Friday, June 02, 2006

Medical Billing - The QA Tester's Headaches

In a previous installment of medical billing software, we covered the many nightmares that a programmer has to go through to get that medical billing software on the market. In this article, we're going to reveal what the poor QA tester has to go through when getting the module fixes from the programmer. In the world of major headaches, this ranks up there with the worst of them.

The QA tester basically takes what the programmer does and makes sure it works the way it is supposed to work. But that's not where it ends. The QA tester, in smaller companies, also has to write up the documentation to show the end user how the software is supposed to be used. Sometimes just one wrong instruction can mean the difference between zero support calls for the software and a hundred calls an hour. What follows is a typical example of how this happens.

The medical billing software company is making an electronic billing module. The module requires that the user use a particular kind of modem with certain settings. The programmer sends the module to the QA tester and the QA tester walks through the procedure. The QA tester determines that the module is working correctly. What he or she doesn't realize is that the modem being used for the test was the wrong kind of modem and would only work on Windows 95 machines and not on 98 or 2000 machines. So the documentation goes out telling users that the modem has to be set a certain way but doesn't mention that this won't work on Windows 98 or 2000 because of a com port problem.

The software gets shipped out and the customers start to install the software, many of them on Windows 98 and 2000 machines. Much to their surprise, they find out that the modems do not respond to the commands from the electronic billing module. The calls start coming in. The first thing support does is ask the QA department if they bothered testing the software. Of course the answer is yes. They then demonstrate and low and behold, they are using a modem that is not supported. Turns out that they discover that this will not work on Windows 98 and 2000 machines. So what now?

The module has to go back to programming and somehow they have to figure out how to make the software work on Windows 98 and 2000 machines because of the com port issue. They ultimate create a program to emulate Windows 95 protocol for com ports and everything is fine.

It turns out that all of this could have been avoided if the QA tester had tested this on both a Windows 95 operating system and a 98 and 2000 operating system. But the software was only tested on 95 on an old modem so of course it was going to work.

In a previous installment of medical billing software, we covered the many nightmares that a programmer has to go through to get that medical billing software on the market. In this article, we're going to reveal what the poor QA tester has to go through when getting the module fixes from the programmer. In the world of major headaches, this ranks up there with the worst of them.

The QA tester basically takes what the programmer does and makes sure it works the way it is supposed to work. But that's not where it ends. The QA tester, in smaller companies, also has to write up the documentation to show the end user how the software is supposed to be used. Sometimes just one wrong instruction can mean the difference between zero support calls for the software and a hundred calls an hour. What follows is a typical example of how this happens.

The medical billing software company is making an electronic billing module. The module requires that the user use a particular kind of modem with certain settings. The programmer sends the module to the QA tester and the QA tester walks through the procedure. The QA tester determines that the module is working correctly. What he or she doesn't realize is that the modem being used for the test was the wrong kind of modem and would only work on Windows 95 machines and not on 98 or 2000 machines. So the documentation goes out telling users that the modem has to be set a certain way but doesn't mention that this won't work on Windows 98 or 2000 because of a com port problem.

The software gets shipped out and the customers start to install the software, many of them on Windows 98 and 2000 machines. Much to their surprise, they find out that the modems do not respond to the commands from the electronic billing module. The calls start coming in. The first thing support does is ask the QA department if they bothered testing the software. Of course the answer is yes. They then demonstrate and low and behold, they are using a modem that is not supported. Turns out that they discover that this will not work on Windows 98 and 2000 machines. So what now?

The module has to go back to programming and somehow they have to figure out how to make the software work on Windows 98 and 2000 machines because of the com port issue. They ultimate create a program to emulate Windows 95 protocol for com ports and everything is fine.

It turns out that all of this could have been avoided if the QA tester had tested this on both a Windows 95 operating system and a 98 and 2000 operating system. But the software was only tested on 95 on an old modem so of course it was going to work.