This is a bit of a rant, and I apologize for people who are going to read this and wonder what I am getting on about. Since I try to tie everything to software testing at some point and in some way, hopefully, this will be worth your time.

I have a drug/convenience store near my place of work. I go here semi-regularly to pick up things that I need or just plain want. I’m cyclical when it comes to certain things, but one of my regular purchases is chewing gum. It helps me blunt hunger, and it helps me get into flow when I write, code or test. I also tend to pick up little things here and there because the store is less than 100 steps from my desk. As is often the case, I get certain deals. I also get asked to take surveys on their site. I do these from time to time because, hey, why not? Maybe my suggestions will help them.

Today, as I was walking out the door, I was given my receipt and the cashier posted the following on it.

Really, I get why they do this. If they can’t score a five, then it doesn’t matter. You weren’t happy, end of story. Still, I can’t help but look at this as a form of emotional blackmail. It screams “we need to have you give us a five for everything, so please score us a five!” Hey, if I can do so, I will, but what if you were just shy of a five? What if I was in a hurry, and there was just a few too many people in the store? The experience was a true four, but hey, it was still pretty great. Now that experience is going to count for nothing. Does this encourage me to give more fives? No. What it does is tell me “I no longer have any interest in giving feedback”, because unless it is something that says “Yay, we’re great!” then it’s considered worthless. It’s a way to collect kudos, and it discounts all other experiences.

As a software tester, I have often faced this attitude. We tend to be taught that bugs are absolute. If something isn’t working right, then you need to file a bug. My question always comes down to “what does ‘not working right’ actually mean?” There are situations where the way a program behaves is not “perfect”, but it’s plenty “good enough” for what I need to do. Does it delight me in every way possible? No. Would a little tweak here and there be nice? Sure. By the logic above, either everything has to be 100% flawless (good luck with that), or the experience is fundamentally broken and a bug that needs to be addressed. The problem arises when we realize that “anything less than a five is a bug”, and that means the vast majority of interactions with systems are bugs… does that make sense? Even if at a ridiculously overburdened fundamental level it is true, that means that the number of “bugs” in the system are so overwhelming that they will never get fixed. Additionally, if anything less than a five counts as zero, what faith do I have that areas I actually consider to be a one or a two, or even a zero, will actually be considered or addressed? The long term tester and cynic in me knows the answer; they won’t be looked at.

To stores out there looking for honest feedback, begging for fives isn’t going to get it. You will either get people who will post fives because they want to be nice, or they will avoid the survey entirely. Something tells me this is not the outcome you are after, if quality of experience is really what you want. Again, the cynic in me thinks this is just a way to put numbers to how many people feel you are awesome, and give little to no attention to the other 80% of responses. I hope I’m wrong.


ETA: Michael Bolton points out below that I made a faulty assumption with my closing remark. It was meant as a quip, and not to be taken literally, but he’s absolutely right. I anchored on the five, and it made me mentally consider an even distribution of the other four numbers. There absolutely is nothing that says that is the case, it’s an assumption I jumped to specifically to make a point. Thanks for the comment, Michael :).