Fog Creek Software
Discussion Board

Quality less important for business software?

I stumbled across this site by accident and found many of the articles on software development to be quite insightful.  One thing that struck me about all the articles, whether about software development or business strategy, was the focus on software quality as a determining factor in the success or failure of a software business.  It has been my experience that the quality of the software was one of the least important factors in the success or failure of a business.  Sales and marketing were considerably more instrumental in the rise and fall of different companies where I've worked.  What may explain this apparent discrepancy is that most of Joel's experience appears to be with mass-market retail software (lots of potential users and relatively low price) and all of my experience has been with enterprise business software (fewer potential users and high license cost).  I've worked on multiple applications that were nightmares from a software engineering perspective.  They also had poor user interfaces, ran slowly and crashed too often.  Customers still bought these expensive packages and didn't want to give them up even after those product lines were discontinued.  I suspect this was the case because these applications were targeted to a specific functional niche, solved a distinct business problem and were costly to replace.  As such, sales and marketing were more important because effectively reaching a smaller market had more impact on the bottom line than a first-rate product.

I'm not sure that the principles in Joel's more strategic articles apply to enterprise business (or just non-retail) software companies.  I would be interested other opinions on the matter as I don't particularly like the conclusion to which I've come.  It's demoralizing to be largely inconsequential to the success of a company if that's in fact the case.

Saturday, January 12, 2002

A large part of that equation is that the enterprise software has tied along with it a hefty support and maintenance charge that insures the user from a good deal of the mess.

That's from the MIS point of view.  The time from their ticket to supplier to fix is reasonable and they gain in experience and skill as they go along.  For the actual corporate user its an ugly nightmare.

The last person MIS, the supplier, the hired consultant seems to take into account is the poor schmuck having to use their software.  The product and service solves MIS and corporate needs, it rarely even touches the individual user's.

That was why first Lotus and then Microsoft started to clean up on the corporate arena.  It was cheap enough to be a departmental buy.  Of course as MIS decided on Lotus or Microsoft as their integrated office they started imposing their own view on users, enforcing templates having them go through hoops to achieve the simplest things.

The wider enterprise systems are still in the fat hourly chargeable arena.  That will change too.

Simon Lucy
Saturday, January 12, 2002

I think this article relates directly to the issues you mention.

To quote:

you can build the best dang computer in the world with a killer architecture, but if it doesn't run Microsoft Excel, nobody wants one.

I think companies with good marketing departments often do better because the existance of a good marketing department is evidence of a companies strategic approach.

You describe companies that have lock in on their customers.  It seems that many of these companies are wide open to loose their lead to a smarter company.

However, rather than focusing on the causes of lock in and relieving them, they instead focus purely on building a technically superior system.

Ged Byrne
Saturday, January 12, 2002

Ah, but this idea of saying you don't need the best talent causes two things:

1) The quality of product is less. Many a success *is*based on the very fact that group of people did accomplish something great...or at the very least 1, or 2 years ahead of the competitor. In other words...don't ever write out talent, and good software. The stats show up to 8 out of ten software projects don’t make it. With talent...those projects do make it. (you might just think the quality was not good...the alternative was to get no product at all)

2) Did you ever work with a incredible talented team? There is NOTHING more enjoyable in the world working with a super star. If you ever worked on a team loaded with special know what I mean.. If you experienced the will work for free just to be part of that team. Of course, by the time you get to work on that certainly will not be working for free!

A good friend of mine is a nurse and she had been working in the research department at the university. Just recently due to shuffling she is now back in the surgery unit. She said she loves working there now. Virtually everyone on the team is educated, and extremely talented. She can’t wait to get up and go to work. Just the wit and humor around these people is incredible.

Anyone who does not understand that working with great talented people is the most enjoyable thing in life probably does not understand the joy one receives from developing software. Groups of talent do special things...and they know it...and enjoy it..

If you can build that special team....nothing else matters!

Everyone here can read and write. Can we write for a living? Tom Clancey writing skills means he can walk into any publisher and get a 1 million dollar advance for a book not yet written.

Software is like writing....and the *difference* in talent makes more of a difference than most professions. If you are producing a product, or service that does not take much talent..then don’t worry. Unfortunately, software is something where talent makes a more difference than most professions.

Yes.....let me work with talent. If you think you can write like Tom Clancey...then go for 1 million awaits you!

Albert D. Kallal
Sunday, January 20, 2002

*  Recent Topics

*  Fog Creek Home