In today’s world, everything moves faster and faster and as we notice everything around us becoming more of a blur, we tend to try and control as much of things as we can. Like many of the beliefs that encourage the use of the waterfall approach, we tend to feel that if we just had all the information, we would be able to make the right decisions. In our pursuit of this lofty goal, we often tend to lean on metrics and data driven decisions to fill in the needed information for us.
Have you ever watched two people in a discussion, and it seems both are quoting statistics from the same study but reaching opposing conclusions from them? From the data scientists I have known, all know and most will admit, that it is possible to make the numbers say whatever you want them to say. To me, therefore it so important to not only get a good data scientist but to get one that will give you the information you need, not the results you wanted. Another distinction I have gained from data scientists is the distinction between data and information. Data being the raw numbers, the information is what the numbers are telling you.
With all that said, we can start to uncover some fundamental issues with most approaches to using metrics. These are:
- Measure everything
- Weaponized metrics
- Metrics for control
Measure Everything
When someone first asks for some way of knowing if the changes being implemented are helping or not. There can sometimes be a push to deliver all the metrics that can be tracked at them. The truth is, you can measure and track anything. The hard part is to figure out what is worth tracking.
The problem with this approach is that it becomes difficult for them to be able to use the mass of data to be able to make informed decisions. Drowning under the wave in data being displayed, it is hard to figure out what the data points mean or what the causes of changes in them could be identifying. In their white paper “Scrum Metrics for Hyperproductive Teams: How They Fly like Fighter Aircraft”, Sutherland and Downey list ten metrics to help teams reach hyper-productivity. For some teams, even this can be too much in my opinion, they simply are not ready for them. However, these are a great starting point at the team level.
Weaponized Metrics
Metrics can be easily weaponized when they are measuring the wrong things. These metrics will encourage gaming the numbers by the teams to keep managers from micromanaging the work. The classic example of this is when managers assign a KPI to team members that the team’s velocity must increase by 10% by the end of the year. There is one inevitable conclusion to this metric. The team will assign larger point values to the stories so the metric improves. However, the result which management was likely after, the increased amount of delivery, will not be obtained. When the metrics measure the wrong things, we will encourage the wrong behaviors from our teams and then have trouble understanding why we are not seeing the results we hoped for. Worse, when the team feels the need to game the numbers in order to have good evaluations, the amount of transparency decreases; making the whole system more fragile and likely to be able to adapt as needed.
Metrics for control
This by far is the biggest culprit in my opinion. This feeds into the prior command and control mindset that those at the top should be making all decisions. It goes like this, if we present our leaders with the right information, they will be able to correct the process, people or solutions to increase our output.
Although I agree that leadership should be supporting the teams, they should not assume to do all their thinking for them. If we do not have teams of skilled and motivated professionals, then we have failed the first test for setting them up to succeed. If we do, why should we assume we know better than they do what they need and when. Metrics should be tracked at the level they are needed to inform the right people.
How should the be used
With all this said, you might be assuming that I detest metrics. That is not true. I have seen them abused often and I think they can be dangerous. But so is fire and look what that accomplished. We should stay focused on what we need to be. As Simon Sinek has said, we no longer need CEOs (Chief Executive Officers) we need CVOs (Chief Vision Officers). Taking from the lesson of David Marquet, if we want our teams to reach their full potential then stop telling them what to do. Tell them what you need done and trust that they will accomplish it. If leaders are asking if their Agile transformation is doing well, they should be looking at metrics that let them know if there are achieving the goals they have set for the organization, and trust that their people are using the most efficient process possible. Trusted teams will tell you what they need to improve, and we should help them achieve that. Only by decreasing the micromanagement of the details from above do we unleash the true potential for innovation and creativity that a team of collaborative professions can accomplish.
If you want to run a restaurant, you do not need to be the best cook around. You need to make sure you hire an outstanding chef and then stay out of the kitchen and run the business. And yet, often we do not do this same common-sense approach for our other types of work. If you have been in a management position for more than a year or two, then there is a great chance that the skills you used as an individual contributor have gotten rusty and you no longer know as much as the people that are doing the work on a daily basis. I have seen managers that have been managing people for five or ten years, telling senior developers on their team what type of architecture they should be using. This sends the very type of disempowering message that we are trying to change.
Conclusion
In most cases, I think the first step in approaching metrics is to understand what information is truly needed and what behaviors we would like to encourage. Then find the metrics that accomplish those objectives. Leave the rest in the toolbox for a time that they are needed. When the teams decide to improve something not covered by the current metrics, we can revisit the toolbox, pull out what we need, and then when they finish the improvement, put it away again until next time.
By forcing an enterprise wide array of metrics, you are telling your teams that they do not know how to improve themselves and worse that you do not trust them to do so. Not to mention that because every team is unique, you are boiling them down to their most common attributes. This can lead to the issues described above of too many metrics and metrics weaponized.
As organizations, if we hope to survive this turbulent world that we find ourselves in, we need to truly lead and throw away the manager mindset. Of course, I am not saying to throw out the people that do what we typically think of as HR people management, but rather to change how we do these roles to be people leaders and developers. After all, as I wise man once told me, “if we do not train someone to replace you, you cannot be promoted yourself.”
Please leave a comment below, I'd love to hear what you think.
Keep learning and keep moving forward!
