Why Proper Checking Systems Are A Good Thing For Performance Quality
Fri 15th October 2010
The best thing for performance quality is to check the work. There are two styles of this, the first is the checker which is another person assigned to briefly examine another person's work to look for mistakes. The other style is the tester which is a person completing in depth system testing on a piece of software for example. The best way to decide what is needed is to look at the output of your team. What exactly is getting sent to the client as something they have paid for? What service is being provided to another team internally or the public? Basically anything that can be used as a measure of yourself and your team should be checked. After all it is far better to pick up any mistakes or problems before things have left. If the client is picking up mistakes then shame on you!
System testing really does depend on the kind of system that needs testing. Testers are a professional group of people that are often experienced with a range of technologies appropriate to the task. They will often have a good level of industry knowledge to aid them in spotting when something is not working. This is an important consideration since you wouldn't expect to pull an ice-cream vendor of the street and get him to complete system testing on a pension transfer system. How would he know if things were coming out right or wrong? The type of testing as well as the quality of the tester therefore needs to be considered.
How you go about testing your product or service is important as well as who tests it. When testing a product like some software or a toy, the aim should be to break it. Sounds harsh I know! Ideally if it is a quality, working product then you will not be able to break it, but if you can then at least you have the opportunity to fix it before it goes out live and you have to recall. For a service, for example a restaurant or shopping facility, mystery shoppers and diners are often used as a way to keep an eye on the staff. When staff are aware that management are watching, they may behave impeccably, but what is the experience of the average user of the service?
When designing and completing a project, always ensure that there is adequate time allowed for testing to take place. It is often the case that a client will be less interested in the testing and may not want to pay for it. Do not be brow beaten by this response, but calmly explain its importance. There must be time for your team to test internally and then for the client to test. When changes are required, they must be made and then tested again. Although this may seem like a laborious process it is vital to ensuring a quality deliverable. There could be a check list to test different parts of the system along with a variety of test data.
Testing and checking must be realistic. A range of data must be entered into a system to check it properly. If issues have previously come up with a specific type of data then it should be added to the test pack as automatic testing for the next release to ensure it does not happen again. Testing and checking is good for performance since you will not have to come back and change things and patch things up at a later date. Proper planning will allow you to manage the performance of your team more realistically.
Original article appears here: