Why bother testing?

As a tester, I don’t practice selling my craft very often. This blog post is an attempt to put together my thoughts when asked the question, “why bother testing?”

Let’s talk about bugs

When was the last time you used some buggy software? How did it make you feel? Did it cause you to swear at your computer in frustration? We are surrounded by software full of bugs, some bugs are minor but some cause us to pull out our hair in frustration. Some bugs when discovered cause nightmare headlines to spread like wild fire. Buggy low quality software is hard to sell, maintain and keep people using. Your customers won’t rave about your products to their friends if they think it’s shit.

Your customers won't rave about your products to their friends if they think it's shit

Testing doesn’t improve quality

Testing itself doesn’t improve the quality of the product, but it can help highlight issues that could impact the perception of quality. Bugs will always exist in software, it’s impossible to test for every possible scenario every time before releasing to production (especially in a world of continuous integration/deployment) but if you knew about some the bugs in you products before your customers find them then you’d be in a better place to make informed decisions about what to release and when.

Testing is active discovery

Everyone does testing on some level, most of the time we are unaware of what we are doing until we discover a frustrating situation. Testing is a skill where you practice looking for quirks in software. Testers are constantly experimenting and observing the product and are well practiced in talking through their thought processes.

There are generally 2 activities people say they are doing when they talk about testing;

1 – Verify the product works as intended

2 – Actively go hunting for bugs

Activity 1 is usually where people talk about automation, when your product can be codified as “working as intended” you might be in a place to build some automation checking to help facilitate faster feedback. However you could build all of the automation checking in the world into your products but people can still think it’s shit. You users don’t care that your unit tests are less than 0.01% flakey.

Activity 2 can be easily practiced, it’s not like anyone knows where the bugs are in software until they are discovered. You can build things into your products that make it easier for people to report the bugs they find or you could do a chaos monkey approach to find out where your product crashes unexpectedly.

There are many other activities involved with testing but I want to keep this high level.

People say testing is risk mitigation

This is not true, testers may use their own internal risk radars to help guide their testing efforts but testing itself doesn’t mitigate any risks. I like the analogy; testing is like an x-ray, an x-ray on it’s own can’t tell you how healthy you are or prevent issues from happening but it’s a tool that can give you a snapshot in time so your doctor can make informed decisions about some element of your health.

Testing can help people answer questions like, “are we comfortable shipping this code to customers?”. Bugs are a fact of life but we can’t fix what we don’t know. Testing helps us discover these potential issues.

1 comment

Leave a Reply