If you're going to disappoint your customers, even if you firmly believe that "in the long run" they'll be better off, just tell them. Flying US Airways last night really brought this home, but there are parallels in our software industry that you'll recognize if you stick with me through the airline part of the story first.

As a traveler, I understand that some of these delays cannot be avoided. Within the same airspace is Philadelphia, Newark, JFK, and Laguardia with Boston, Baltimore, Washington National, and Dullas all close by as well. The air traffic pattern up and down the east coast is just jam packed and a little bad weather can play hell with the schedule. Fair enough. The problem is the Airline plays games with its customers instead of just admitting the problem and posting a realistic expectation for when a flight will actually be leaving. Instead, they don't seem to post ANY delay until the previous flight has already gotten behind even though the same plane is used to go back and forth all day, only the very next segment is impacted on the flight status boards. Then, they start with small delays of just a few minutes and every time you think you're getting close they add another small delay and another until by the end of the evening you've been in the damn airport for hours and hours and still may or may not be going anywhere.

Sitting in the airport waiting as they repeatedly add more and more delay time to your flight while you have no information and can make no plans leaves you, the customer, feeling like a helpless supplicant at the mercy of the big corporate airline. That is the worst thing you can do to a customer.

If you ask the airline why they don't just post a realistic estimate on the board, they'll tell you it is because they don't want you to leave the gate area just in case they can somehow get the flight in sooner. The truth, as I understand it, is that they don't want you making other plans. They're afraid that if they tell you it's going to be three of four hours delayed that you'll decide to cancel and take another route to get there. The airlines KNOW the delays are going to get longer. They have plenty of smart people there who have seen the same situation develop over and over, several times a week, year in and year out. On the long haul flights with connections, they don't want customers demanding to be put on other airlines to make their connections or get to their destinations closer to on time. On the shorter commuter flights this is even more important for the trip inbound from the regional airport as many of those customers are making connections at the hub. For the return flights out to the regionals, customers have other options. For many of us, a short flight can be replaced with a longer (but still reasonable) drive, bus trip, or train trip.

Yesterday, for example, I finished early and got to the airport at about 2pm. My flight was scheduled for 6:25. That's already a 4 and a half hour wait. I can do the drive from PHL to PWM in about 7 hours according to Google maps. I checked, and the earlier flight was full, so I decided (fool that I am) to wait for that 6:30 flight. If they had just admitted at that point that my flight home wasn't going to get out of the gate until 9pm I would have opted to drive. I'd have been home at least an hour sooner and it would have been a far more comfortable and enjoyable a trip. Instead, I spent nearly 7 hours stuck in an extremely over crowded commuter terminal with limited food choices, over used bathrooms, no place quiet to sit (in fact I was sat on the floor for a few hours) and very poor air conditioning. I could have take a shuttle bus to another terminal where at least there was better food and I may have found some open chairs or paid for an airline lounge pass -- but the way the delay system worked I never knew how long I'd have before having to head back. It's not a short trip.

US Airways did save themselves the cost of my one segment not cancelling, but they've increased my chances of finding alternate travel arrangements whenever possible. The train or car is still longer than a plane trip when everything goes right, but because it goes wrong so badly, it tends to even out. I'm more likely to drive in the future when I can take the time to do so. Flights to New York have already gone this way. What used to be a simple commuter flight into the city now takes so long due to parking, security, and ground transportation that it's faster for me to take the train from Boston. That's lost revenue for the airline because they've made it such a hassle.

What could they do? They could tell the truth. They could, if they wanted, put a single really long delay on one flight and use that plane to get the next round trip back on schedule. This works for commuter flights where the same plane is used to go back and forth all day. The customers on that one segment would have a very long delay unless there were seats on the other flights available to put customers on to -- but they could then post this one really long delay and that group of customers would have a choice. If I got to the airport and saw that I was going to be several hours delayed -- but I was confident that it was the truth -- I could leave the airport and go have a nice dinner or see a movie or a museum and come back later. Maybe if I'd been in the middle of a long day of travel I'd take a hotel room and get some sleep. Maybe I'd find another way home. It would be my choice. Even if swapping out one flight isn't realistic for one reason or another, simply posting a realistic estimate that they can commit to would give the customer the real information they need to make a choice about what to do. It would remove that feeling of being a helpless supplicant to the process.

Here's where it ties in to the software business.

Sitting in your office waiting through more and more delays, more and more release updates that don't solve your important problem leaves you, the customer, feeling like a helpless supplicant at the mercy of the big corporate software company. That is the worst thing you can do to a customer.

If you're not going to bother updating large areas of your product that customers still care about -- even if you think "in the long run" it will be better for them, just admit it.

If you're not going to write documentation to include with your product because you don't want to spend the time and money to translate it, just admit it.

If you aren't going to ever fix bugs in some areas of your product because you no longer even have a single developer on staff assigned to that area of the product, for everyone's sake just admit it.

Many of us know this is the case with some software. Through back channels we even hear about which areas of the product have been relegated to this status. It isn't the secret you think it is. All you're doing is angering your loyal customer base in the futile hope that you'll get new and different customers for something else. That's bad business.