Best practices are the worst

Working as an agile coach I am often asked to present best practices. What is the best way to run a standup? What is the best way to create a plan? What is the best way to run a retrospective? What is the best way to involve your stakeholders?

My usual answer is the classical it depends. This is where people give up on me and start asking what other companies of the same size and in the same line of business are doing.

Inspired by the book Team of Teams by Stanley McChristal I realized why people go looking for best practices, and perhaps how to address them.

Let us start with some important history.

Taylor

Frederick Winslow Taylor revolutionised the 1900 American manufacturing industry with his practices as he described them in The principles of scientific management. Before Taylor came along workers in the industry mainly learnt according to a master apprentice relationship, inherited by the guilds and craftsman that made up the workforce. Every worker had their own way of doing things that they had learnt from their master. Many of the procedures they performed they didn’t really know why. Taylor’s revolution came when he started studying and measuring these procedures. He documented them and researched every step, breaking it down into small components. He would then create a complete, detailed process of how to build the end product in such a way that anyone could do it. He would optimise each step, research why it was performed and experiment to find better ways of doing it. The result was simply a best practice. The best way to create the end product.

Taylor’s revolutionary way of approaching management of work increased the efficiency of plants and factories by orders of magnitude. It also did away with the tedious and slow learning of apprentices. Because of this revolution the price of products dropped significantly, giving the larger masses access to what previously was considered luxury products. Some people even credit Taylor’s methods for being an important part in the allied victory of WWII as US factories could crank out tanks and planes at an unheard rate.

However, in Taylor’s system workers were seen as dumb, replaceable cogs in a machinery. They were being burnt out doing physically hard, repetitive tasks, and quickly replaced by new workers. Workers were being specialized to perform their small part of the process as quickly as possible, without the need to understand anything about the end product.

Heritage

As Taylor’s methods were spreading across the world, from factory to factory, so was the mindset that came with it. Anything can be broken down into small steps that any idiot can perform, and there is a best way, scientifically proven, to do each of these steps. To this day, more than 100 years after Scientific management was published, this mindset holds a steady grip of many organisations. However, the environment we operate in today is very different from factories 100 years ago. The pace of change is increasing and the work we do is less and less physical labour. We have moved far away from the complicated and predictable environment that Taylor assumed, into the complex domain of unpredictable systems. However, we keep trying to apply Taylor’s methods to find a best practice, where there is none.

There is no best way to perform a retrospect, a daily meeting, planning or involving stakeholders. *It depends *(on your context). Because something worked well for one organisation or team it does not mean applying the exact same in your context will have the same effect. In contrary, you can be almost certain that you will get a different outcome.

Today almost all organisations operate in a complex environment. The only way to successfully navigate in that space is by probe-sense-respond, inspect and adapt. That usually means small experiments, evaluate the result and act on the data (by defining the next experiment). Trying to copy a procedure because someone else had success with it, without understanding the ideas behind it is doomed to fail. Spotify does not use the spotify model. They were just describing how they were organised at a specific point in time. They had a culture, built on values and principles that allowed them to keep evolving. No success has ever been reached by simply copying the procedures and organisational structure.

One of my biggest learnings is that each team and organisation is as unique as the individuals in them. Therefore, trying to copy what another team or organisation is doing will not give you the same results, even if frameworks like SAFE provides plenty of promise. Instead you must understand the underlying values and principles, the ideas behind. Only then can you try to apply a similar method, and start adjusting it to your needs.

A better practice

Before seeking best practices as inspiration, seek to understand what it is you are trying to create. Here are a few guiding questions:

• What are the outcomes you are looking for? Use at least a few layers of why.

• What would reaching this outcome be worth for you? How much are you willing to invest?

• How is the culture of your organisation support what you are trying to achieve?

I will use retrospectives as an example.

What are the outcomes?

My team wants to start doing retrospectives. *Why? *We are looking for a good way to continuously improve the environment we work in. *Why? *Because we have realised that we can work better, but we are not sure what to improve and how.

What is it worth?

If we can find time for the team to keep improving, and learning we are hoping to deliver faster, with higher quality. We are willing to invest at least a day per sprint for the whole team to spend working on these improvements.

Cultural support?

In our organisation we are encouraged to speak up when things are not working well. We are never punished for admitting failures. Everyone’s suggestion for improvements is always listened to, but the team will only go with the ones that have been proven to work in our context.

Experiment

For every outcome you are trying to achieve, define a number of experiments that you and your team are ready to take on. Try things out for a limited period of time. Create a hypothesis, and find ways to prove it. Decide what indicators you will use to tell if you are going in the right direction. Evaluate the result together with the team and decide how to proceed.

We believe that by creating a backlog of improvements, and spending time implementing these improvements it will have an effect on our lead time and quality. We decide to spend an hour at the end of every sprint finding new items for this backlog, prioritising it and then pulling at least two items from this backlog every sprint.

Steal with pride

Even though I don’t believe in copying practices I am a strong believer in stealing, and adapting. Once you understand why you want to change something there is an endless stream of inspiration. Steal what other teams and organisations are doing. Be inspired by books, podcasts, blogs and frameworks (yes, like SAFE and LESS). Go outside your own field to look at how others are solving the same problem. Just as in a Design Sprint you are encouraged to look for solutions in new domains, you might find the best inspiration somewhere you never expected. Once you start looking for it, you will find good inspiration in the strangest of places (thanks to selection bias). Pick things up and experiment with them. Keep evolving your practices as the world around you keeps changing, and make sure to be explicit about what you are doing so that others can understand, learn from it, improve it, and steal it.

Do not let the search for a best practice limit you and kill your creativity! Assume there is no best but set out to create an even better practice.