The company I'm employed at is involved with client based web application development. Currently,for qa, only 15% of development resources are allocated. The argument for resisting the increase in allocation is the problem management has with convincing the clients to accept increase budget and cost associated with greater qa time allocation. How would I go about making a good case for increasing the 15% resource allocation to ateast around 30%? Also, due to the scope of most projects, they do not require much, if any automation.
Normally to convince to get either more resources or more time, best way is to come up with detailed granular level plan and estimates. That gives better idea about the exact requirement for resources and time required. But if you are in early stage of project then may be some quantitative measurable numbers from past experience should help you. Like E.g. in past you had a web application with 100 screens and with say 5 resources and it took you 10 days to complete. that means with 1 resource the estimate would be 50 days. Based on these, also just keep in mind the complexity level of the applications as well you should be able to put a rough high level number.(with deviation of 10% as you are in early stage where all planning and estimations are yet to be done)
This is my experience, if we give a quantitative measurable numbers, people get a clearer idea and negotiations become easier.
Do let me know in case I have misunderstood your questions.
Also suggestions are always welcome [img]/images/graemlins/smile.gif[/img] feel free to information
Why isn't 15% enough? Why is 30% the right number?
Once you can articulate the answer to those questions, you can talk about it to management and see if they agree. Then, they can sell it to the client.
In my opinion, determining the QA resourcing based solely on a percentage of the development resourcing makes no sense at all. The time required to test something is contextual - it depends on factors that may have little or nothing to do with how long it takes to develop that feature.
One way is that you can make a case by showing some supporting evidence. There is a Gartner report called "Q&A for Sizing the Test Team" by Thomas Murphy that talks about how much testing is enough and how many testers are required.
Also you can make a business case talking about the impact of quality i.e. risk involved in going for the release without the required quality.