Are you confused by all the different ways people talk about metrics? Leading vs lagging. Inputs vs outputs. Activities vs outcomes.
Take the famous five a day health advice. Basically eat five portions of fruit and vegetables a day. Is that an input or an activity? Is it a leading measure? If I eat five pieces a day have I achieved an outcome or is it an output?
All most people want to know is what does eating five a day do? How does it positively impact their life?
They want to know the relationship between what they do, how they do it and why it is important. To create a ‘line of sight'. Yet the language between metrics, whilst well intentioned, hinders rather than helps these things.
Which is why I propose a simple model to solve that problem: practice, product and performance. It is built on the same underlying concepts behind the various terms around metrics. The difference is that it explains it in a way which connects the way a team or individual works to the results they are trying to achieve.
Let's return to our earlier example, eating five a day. This is a daily practice we can all follow. It's immediately within our control. We can start right now. We can measure our current state by remembering what we ate over the last week. If we find we aren't doing so well we can take action to try and improve.
This makes practices something completely in our control. We can easily measure ourselves against them and focus on improvement.
Practices match with the terms 'inputs' or 'activity' measures.
By following a practice something happens. Just like your teacher or parent always told you ‘practice makes perfect’. But what are you making perfect?
Eating five a day produces different benefits. It provides the necessary vitamins and minerals (including folate, vitamin C and potassium) and enables you to maintain a healthy gut. The product of five a day is a healthy you. Specifically a healthier gut and a healthier level of nutrients.
Unlike five a day, a healthier you is the result of a series of actions rather than a measure of the action itself. This makes them harder to measure.
Product is not something actionable. You can't just say 'be healthier' and it magically happens. What you can do is decide to 'become healthier' and come up with ways (hypothesis) to do it. And there may be valid alternative or additional ways to improve your nutrient levels or gut microbiome you need to explore.
Product matches with the terms 'outcome' or 'leading' measures.
You're eating your five a day. You're producing good nutrient levels and a healthy gut microbiome. What's the extent of the impact of that? What did it achieve?
Hopefully you live longer. Hopefully you don't get heart disease or stroke. Hopefully you avoid certain cancers. Hopefully you've reduced these risks.
These are all measures of your performance as a human being. They are the ultimate goal. The reason why you are eating five a day.
But you won't know. Not until you reach the end of your life and they don't happen to you.
This is the challenge with performance. 'The game isn't over until the final out'. You only know how you performed after the fact. Of course, having a long life is an extreme version of this. Some performance goals may be a lot shorter. But they are only ever achieved after the fact.
This makes them an easy measure but difficult to change.
You can forecast performance though. You can work backwards from what we know influences long life and make predictions based on that. Such as maintaining nutrient levels or gut microbiome. One way of guessing this is by asking ‘do you regularly eat five a day?’.
Performance matches with the terms 'lagging' or 'Impact' measures.
Let's summarise this so far.
Eat five a day (practice) leads to >
Healthy nutrient and gut (product) leads to >
Long life (performance)
By creating the line of sight between practice, product and performance it enables you to make short term decisions to meet long term outputs. It enables you to focus on what you can do today (improve your nutrition and gut microbiome by eating five a day) to reach something which you only know you will achieve sometime in the future (long life).
Playing with the method in other domains is useful. Some simple examples:
Speed sessions (practice) >
VO2 max (product) >
Beat 5K time (performance)
Play to a click (practice) >
Sense of rhythm (product) >
Entertain audience (performance)
Test driven development (practice) >
Fast feedback (product) >
Short lead time to change (performance)
A common trap of other models is single cause fallacy where you attribute a particular impact (live long life) with a single solution (eating five a day). Essentially a one-to-one relationship.
Practice, product, performance intentionally enables you to model more complex relationships between the three. One practice can lead to multiple products. A single product can be produced by multiple practices. Likewise, performance is often the result of multiple products and one product can contribute to different performances.
As an example, test driven development (practice) can lead to fast feedback and safer changes (product). These can improve Lead time to change and change failure rate (performance).
Working backwards, lead time to change (performance) can also be improved by focusing on reduced rework or small batch sizes (product).
Once you have created a map between practice, product and performance you can use it to understand the constraints you are operating under and the relationships between them. This enables you to select areas to focus on.
For example, if you want to improve your software delivery performance, rather than starting with the four key metrics (performance) you can look at product outcomes. By understanding those outcomes you understand the levers you can take action against. You might realise that you suffer from large amounts of re-work (product) which will eventually harm your lead time to change and change failure rate (performance).
By selecting low re-work as your product outcome you can use practices to understand where to take immediate action. Perhaps you aren't automating deployments? or automating tests? or writing clean code?
By building out your map between practices, product and performance you create the necessary line of sight between action and results in a way you can confidently measure improvement.
Disclaimer: The statements and opinions expressed in this article are those of the author(s) and do not necessarily reflect the positions of Thoughtworks.