Learning how to estimate testing assignments
Submitted by aakashvakil on Fri, 14/07/2006 - 09:50.
metrics | project management | this.site
Last week was very important in my life span as a Software Tester. I realised that no matter how precise you are at estimating efforts and schedule - estimates are just that - Estimates - rough idea initially, but as you refine it - you get a more solid view of what is realistic estimate.
I Was surfing on QAForums -> Estimation & Planning section and read that estimation is all about experience. Tools and Templates help you to some extent - calibration needs to be done by you.
Though I got most of the stuff written there, I was not totally sure I agreed on all of that. The reason for my 'not sure' feeling is that most of the techniques are from an assumption that you are going to propose the estimate to management and client and based on that the go-live dates will be decided.
I thought to myself - how many product dev companies are there who would care for your testing estimates ?? I am not telling that product companies dont care for estimates - they are ones who care the most. What I am pointing is that product companies who create-test-sell their own product form a niche percentage of market - the major part outsource some part of dev-test work to the Service Vendors / Offshore Companies. They dont have their own products...but provide testing service to others.
That got me started and I have decided that I am going to learn practical ways to estimate better - so that we, as vendors can help the client realise that we need more time for testing or we can add value by reducing the testing time and subsequently saving client money.
So thats it for now....I have a target that I need to achieve. At the end of this - I hope to atleast understand what are the problems in our current estimation models and how can estimations help in educating the client and relationship building.
Stay tuned for more as I proceed on this very interesting journey....
Got a few reference sites for Estimations and one book. It says estimates are done at two levels - Macro and Micro.
I Was surfing on QAForums -> Estimation & Planning section and read that estimation is all about experience. Tools and Templates help you to some extent - calibration needs to be done by you.
Though I got most of the stuff written there, I was not totally sure I agreed on all of that. The reason for my 'not sure' feeling is that most of the techniques are from an assumption that you are going to propose the estimate to management and client and based on that the go-live dates will be decided.
I thought to myself - how many product dev companies are there who would care for your testing estimates ?? I am not telling that product companies dont care for estimates - they are ones who care the most. What I am pointing is that product companies who create-test-sell their own product form a niche percentage of market - the major part outsource some part of dev-test work to the Service Vendors / Offshore Companies. They dont have their own products...but provide testing service to others.
That got me started and I have decided that I am going to learn practical ways to estimate better - so that we, as vendors can help the client realise that we need more time for testing or we can add value by reducing the testing time and subsequently saving client money.
So thats it for now....I have a target that I need to achieve. At the end of this - I hope to atleast understand what are the problems in our current estimation models and how can estimations help in educating the client and relationship building.
Stay tuned for more as I proceed on this very interesting journey....
Got a few reference sites for Estimations and one book. It says estimates are done at two levels - Macro and Micro.