Quality On Time.

I have talked about Quality in my earlier article ( https://www.dhirubhai.net/pulse/quality-varun-maheshwari/ ) and here let us discuss how can we move towards Zero defects & deliver projects on time with the best quality possible.

By the way, all of the credit for these learnings goes to Tom Gilb & Niels Malotaux.

Ok. Now, Zero defects is a mindset like agile. I agree that it is impossible to have Zero defects but the trick is : Even if there are defects , the customers should not find out.

People make mistakes. We are people. Hence, if we produce something, there are defects (caused by these mistakes). The longer these defects stay in the product, the more it costs to find and fix. So, when do we remove these defects: immediately after injecting, or even better: by preventing making the error in the first place.

The basic premise in all of below points is : Everyone is trust worthy & doing their best. It sometimes turns out that people do not realize where to do their best and that causes problems. Deming : “It is not enough to do your best; you must know what to do, and then do your best.”

Some things to consider, only if we are serious about Quality:

• Do requirements specify certain number of defects to be produced ? If not , should we produce them ? And do testers check that we produced the required number of defects?
• Our business is paying us to produce things which our customers love and will pay for. Will customers pay for defects ? If not , should we produce defects ?
• We all know that acceptable number of defects in production is ZERO. So, what happens in reality is : First we put defects in our product and then we ask someone to find it and then we fix it. So, how about : Do not put defects at the first place , so then obviously it cannot be found because it does not exist. Time savings. Better Quality costs less.

Example : There is a huge budget to work towards preventing defects in the first place, and catching them immediately after injection (numbers are rough estimates ??):

a)     Doing it right costs 10 $

b)     Doing it wrong costs 10 $

c)     Finding out it is wrong costs 10 $

d)     Finding out where it is wrong costs 10 $

e)     Doing it right costs 10 $

f)      When working on b) - e) we are not working on the next a)

g)     Hence not doing it right the first time costs 4 times (or more if caught later) as much as doing it right the first time.

• Definition : A defect is the cause of the problem experienced by any stakeholders while relying on our results. Root Cause : What caused us to make that error that created the defect.
• One of the cause of poor Quality is 'weak' on unclear requirements. Let me explain what i mean by 'weak' or unclear. Have you ever seen requirements like : 'The system should be very user friendly' , 'The system should satisfy the end user', 'The system must be better than before', 'The system should have high security' etc. Now, look carefully and you will realize some ambiguity. Like : Very, Satisfy, better and high, to name a few. These words have different meanings to different people.
• The requirements should be Unambiguous, Clear to Test and Quantified(put a numeric objective).
• The purpose of Quantification is to force us to think deeply and debate exactly what we mean so that other later cannot fail to understand us. For Example : Choose which requirement is more clear to you.

a) System should be highly secure by year end.

b) System should be so secure by Dec 2020 that we get 95% probability of detecting a hacker with in 5 seconds.

• Every team member should be clear on the goal of our work,which is : Providing Customer with what he needs(which is not usually what he says), at the time he needs it (which may be earlier or later than he says) , to be satisfied (so that he wants to pay) and to be more successful than he was without our results (because if he is NOT successful he cannot pay and if he is not more successful WHY would he pay).
• Deming famously said : Testing does NOT improve Quality, nor guarantee Quality. Quality, good or bad, is already there in the product. Quality comes not from testing but from improvement of development process. Now, if you cannot find any defect that simply means that you cannot find any defect. That does NOT mean that customers will also not find any defects. So, if you have found a defect, ensure your team fixes the defect and it's root cause so that similar defect does not come up again.
• Always remember : The longer a defect stays in the system, the more it costs to find and repair. Why is that? Do you want me to tell that ? :))