We've all been there, on a project where the inevitable happened. You guessed it, the code is late in the tests, pronounced the 11 hours. Inclusion criteria were under pressure and may have been completely ignored. Your test environment is delivered, but was not verified because the code was not available. The supervisor had been harassed by the stakeholders of the company in case of late development and is not interested in your problems. IT Management urgently to see the project on time and to top it off, Marketing has organized a campaign launch to occur under the required delivery date. As Test Manager you are now the main obstacle to the launch, the project's success rests on your shoulders. Oh yes, and your test window six weeks was reduced to four.
This is the stress (no pun intended), the phase of the test. Not only do we now need to think outside the box to complete the test, but you also have to think much broader level than just a test. E 'to be undertaken at the project level, which can make a big difference in work-tester.
Let's start by looking at this from a broader perspective. I remember a situation at a very early project that I led. Sad in a meeting of the project and to ask how we can try with a window of execution reduced from four weeks to two. I dug my heels in and refused to budge, leading to a separate meeting immediately after the PM, have been advised that this was not the right way and get training at the beginning of art to the test.
Back to our problem, there are several actions that can be implemented at project level to help the situation and may be able to PM to help with some recommendations. (1) is the application for the need to invest for a living? It is the application of factors that could live as part of another publication, reduce the scope of the required tests? (2) the issue of internal users? This will minimize the damage if the production of errors, which means that the exit criteria can be checked. (3) will be the further development can be applied to the development of new?
Whatever the answer at the project level, may be implemented by the Test Manager to manage the situation in the tests:
(1) Insert testers in the development teams and increase the unit value and integration in the small test. Improve the quality before it hits the test, reduce the scope of errors and so the duration.
(2) intends to make application components to be finished before the deadline specified. It is likely that only some parts of the application that are causing the delay, and not the entire application. Bring in some regions at first, increasing the test window and may restore some lost time.
(3) Apply a technique of risk analysis. Much of the first test question and work through the testing of a risk. Upon expiration of the time, this should mean that only the elements of risk than were removed.
(4) is added to the hours worked on the team. View options around overtime and on weekends. If you use the ability to see off-shore, then a two-day gathering. (A word of caution at this stage, if the testers do not work, they need to support staff development and support environment. Test alone will only increase the blackout makes it difficult to keep pace with development and environmental issues could stop all off-hours.)
(5) Consider some overlap in the stages of testing. For example, if SVS is managed as a separate step, functional testing, to see some of the overlap of functional testing during UAT.
(6) Ensure that the focus is on the high priority the lack of activity. Make sure that the developers are fixing what needs to be fixed first. (Do not ignore the seriousness of this point.)
(7) Monitor the fault of turnaround. If development is delayed, it is a sign of problems and an error will be slow to turn cripple the project, while the test can be completed only if the exit criteria have been damaged.
(8) Can multiple environments made available? There are probably already a requirement for multiple environments, but if you start overlapping phases of testing, functional non-functional, with ATU, so the amount environments needs may increase.
(9) Carry out a review of the exit criteria for the project. Note that this set before the problems occurred and even if the desired result, a compromise may have to go. Work is acceptable and we must not forget that if the coverage is reduced, the number of defects are indicative of only a portion of the test. coverage is 80%, which means you can have only 80% of the defects found and it is good to think that 20% are.
Shares (2), (3) and (5) above, all increase the risk in some way. Ensure that the project and stakeholders have agreed that these risks and, if possible mitigation measures are in place.
In summary, there are many actions that can be tested for slip-face, while maintaining the original delivery date. Do not forget the elements of the project that can make a difference. I'm sure there are more, but we ensure that all aspects are studied and only then begin to yield in the test.
