Coding Horror: The Great Enterprise Software Swindle

March 02, 2005

The Great Enterprise Software Swindle

After nearly four years working for a Fortune 50 company, I am now completely convinced that the term "Enterprise", as applied to software, is synonymous with "crappy".

Clearly I'm not the only guy to notice the apparently linear price to bug ratio in the so-called "Enterprise" software I've been exposed to. It's an embarrassment. But it also seems to be a standard part of the business model for these companies. The rampant bugs in their Enterprise Software-- which we already paid millions for-- force us to contract with the "professional services" arm of said company, who then get to work busily fixing the bugs in their own product for $100/hour.

Worst of all, if your organization has used this Enterprise Software for a year or two, you've likely customized the heck out of it. The idea of a one-size-fits-all software package-- particularly one of this magnitude-- is laughable. And yet that's one of the illusions that drives sales of these packages in the first place. It's as if the salespeople watched one too many Warner Brothers cartoons, and they're expecting to get another delivery from ACME Corporation any minute now.

Wile E. Coyote and ACME Corporation

So, once you factor in natural inertia, plus the customizations required to get the functionality you paid for, the cost of conversion to a competing product is brutal. It's easy to get sweet-talked into yet another version of the devil you know, on the off chance that hey-- maybe this year it might not suck. As much.

It's madness. As far as I'm concerned, the word "Enterprise" is now so tainted that it's best used as an epithet. Dude, your software sucks so much, it's enterprise software.

Posted by Jeff Atwood View blog reactions

« On Interviewing Programmers To Compile or Not To Compile »

Comments

Good comment, Jeff! :-)

Oliver Sturm on March 3, 2005 05:10 AM

Many years ago I did work on programmable logic for a system I was working on. At that time, I formulated a rule of software: The more it costs, the less it works. We paid $5000 for a piece of software which allowed one to program and simulate these devices. What the app does is NOT simple, but the bug count was astounding.

I think the more it costs, the less users you have, and somehow that turns into 'spend less on QA'. Sad really.

Michael Kohne on March 3, 2005 07:46 AM

Agreed! In addition to being a programmer, I have been an adminsitrator for both Oracle Financials and Peoplesoft. I can tell you that both of these products are absolutely horrible. There's no excuse for it, but the big corporations just keep buying the junk lilke it's candy...

Of course, both of those products are implemented in Java, but that's another story (or is it?).....

Jeff M on March 3, 2005 08:43 AM

Well, when you have an extremely complex multi-server distributed application being written b y a friend of the boss who once read "Learn Vistula* Basic 4 in 21 Days", you have to expect certain sacrifices...

*(No, that's not a typo; the title refers to a knock-off product from a company in the CIS that closed it's doors in 1999.)

Joseph Osako on March 3, 2005 10:25 AM

I used to work for a software company exactly like this. They sell 'enterprise' systems for the insurance market. The software is terrible, and riddled with bugs, but they continue to sell it to this day and make an enormous amount of money from bug-fixing. They have no formal design process, do no R&D and have no formal development process. (The application is written in VB6 and is comprised of 1500+ (count 'em) COM components >)
Hence, out of a company with 50+ developers, at any one time at least half of them are bug fixing.

It catches them out of course - out of the 6/7 customer implementations I worked on while I was there, every single one of them was late, over budget, and about to result in litigation. The final project I was on was a £3m+ implementation for an insurance company in the Southern Hemisphere - it was 3 years behind schedule and milions of pounds over budget. Eventually the customer canned the project and wrote off the cost, deciding to extend the lifetime of their current inadequate system instead.

But then, the company still turns over £20m+ and the directors live in big house and drive Porsches and Aston Martins to work. It's then I think to myself 'who is the mug here?'

Andy on March 4, 2005 08:02 AM