Topic: Estimating Test Time and Resource Needs

Message

I've reviewed some past postings on estimating QA test time/effort and
resource needs on software projects. Boy, looks like everyone mainly
uses the percentage of development time approach OR guessing test time
and resources based on past experience.C'mon!
There's got to be a more accurate approach. First, a VERY accurate
estimate of full test cycle time should be possible just be estimating
the time required for one person to execute each of the listed test
cases. From there, I guess you could look at the developers involved
and see what their "single fix rate" has been on past projects.
Meaning, what percentage of a developers bugs pass regression testing
on the first pass of testing.
Where I
get stumped is on number of test cycles. I know that the times for each
test cycle should ideally reduce as you progress through the testing
phase, but by how much and what are the total number of test cycles?
Curious
if anyone has come up with a really accurate method of estimating QA
test time/resource needs for projects. My current company is using the
CLass C, B and A estimate rule. Class A estimates can be made after the
design phase is complete and theoretically should be +/-10% of the
actual time needed.

Author	Message
tanu_2006 Newbie Joined: 29Apr2007 Online Status: Offline Posts: 15	Topic: Estimating Test Time and Resource Needs Posted: 05May2007 at 4:40am
	I've reviewed some past postings on estimating QA test time/effort and resource needs on software projects. Boy, looks like everyone mainly uses the percentage of development time approach OR guessing test time and resources based on past experience. C'mon! There's got to be a more accurate approach. First, a VERY accurate estimate of full test cycle time should be possible just be estimating the time required for one person to execute each of the listed test cases. From there, I guess you could look at the developers involved and see what their "single fix rate" has been on past projects. Meaning, what percentage of a developers bugs pass regression testing on the first pass of testing. Where I get stumped is on number of test cycles. I know that the times for each test cycle should ideally reduce as you progress through the testing phase, but by how much and what are the total number of test cycles? Curious if anyone has come up with a really accurate method of estimating QA test time/resource needs for projects. My current company is using the CLass C, B and A estimate rule. Class A estimates can be made after the design phase is complete and theoretically should be +/-10% of the actual time needed. Post Resume: Click here to Upload your Resume & Apply for Jobs

	IP Logged


One Stop Testing Forum : Quality Assurance @ OneStopTesting : Estimation and Planning