough money then they’ve succeeded ;-)

Posted: Aug 31 2004, 08:52 PM by gunnarku | with 11 comment(s)
Filed under:

Comments

Stuart said:

For people outside microsoft, could you give a one-sentence summary of the meaning of SDE, SDE/T and STE?

From the context, I'm guessing that SDEs are developers and the other two are different kinds of tester.

But the acronyms themselves sound like they should mean, respectively, developer, "developer who also does test work", and tester. But that would mean that the developer/tester ratio is nowhere near the 50/50 you suggest.

Bottom line: I'm confused ;)
# September 1, 2004 10:46 AM

Michael Russell said:

SDE: Software Development Engineer
SDE/T: Software Development Engineer in Test
STE: Software Test Engineer
STL: Software Test Lead

By the way, not every division of Microsoft has a nearly 1:1 test/dev ratio. When I was working in games, it ranged from 1:3 to 1:13.
# September 1, 2004 11:16 AM

Stuart said:

So an SDE/T is essentially a developer whose responsibility is solely to do test-related development work?
# September 1, 2004 12:28 PM

Gunnar Kudrjavets [MSFT] said:

Russell is correct about the definitions. Yes, SDE/T is a developer who does test-related development work. As test organization needs also tools to be written, infrastructures to be developed, testing at very deep technical level done etc., that's where SDE/T-s come to play.

Everything I'm writing about is based mainly on the knowledge I acquired from my immediate workgroup and here we tend to have very similar sizes for both organizations which of course doesn't mean that ration in other teams can't be different ;-)
# September 1, 2004 12:43 PM

Amit Kakade said:

I want to know is there any specific development to testing ratio present or not

# July 30, 2007 12:56 AM

Belal Raslan said:

Well, i think that 1:3 is the most common ratio, but this can't be applied for all organizations. you may check this related URL: www.riceconsulting.com/.../tester_developer_ratio.htm

In an effort to develop some findings concerning the tester to developer ration, I took and informal survey at QAI's 20th Annual Software Testing Conference in September of 2000. Here are my findings:

There were 29 respondents

The minimum ratio was 0 testers to 1 developer

The maximum ratio was 1 tester to 30 developers

The most common ratio was 1 tester to 3 developers

The average ratio was 1 tester to 7 developers

The median ratio was 1 tester to 5 developers

# November 28, 2007 10:37 AM

Tom J. said:

Thanks Belal, i think this is the most realistic ratio and its close industry standards today.

# December 13, 2007 1:26 PM

erw said:

everyone talks about code to test ratio but does that go beyond development and include any functional , or system, testing elements?

# April 26, 2008 2:40 AM

Terry said:

The development to testing ratio has been a point of controversy for years.  In a recent project, given the complexity of the landscape, with 72 countries, 12 languages, a 100 plus interfaces and reports, and 7 functional modules, I found you had to have 1 test hour to 2 development hours.  I've always used 1:3 ratio on other projects, but be cautious as not to quickly apply this industry standard, until you get a feel for what the landscape look's like.

# May 8, 2008 9:42 AM

Chris W. said:

Even within a single organization, the ratio can need to change by project.  Consider the following two scenarios:

1. Brand new to the world product.

2. A one line change to a legacy product.

Certainly every organization will eventually be supporting an old product while developing a new one.  

The first project seems like it would need relatively greater development resources vs. testing than the second.

So there may be industry averages, but the 'correct' answer would seem to depend on the project in question.

# July 25, 2008 12:44 PM

Belal Raslan said:

Factors such as Function Point (FP), team experience and project schedule should be considered when defining the ratio.

# August 16, 2008 6:06 AM
Leave a Comment

(required) 

(required) 

(optional)

(required)