Do You Even Love Me?

Do You Even Love Me?

Have you even had a child put on their grumpy face, cross their arms and say something along the lines of “you don’t even love me”? Maybe you said no to them having more candy or insisted they need to do their homework before they can play video games or any of the millions of things that might upset a child.

The reality of course is that you do love them.  In fact, it is precisely because you love them that you are try to keep them healthy and make sure they are learning the things they need to know to be successful in life. They want something now, but you are looking at the bigger picture and you know that there is something better to be gained by giving up the immediate want.

Sometimes being customer focused can be like this.

Don’t get me wrong, I think the customer is ultimately the one who dictates the value of your software. If they don’t use it and find it valuable you aren’t really making good quality software, regardless of how well you’ve met the specifications. But sometimes customers ask for things and for you to be really customer focused you have to say no. Much like the parent who sees the bigger picture, you know things that your customers don’t.  You know the cyclomatic complexity of your code.  You know the state of your build system.  You know the risks that are there due to lack of test coverage.

Your customers don’t really care about that.  They just want the candy now, but you know what that candy will do to the health of your code. You know that you need to take that code to the gym and get a personal trainer to yell at it for a while first. Or maybe it’s even worse than that.  Maybe you need to call an ambulance and perform an emergency procedure.

Ok I’ll stop with the metaphors now, but the reality is that sometimes investing in underlying code quality issues is caring for the customer.  Sometimes slowing down a bit to clean up your processes is required to allow you move more quickly in the future.

There is a balance here of course.  The point still is to be customer focused and so you need to consider how healthy your code needs to be to meet those needs. The goal isn’t have code with a six pack, rippling abs and  a pearly white smile (sorry the metaphor is too fun – I had to come back to it).  The goal is to have code that is healthy enough to consistently deliver customer value.

As with your health, it’s a lot easier to do this if you make it a part of your regular routine.  It is a lot easier to stay healthy than it is to get healthy. Keep your code fit. Focus on the customer and make sure your code is able to help you do that. And the next time your customer asks ‘do you even love me?’ you can assure them that you do.  You are doing everything you can to have your code be around to help them out for as long as they need it. And if they’re patient, they might even get a lollipop.

Success and Failure

Success and Failure

I hate failure.  And not just in myself.  I don’t like watching others fail. It’s really hard for me to watch someone doing something ‘the wrong way’ without correcting them. As a parent I know I have to let my kids fail sometimes so that they can learn lessons I could never otherwise teach them. My son has been learning to read this year and as I sit with him and listen to him struggle with sounding out some simple words, I want to just read it for him. I want to take over and read it for him – it’s so much faster anyways. But I know that if I don’t let him struggle and fail I’ll actually be slowing down the learning process.

Work isn’t exempt from this.

How often have you jumped in to show someone what they were doing wrong or just taken over and done it for them? Or how many times have you stopped someone from making a mistake?  Sometimes it is a necessary thing to do, but have you ever short-circuited the learning process? I think that the idea of failure as something that leads to success is worth embracing.

If we leave it there though, we are in danger of missing an important point. Important lessons come from failure, but we need to set people up for success.

Now what gives on that? Have I given up on Aristolean logic? We need failure and success?


You see to learn from failure, you need to be set up for success.  Let’s return to my son learning to read. He’s been setup for success.  He’s been taught what sounds letter usually make in combination with each other. He’s been taught some basic rules of reading and he gets feedback on what he is great at, as well as feedback on how to correctly say words he has messed up.  He has my finger under the word guiding him.  He has a teacher and parents that help him with this.  So do I let him fail? Yes of course, but only in a way that I think is setting him up for success.

I don’t want to see failure for failures sake. The goal is learning not failure and so the environment needs to be conducive to learning.  Learning from failure.

This blog is about testing.  Hopefully by now you have been able to draw a few conclusions on your own that can help you in the way you work with your teammates, but let me pull out a few of my observations as well.

How can you as a tester set up others for success? One time I was riding with a friend and we were coming up to a stop sign. Based on how fast we were going, I was pretty sure he didn’t see the stop sign. I quickly looked both ways and there were no cars coming on the cross road, so I waited. Then just as he was entering the intersection, I casually said “That’s a stop sign eh” (yes I am a Canadian). The look of panic on his face as he realized what he had just done was priceless, but why tell this story here? I want us to think for a minute about the key factor in there – I looked both ways.  I wanted to let him learn a lesson about paying attention that he wouldn’t soon forget, but what if there had been a car coming down the cross road? If I had let him smash up his car would I have been setting him up for success?  No of course not! The whole point of letting him run the stop sign was so that he wouldn’t do again sometime when it would be more dangerous.

Setting someone up for success means letting them fail in safe ways.  Letting a serious bug go live so that a developer can ‘learn to write better unit tests’ is not setting someone up for success. Working with your team to help them do more testing and putting together a transition plan that moves away from you being a safety net is setting them up for success.  There will be failures along the way of course, but you need to be doing everything you can to make sure everyone has the tools they need to use those failures as learning experiences.

Some other examples of setting yourself up for success in failure include things like shortening your release cycle so that you can better respond to failure.  A bug found in the wild? Being able to quickly respond gives you success in the failure.  Or another example is having instrumentation in place that helps you understand your customer’s pain points.  You could think of those pain points as failures in your design or code, but if you have set yourself up for success you can respond to them and learn from them. These are just a few examples, and I’m sure that if you think about it, you can come up with many other ways to succeed in failure.

