Fog Creek Software
Discussion Board

What  are some reasons companies give for not tes

Please help me come up with some reasons

Thursday, May 8, 2003

Please fill out the subject line in another message or repost with a shorter title.  Thanks :)

Thursday, May 8, 2003

not enough time/resources
programmers should do the testing / release bugfree work
we'll do it later

Thursday, May 8, 2003

some comapnies think: "our customers are stupid"

Daniel Shchyokin
Thursday, May 8, 2003

Why are you writing bugs? You're supposed to be writing code!

Nat Ersoz
Thursday, May 8, 2003

At the first company I worked at out of college (as a tester) there was in an eternal struggle to improve their testing and quality efforts (genrally in a three steps forward, two steps back fashion).

The company had been the first out with products to serve their target market, and had done so without any sort of real QA or testing. Yet, by being first, having a product that despite its warts was very useful and agressive and connected sales they rose to (by their estimate) about 80% saturation of their potential customers.

So, the old guard had a real 'we got this far without testing, why do we need it now?' attitude - even as the flagship product got more complex and they added new ones that automated even more critical parts of our customers' operations.

(Long story short, the new product was a support nightmare and a very hard sale - and by the time I left there was nearly a one-to-one tester-to-developer ratio for it and there were more unplanned midnight shifts to fix problems in the field.)

Thursday, May 8, 2003

d'oh, that should've been "there were NO more unplanned midnight shifts to fix problems in the field.)"

Thursday, May 8, 2003

Business managers have one purpose, make money. In the eye of a business guy, Q&A costs money and they have no evidence that it will save them money, in any meaningful way. So they tend to think that Q&A is a money sucker.

Of course as developers we could reason that for every bug found in-house by Q&A there's one less potential bug that could blow up in the field. How that translates into money, you are going to have to figure it out if you want to convince them. Presumably it will save money in tech support. But you are going to have to detail how many Euros you save on tech support per Euro invested in Q&A, or they won't buy it.

That is, if you are lucky enough to work for a shop where they have tech support professionals. If developers are on salary and happen to be, developers, testers and the tech support team all in one, then I wish you good luck trying to get management buy-in about the benefits of Q&A. For them it makes little difference to overwork some of the developers before the release to do Q&A, or overwork them post-release to do tech support. When their pretty spreadsheets add the numbers up, it's all the same to them.

All they risk is their reputation by releasing buggy software onto the public. Sadly enough, the public is accustomed to poor software releases and are willing to apply patches if a bug bites them. The internet made matters worse in this regard, because patches can be offered freely in the company's website at little cost for the company.

Given the choice of being first to market with a buggy product or pushing the release date in favour of a strong development process. Most business people will happily contribute to the amputation of Q&A from the process (Q&A or whatever other expense they deem unnecessary).

Of course, not all software shops are this pathetic but many of them are.

Beka Pantone
Thursday, May 8, 2003

Because customers select on price and  features, not on quality.
Hey, Open Source even managed to make this into a feature: Let the customer handle the debugging, documentation and maintenance. :-)

The main problem is that quality of software products is very hard to quantify. Most of the quality issues are based on reputation, and in an almost fanatical religious context such as the world of software, reputations are based on all sorts of dogmas and anecdotes that are often a far cry from any objective reality.

Just me (Sir to you)
Friday, May 9, 2003

Reputation: Potential FTC fines of $2.2 trillion for passport security breaches.

I'm guessing just the tip of the iceberg, eh?

Nat Ersoz
Monday, May 12, 2003

*  Recent Topics

*  Fog Creek Home