Fog Creek Software
Discussion Board




User Acceptance Testing

I have recently been hired as a computer consultant at a rather large corporation. Among a number of tasks that have come up is the process of performing User Acceptance Testing. I have never done anything like this on a formalized scale, having preferred hallway usability & rapid prototyping to illustrate a program's function to my clients in the past.

This client is currently using a more formalized process for projects and would like me to deliver formal documentation in addition to my more agile methods. Unfortunately, all that I have been given as an example is a worksheet from a different project that hardly pertains to my current system.

Does anyone here know of a good online resource for technical UAT methodologies? I need quick tutorials and research discussions versus paid courses and examples. I'm looking for things along the lines of what can be found on Alistair's site for Use Cases as opposed to the marketspeak you can find on Rational's page.

Alistair: http://alistair.cockburn.us/

Dustin Alexander
Wednesday, April 09, 2003

Do they have a formal requirements document and/or use cases? There are your templates for testing...

Philo

Philo
Wednesday, April 09, 2003

Not really. This project was an ad hoc solution to a reporting problem that grew into a full solution through featuritis. As such, I coded it with a bent towards its output (the reports) as opposed to the usability, which was placed second. I have high level use cases (basically ten or so scenarios for using the system), but no standard requirements documentation or project doc.

Originally the solution was designed to accomplish what it was in a 1-3k lines of VB code. It is now a behemoth weighing in at 25-35k of code (depending on legacy code  and comments).

Dustin Alexander
Wednesday, April 09, 2003

You'll need to go back and reverse engineer something like Use Cases in order to generate the right questions to ask. 

The problem with formal User Acceptance is that its treated as a finishing line rather than a process, often because it triggers payment :-).

Simon Lucy
Thursday, April 10, 2003

This I am fine with. I think my baseline cases give me good grounds to start with and I'm certainly familiar with more formal use case / project planning processes than were used on this system. It shouldn't be a problem to get the user acceptance worked out as far as determining which business flows to place in the documentation.

What I'm looking for is a discussion on the formalized methodology behind acceptance testing. Given a pre-established set of uses for a system, what types of measurements and output formatting have developers and managers found effective in the field? Has any work been done in the area since FURPS came out of HP?

I want a more fundamental understanding of the measurements and functions behind the process. Has anyone had direct experience with this process or something similar?

Dustin Alexander
Thursday, April 10, 2003

Isn't "User Acceptance Testing" an odd phrase? I've never found a user yet that I would accept. Maybe I am not testing them thoroughly enough.


Friday, April 11, 2003

*  Recent Topics

*  Fog Creek Home