The Hidden Costs of Bad Software

When software is frustrating and difficult to use, people will avoid using it. That is unremarkable until you realize that many people's jobs are dependent on using software. The corporate cost of software avoided is impossible to quantify, but it is real. Generally, the costs are not monetary ones, anyway, but are exacted in far more expensive currencies, such as time, order, reputation, and customer loyalty.

People who use business software might despise it, but they are paid to tolerate it. This changes the way people think about software. Getting paid for using software makes users far more tolerant of its shortcomings because they have no choice, but it doesn't make it any less expensive. Instead while the costs remain high they become very difficult to see and account for.

Badly designed business software makes people dislike their jobs. Their productivity suffers, errors creep into their work, they try to cheat the software, and they don't stay in the job very long. Losing employees is very expensive, not just in money but in disruption to the business, and the time lost can never be made up. Most people who are paid to use a tool feel constrained not to complain about that tool, but it doesn't stop them from feeling frustrated and unhappy about it.

One of the most expensive items associated with hard-to-use software is technical support. Microsoft spends $800 million annually on technical support. And this is a company that spends many hundreds of millions of dollars on usability testing and research, too. Microsoft is apparently convinced that support of this magnitude is just an unavoidable cost of doing business. I am not. Imagine the advantage it would give your company if you didn't make the same assumption that Microsoft did. Imagine how much more effective your development efforts would be if you could avoid spending over five percent of your net revenue on technical support.

Ask any person who has ever worked at any desktop-software company in technical support, and he will tell you that the one thing he spends most of his time and effort on is the file system. Just like Jane in Chapter 1, users don't understand the recursive hierarchy of the file system the Finder or Explorer on Windows, the Mac, or Unix. Surprisingly, very few companies will spend the money to design and implement a more human-friendly alternative to the file system. Instead, they accept the far more expensive option of answering phone calls about it in perpetuity.

You can blame the "stupid user" all you want, but you still have to staff those phones with expensive tech-support people if you want to sell or distribute within your company software that hasn't been designed.

The Only Thing More Expensive Than Writing Software Is Writing Bad Software

Programmers cost a lot, and programmers sitting on their hands waiting for design to be completed gall managers in the extreme. It seems foolish to have programmers sit and wait, when they could be programming, thinks the manager. It is false economy, though, to put programmers to work before the design is completed. After the coding process begins, the momentum of programming becomes unstoppable, and the design process must now respond to the needs of programmers, instead of vice versa. Indeed, it is foolish to have programmers wait, and by the simple expedient of having interaction designers plan your next product or release concurrently with the construction of this product or release, your programmers will never have to idly wait.

It is more costly in the long run to have programmers write the wrong thing than to write nothing at all. This truth is so counterintuitive that most managers balk at the very idea. After code is written, it is very difficult to throw it out. Like writers in love with their prose, programmers tend to have emotional attachments to their algorithms. Altering programs in midstride upsets the development process and wounds the code, too. It's hard on the manager to discard code because she is the one who paid dearly for it, and she knows she will have to spend even more to replace it.

If design isn't done before programming starts, it will never have much effect. One manager told me, "We've already got people writing code and I'm not gonna stop." The attitude of these cowboys is, "By the time you are ready to hit the ground, I'll have stitched together a parachute." It's a bold sentiment, but I've never seen it work.

Lacking a solid design, programmers continually experiment with their programs to find the best solutions. Like a carpenter cutting boards by eye until he gets one that fits the gap in the wall, this method causes abundant waste.

The immeasurability and intangibility of software conspires to make it nearly impossible to estimate its size and assess its state of completion. Add in the programmer's joy in her craft, and you can see that software development always grows in scope and time and never shrinks. We will always be surprised during its construction, unless we can accurately establish milestones and reliably measure our progress against them.

Opportunity Cost

In the information age, the most expensive commodity is not the cost of building something, but the lost opportunity of what you are not building. Building a failure means that you didn't build a success. Taking three annual releases to get a good product means that you didn't create three good products in one release each.

Novell's core business is networking, but it attempted to fight Microsoft toe-to-toe in the office-applications arena. Although its failed efforts in the new market were expensive, the true cost was its loss of leadership in the networking market. The money is nothing compared to the singular potential of the moment. Netscape lost its leadership in the browser market in the same way when it decided to compete against Microsoft in the operating-system business.

Any developer of silicon-based products has to evaluate what the most important goals of its users are and steadfastly focus on achieving them. It is far too easy to be beguiled by the myriad of opportunities in high tech and to gamble away the main chance. Programmers regardless of their intelligence, business acumen, loyalty, and good intentions march to a slightly different drummer and can easily drag a business away from its proper area of focus.