So don’t be afraid of failure, focus instead on setting yourself and your team up to learn and grow from failures.



During World War II, the CIA wrote a manual called the Simple Sabotage Field Manual. This manual presented ways that those in German occupied territories could sabotage the Nazi oppressors. There is a section in the document about “General Interference with Organizations and Production.”  The sabotage methods that are shared in this manual are fascinating, not least because I think we see so many instances of these happening in companies today. This is a document outlining the ideas of people who sat down and deliberately thought about how to make things less efficient.  We could do well to learn from it.

Let’s look at the first one.

Never permit short-cuts to be taken in order to expedite decisions.

Sound familiar?

“We can’t release this until we have all the test cases signed off.”

Approvals need to go through your manger who passes to her manager who passes it to the director, who make a decisions and passes it back down the line.

The developer can’t get started until the designs come in.

The tester has a huge backlog of items to get through before we can ship.

Having processes in place is often a good thing, especially as a company gets bigger, but we need to allow for flexibility in the processes.  A process can’t anticipate everything up front, so sometimes producing value requires shortcuts.

So can we rephrase that? “Always permit shortcuts to be taken in order to expedite decisions?”

No, that defeats the entire purpose of having processes in place at all.  If we always permit shortcuts, the shortcuts themselves are the process. How about this?

“Permit shortcuts to be taken in order to expedite decisions”

I put that in a quote, so you know it must be good right? We need to be ok with people short cutting the process sometimes.  If this freaks you out, perhaps your team needs to work on trust.  It is true that there are many times when it would not be helpful to take a shortcut.  For example, if you have legal auditing standards you need to meet, or if the shortcut would add a lot of risk without much benefit. But there are also time when it would be good to take a shortcut.

In order to be comfortable with people taking shortcuts you need at least two things.  You need to trust that those taking shortcuts are competent and are working towards the same goal.  

Competent doesn’t just mean good at their job.  It also means they have the information they need to make those kinds of decisions. If I shortcut a process, I need to know why it is there.  What is the purpose of it?  Will this shortcut violate the intent of it or will it help us?  To make decisions like this we need information.  Part of having a competent team is having a well informed teach

The other important factor is that we are all working towards the same goal. If you want to allow people to take shortcuts you need to know that their shortcut will lead in the right direction. Sometimes, I think we can get lazy and let process be a substitute for vision. As long as people conform to the process we know they are heading in the correct direction right?  Even if it is slower.

To allow people to go outside of the process means to know that they are aligned with what you are trying to do.  This is a harder thing to do.  It is relatively easy to force people to comply with a process.  It is much harder to keep a team of people all rowing in the same direction.

Now let’s connect this all back to testing. Are your testing processes flexible enough to allow for shortcuts? Are you a competent tester? The kind of person people can trust with doing the right thing and getting the job done well. Are you aligning yourself with the business goals of the company?  Do you know what they are?

Sometimes we can complain about how there are processes in place, but if we want to be a part of a company that allows shortcuts in those processes, we need to be the kind of testers that can be trusted with this. Up your skills.  Learn new things. Build relationships. Learn the business.

Take a shortcut.


When Should you Automate?

When Should you Automate?

When should you automate?

Well let’s think about it for a minute. What is automation good at?  Repeating the same thing over and over again.  So when should you automate? When you have something you want to repeat over and over right?

But let’s think about it a bit more.  What does automation do?  Repeats the same thing over and over… What if you want to change something?  Your automation might just lock you into something you don’t want any more.

Automation is a powerful way to add leverage. If it is done well it can allow you to get way more done in the same amount of time.  However, it can also reduce flexibility. Let’s look at an example. You’ve created a high level test that checks for the existence of a certain value in a table.  Now you want to change something in your app so that the value will be in a div instead of a table. At this point you either have to get rid of the test, re-write it completely, or not make the change all.  Your automation has made you less flexible.

So we still haven’t answered the question: When should you automate? Let’s take a stab at a couple of heuristics that can be used to help answer that question

When you know what you want

If you have confidence that things will be a certain way for a significant amount of time, it probably makes sense to do some automation.  The designs are settled on and we are fairly certain this is the way we will go for the foreseeable future. In other words what we are looking at here is an estimate of the shelf life of the automation. How confident are you that this is something we want to lock ourselves into?

When the automation is simple

Sometimes though, automation still makes sense even when we don’t know what things will look like. For example, when the automation is simple to make.  If it only takes me two minutes to automate something it doesn’t really matter to much if the design changes next week and I have to throw that test out.


In a sneaky way, what I’ve been talking about here is your return on investment. How do you know when to automate?  Well it comes down to understanding how much leverage your automation will give you and for how long, as compared to the cost of creating that automation.  If it is easy to automate something, the ROI comes more quickly and so we can create it in more uncertain environments.  If it is hard to automate something, we need to have a lot more confidence that we will be able to use that automation for the long term.

I want to take a step back from this and think about how it applies to test automation in general.  I think there is a principle here that bears some reflection.  Software creation is an inherently unstable process. We are constantly changing things in response to customer and business needs. In many cases the shelf life of test automation is going to be quite short. It won’t be long before a test fails.  If we think about this in terms of ROI doesn’t that imply that we should bias towards smaller tests?  Smaller, easier to create automated tests give a return on investment more quickly. The more difficult a test is to create and maintain the more hesitant we should be about automating it.

So when should you automate? When it gives you a good return on your investment.