KeyCruncher.com

Joel Spolsky of JoelOnSoftware.com and FogBugz fame recently wrote a brief blog entry describing the exciting new Evidence Based Scheduling feature of FogBugz; this is a revolutionary concept that I have yet to see implemented well in any other COTS software and one that I would love to see extended to trouble ticketing systems.

---

You cry some and then realize that the computer might still be useful for something: you can draw yourself a bath, get in and then drop the damn thing in with you to end the whole mess now…

---

As a customer I, like many of you, loathe trouble ticket systems, especially when they are used in place of real human communication. Trouble ticket tracking can be a very useful tool, but far too many technology companies use them as a moat around their castle, treating the customer like a marauder attempting to breach the gates. “Whatever you do, don’t talk to the customer!” seems to be the motto of too many tech companies and this is exactly the reason our industry has such a well-deserved reputation for bad customer service.Secondly, I have yet to see a trouble ticket system which does a good job of analyzing and reporting on estimates vs. actuals, nor have I seen a trouble ticket system that is decent at providing feedback on how changes in priority can affect promised outcomes. These predictions are precisely what is needed to gauge confidence in a plan, perform more accurately in the future and exceed customer expectations.

If you have ever used a trouble ticket system as a customer, you probably experienced something like this:

• You try to call or find someone to ask about your problem. Of course there is not a phone number. We only call you. If you are lucky, you find an email address or a form to fill out to request help.
• Email the issue or fill out the form to create a trouble ticket. It is unfortunate that you usually cannot actually speak to a technical person at this point (precisely why an “automated” trouble ticket system exists) because your issue is usually hard to explain. Of course, you have put off reporting the issue exactly for this reason, putting up with the problem and becoming more and more frustrated until you decide it might actually be a good idea to ask for help. By this point, you are pissed and you want the tech guys to know. Your email probably contains a profanity or two to make your frustration level abundantly clear. Never mind the fact that your problem may be compounded by the fact that you tried to fix it yourself.
• You receive an automated response. At this point, it is highly unlikely that anyone has actually read your plea for help; instead, your message has probably vanished into the dark hole of automated queues. Perhaps someone will read it within a few days. Perhaps nobody cares and your message will never be read. Not surprisingly, the automated email contains not one hint of realistic promise as to when you can expect a resolution or a response. You cry some and then realize that the computer might still be useful for something: you can draw yourself a bath, get in and then drop the damn thing in with you to end the whole mess now…

Now consider this scenario instead: You recognize a problem, call it in to a live person who understands your unique compu-term dialect and while you are on the phone with them, your issue is appropriately routed and you are provided with an accurate promise of resolution time based on the team assigned to your issue, the accuracy of their past estimates, and the severity of your issue as compared to others already in front of you in line. Evidence based scheduling could allow this dream to become a reality.The great thing is that the evidence based scheduling concepts found in FogBugz would not be at all hard to port to a trouble ticket system. In fact, the code is most likely already abstracted enough to copy-and-paste into an interface tailored to service delivery instead of product delivery and one might argue that FogBugz could be used in current form to do just that. So, the question becomes, when will Fog Creek Software recognize the opportunity and seize it?

Filed under: Business, Technology