Comments on: Software Test Estimation – 9 General Tips on How to Estimate Testing Time Accurately

By: Franklin

Franklin — Thu, 08 Sep 2011 04:07:12 +0000

The Points Which have been Mentioned in this Article is the Factors which will influence the estimation but is not the estimation technique.

By: Viplav Anand

Viplav Anand — Fri, 01 Jul 2011 09:22:34 +0000

I have heard that in general we can take 40% of the development effort as the baselined effort for estimation of the testing time. But in current day scenario, wherein developers have various codes at stake of their googling, can we really benefit from that approach. However even some arguments have come in this regard, klike we have test cases for sign in features also avaialble with us, so it is indeed very tough on this count.

By: Deepti

Deepti — Fri, 10 Jun 2011 09:35:32 +0000

One more menthod of testing estimation is that the tema lead can make the tem to fill a task sheet every week in which they can fill the time spend on various test activities . Based on this sheet the time aestimation can be determined by dividing the total time spend on activities by the whole team by the number of test cases executed.
For example if total team spend 5 hours for 58 test cases to execute then in one hour the team execute around 10 test cases.Based on this the future testing estimation can be made.

By: RAJAN SINGH

RAJAN SINGH — Thu, 21 Apr 2011 08:47:29 +0000

Hi all, I reviewed the complete conversation of this article and feel that the effort estimation is something which depends on the Test policy of the organization too. Agree that all the factors mantioned above are also considered in estimation but the experience is always the key for good estimation.
Leaving one question here…
What is the phase when estimation happens in STLC. If you can share how and when happens in ur organization it would be great.

By: Sushitha

Sushitha — Thu, 03 Mar 2011 07:30:34 +0000

Hi Sandhya,

You have written a simple article which is very practical and gives light to points which have to consider while preparing a test estimation. I totally agree with you on points 2,6 and 9. Keep writing…..

By: Subrata

Subrata — Fri, 25 Feb 2011 09:18:26 +0000

@Amit
Thanks for your comment. before puting my thought I needed to go through all the comment twice to analyse the mentality of those guys-
@ Pradeep Soudararajan and @ Ambarish Karnik.
I have played different role in different time. Estimation at the end depends on the people who executes.
@Ambarish: lets say you are the best person in the world in estimation and you are not wel known with your team or might you have been approached by one client to do the estimation. You have done it. How much assurity you give that it is going to as per your estimation? It really depends on the team co-ordination and the level of skill.

By: nitesh

nitesh — Thu, 20 Jan 2011 11:39:20 +0000

Good article..Sandhya
I appreciate the flow and simplicity of this article. Its Practicl and applicable.

By: Amit

Amit — Wed, 19 Jan 2011 16:59:31 +0000

@ Sandhya,

Your post is great. I am an SDE at Microsoft, and with all my experience over the years… I can say that you nailed it right…

@ Suhas
@ Ambarish Karnik

The Function point analysis and other methods are quantitative and theoretical. What Sandhya has mentioned is purely qualitative and practical. These are tips, I am assuming you know the difference btween a tip and a theory…
For example, you can be an expert in the estimation process… but if you dont know your team “well”.. some assumptions will go upside down… and they won’t be able to deliver as fast as you’d like to believe. In the end you’ll screw up big time. Your Use case analysis or whatever was spot on, but your team dynamics failed.

Appreciate the work Sandhya has done, and complete it with the quantitaive stuff… that will make this a still better post…

@ Pradeep Soudararajan

“at least discuss it with people like me who would test your ideas up front before you consider writing more.”

And you thought discussing something with you will make the artifact absolutely flawless… Grow up man… what do you bring to the table in this thread… except for a made up story involving your cricket star… Stop ranting and get a life!

By: Debs

Debs — Wed, 15 Dec 2010 10:12:05 +0000

The given artical for the estimation is quite ok but it highlights pre considerations before doing the test or effort estimation. It will be useful if some more details are provided with an example

By: Shakil

Shakil — Tue, 19 Oct 2010 10:33:52 +0000

Hi Sandhya,

your point, ‘Parallel Testing’ is not clear to me. You said, if we have previous version of a product so we can compare the output. Don’t you think Comparing with previous version of product will be repeated task? Parallel Testing, if it starts parallely with development then it will save our time.

Please elaborate it so that i can understand what exactly what this point says.