“No one knows what to do”
Those who attend my talks know I ask folks to fill out a form with several questions on it. This morning I looked through responses to these. My original intent was to find something interesting for you. I thought, if I break down the data, is there something interesting I can share? Is there something we can learn to will help us with test automation?
To the question, “What is your biggest challenge with test automation?” one response caught my attention. After tweeting about it and seeing the strong response, I decided to post on it. I still intend to post the break down for this question soon, so stay tuned.
Here’s the tweet:
Reviewing surveys. Q: “What is your biggest challenge with test automation?” Favorite answer: “No one knows what to do.” 😟 Will post soon.
— Paul Merrill (@dpaulmerrill) May 4, 2017
Ever heard that answer?
I had a visceral reaction. Immediately, I was sucked back into memories of times I’ve heard it before. Frankly, the statement grates on me. I’m not the type that says something like this. When I’ve heard it said, I’ve noticed the people saying it are VERY different from me. If I’m lucky, when I have a reaction like this one, a little flag goes up in my head. A voice says “Paul, you’ll need to slow down and put in extra effort here.”
My first instinct is to assume that this answer was written by people like the ones I’ve heard say it. It’s easy for me to assume they said it with the same intonations, the same gestures, the same conviction of helplessness I’ve heard before. My instinct is to assume they wrote it for one of two underlying factors:
1) Leadership is missing and no one knows what to do, OR
2) This person doesn’t know what to do and is projecting that feeling on to everyone else.
If I were to approach the person who wrote this assuming either of the assessments above were correct what would be the outcome?
For #1, I could easily end up undermining leadership. I could degrade any potential for a relationship with a leader before it begins! Of course, I may end up endearing myself to the respondent – but what would we gain from that? A temporary alignment based on a sentiment that may or may not be justified?
For #2 I could easily position myself as “knowing what to do” and powering-up. The all-knowing-all-automating king of automation is here to help (that’s satire, by the way). But who wants to be around a “know it all” that’s already discovered all the answers?
When consulting, coaching, teaming up, or training, I have found that is is generally better to assess the situation while listening carefully to my gut. If my goal really is to make a positive impact on the people around me through test automation, how would I approach this respondent? Could there be something I’m missing?
What if test automation is just a minor part of the team’s everyday world? What if it’s something they want to do, but just don’t know how to do it? What if the leader has intentionally put the individual or the team in a position to have to figure out test automation on their own with little to no guidance? What if the entire organization is made up of people with fewer than 5 years of experience in software development? What if the organization is adding headcount so quickly, that new folks haven’t had time to catch up and old folks don’t have time to help them (making it look like “no one knows what’s going on”)?
So many “what if”s…
Getting the Context
It’s likely I’ll never have all the context around this answer. In general, I think we rarely (if ever) have all the context around anything. We function everyday without all the context. We use assumptions and hard-coded responses to allow us to act without all the context. Assumptions are necessary and helpful. They are useful things. Imagine waking up every morning and having to study gravity in order to determine whether stepping out of bed would be safe!
On the other hand, there are times when recognizing we’re acting on preconceptions, biases, and overly-poignant memories can be helpful too.
A balance is necessary. For me, for the respondent… The balance between use of assumptions and gathering context differs for each of us over time depending on situations and constraints.
I can’t fit all of this into 140 characters on twitter. I’m not sure I did my thoughts justice in this post. But I’ll tell you one thing, it’s still my favorite answer.
So what is your biggest challenge you face with test automation?