There has been lots of talk recently about “failing” and “failing fast”. This is actually an extremely complex topic that can be difficult to grasp.
It’s not actually about failing.
The topic of “failing” is actually all about learning. When we think about the properties of learning, we can break it down into various properties: knowledge, skills, experience and arguably the most important thing, attitude.
When you think of these as seperate entities, our attitude of learning varies for each of the other properties of knowledge, skills and experience.
When thinking about each property and how failure can relate to them, the concept of failure is completely different between them.
For example, with knowledge, we can gain knowledge through various means – a common avenue for gaining knowledge is reading books and blogs. Obviously reading a book or blog allows us to gain knowledge and learn – even if it’s learning that we disagree or can’t use what’s being said in the book or blog. But if we think about failure relating to this activity of reading, what would that look like? Not reading the book or blog? Or half reading it?… The word “failure” feels like it doesn’t quite fit right with this kind of learning.
Our attitude to failure tend to apply more in a context of gaining knowledge through practice, experimentation and decision making.
“Safe to fail” and “fail fast and often”… But what about the different types an severities of failures?
Lets hone in on this for a second. When we practice or experiment, there seems to be a push to “make it safe to fail” and to “fail fast” or “fail often” but this misses taking into account the variables in the types and scales of failure… Some of which can be very detrimental and even harmful, which we should probably avoid at all costs!
There are many different variables relating to the types of failure:
- Experimentation Failure – This is when what you planned for the experiment doesn’t work out how you wanted it to (i.e. it’s unsuccessful). This may be safe or unsafe depending on context and how big the experiment is.This experimentation failure could possibly have a knock on effect for…
- Failure For The Business – How much does it harm the org. There is definitely an acceptability scale here, and sure, there is an objective to make it “safe”, but there are a huge amount of unknowns here, and as a human race, we’re inherently pretty bad at uncovering and assessing risk.
- Failure in learning (from 1 & 2) – by this I mean failing to gain any knowledge, experience or skills after you’ve previously experienced failure (at experiment or business levels)In some ways, this type of failure is worse for us at a personal level, as it can lead to personal reputational damage if we were to make the same mistakes over and over again without learning from them.
This type of failure can be common in some environments. Especially when there are time pressures… Time might be afforded to do the experiment, but not afforded to reflect on the failures or the lessons from the experiment.
In fact, the fact there are continued time pressures is a sign that the company might not have learned from their past failures too).
Related to this is…
- Failure in thinking about the consequences – This actually relates to failure prevention as much as it relates to learning from the failure.If you have no awareness of the possible consequences of the experiment failing, then there is a higher risk of being unable to move out of the way of the bigger failures relating to the business.
Another way to describe this is if you fail to learn about a possible failure before it happens in order to act quickly to prevent it.
I’ve seen all of these kinds of failure personally and within many companies that I have worked with in the past.
One company that I used to work with was experimenting with an org change. There were some big “red flag” concerns and lots of confusion that came out part 1 of the experiment, which was at a small scale with one team. But the company actively didn’t take on-board those concerns and risks and ploughed ahead with part 2 of the experiment, which was to scale it up to other teams.
At that point, everyone could see the failure coming like a double decker bus speeding towards you from 10 miles up the road – we knew we were in danger, but we kept standing in the middle of the road, just waiting for the bus… watching it get closer. Prepping ourselves for the impact. The reason that no corrective action was taken (aka no quick learning occurred), was because the company kept on saying: “It’s fine – failure is ok. There’s no need to get out of the way of the bus… Let’s see what happens when it hit us and them we can make our next decision after that”.
The final outcome? The metaphorical bus hit hard. The experimental failure caused a business failure as lots of people ended up resenting the company and leaving. This in turn affected the product and ultimately caused reputational damage to the company too. On top of that, the company struggled to re-hire people as it had also gained a bad rep within the software development world too. Double whammy… 😦
Different severities based on different contexts.
Context also plays a big part in failing and the levels of acceptability to fail, and the severity scales of failures too.
If you are working on medical software, aeroplane software, banking software, government software, etc… then experimenting in this context should be treated completely differently compared to working on a small independent mobile app that’s for entertainment.
In certain contexts, more emphasis is certainly needed on investigating the risks and effects of failing to raise awareness
Let’s focus the conversations on learning.
Having a focus on making our failures safe doesn’t often take into account (or at least make explicit) the boundaries of acceptability or the risks surrounding failure at all levels (beyond the experiment itself), so I would suggest we shift the conversations to focus on learning, and making learning safe.
Don’t just think about the positive things you could learn within the knowledge, skills or experience you could gain through experiments and prqctice, but think about what we can learn from the possible failures too (see what I did there? I didn’t say think about the failures, but I said think about what we can learn from the possible failures). Having an awareness of the risks of failure – actually putting some focus on learning about these possible risks allows us to
This focus on learning helps us to think of the risks and the possibilities of failures. Ask yourself what would you learn from failure and then ask if and how you can learn that lesson sooner that the failure itself.
Also ask yourself what the wider impact and consequences would be from the failure (from failing an experiment, to the knock on effects regarding business failure, and even at that personal level too). And think about whether the learning opportunities justify accepting the risks and dealing with those possible consequences or whether they provide an insight in allowing us to make a course corrective, preventive decision regarding the experiment.
Only when we understand what it means to learn and the possibility of our experiments failing at this deeper level of understanding, can we then start to think about how to reduce the risks and consequences to be able to answer the questions: do we need to fail in order to learn these lessons and is there a safer way to learn quicker.