How is Making Metrics Like an Organ Transplant?

Audio Version

Most businesses know the importance of using metrics to manage, but a lot of them struggle with creating metrics that actually fulfill their purpose. How do they go wrong, and how is hiring an external consultant to tell you the “right” metrics like getting an organ transplant that fails?


The point of metrics is to allow managers to manage (make day-to-day decisions to support their people in creating value for the customer) and to lead (take an organization somewhere it would not otherwise go). The big question for many companies is what exactly those metrics should be.

Some (perhaps most?) companies don’t have any formal approach to creating metrics. That allows everyone to make up their own metrics (or have none). This has the advantage of allowing local managers to come up with something that makes sense for them.

There are a number of disadvantages of this approach, however.

Self-generating metrics without a process means that there are no necessary relationships between what a manager chooses and what the business is trying to do. This has consequences. Just because a metric can be measured doesn’t mean that it should be measured.

For example, managers often choose a metric because it is “easy to measure.” An easy-to-measure metric that is unrelated to the business’ objectives is one that is wasting time and money. This brings to mind the old story of the man looking for his car keys at night under a streetlight because it was lighter there. The problem was that he had lost his keys in the dark alley. Just because a metric is easy doesn’t mean it is helpful. Metrics should be a condensed set of what is needed to align your area with what the business is trying to do today and into the future.

Managers may also choose metrics that tell them something important about their process but provides an incomplete picture. I once saw a process that only measured scrap. Metrics drive behavior, so since scrap is a bad thing, the supervisors did everything they could to drive that number down. The problem is, in the absence of other metrics like speed or efficiency, the cost of lowering the scrap rate far outweighed the benefit of all the scrap they prevented.

Even with a better set of metrics, managers can make decisions that hurt the company if they don’t understand how their metrics relate up into the organization. By maximizing the performance of their own metrics, they can hurt the organization as a whole. This is known as “suboptimization.” At a company I worked, the manager for a very large area showed millions of dollars of cost savings – enough to get promoted to even higher levels of management. The problem was that the money he was saving was by cutting back on maintenance. So for the short-term, it looked like a big win, but once he had been promoted, the consequences of that decision ended up costing the company far, far more than he claimed to have saved them.

Another version of this is when a manager focuses on improving something like productivity. Improving productivity in an area sounds great, until you realize that improving productivity in a non-bottleneck just means piling up work-in-progress ahead of the next operation and hurts the company as a whole.

Many companies take a different tack with metrics. Rather than let their managers create their own, they focus everyone in the company on a few, high level metrics. These metrics may even be good metrics for the company, but there is a major problem with this approach.

Let’s say that a company wants to focus on profit. It is hard to argue that some amount of profit is important in business, so having everyone measure profit seems like a good idea, right? In their effort to do so, they tell every manager that they are to measure profit and make decisions that contribute towards it.

Now, even leaving aside the issues I have already raised, this has the added problem that many people may not know how, or if, they affect profit. I know that the maintenance group affects my profit, but how do I relate what they do to that? I probably can’t show a profit relationship to helping prevent a possible problem, but I think that there is one there.

Finally, companies often throw up their arms and hire some big-name (and expensive) consulting firm to come in and tell them the right metrics.

The problem here is that the metrics are coming from outside of your organization. The chances that they are right for your people, processes and market is vanishingly small. Imagine somehow getting ahold of your competitor’s metrics. How would those apply to you? Some might make sense, but most would not fit with the way you do things.

Further, even if they were the right metrics, you are asking for people to change how they measure their own process. They are not going to be thrilled to have someone else come in and tell them they have been doing it all wrong, even (or especially!) if they have been.

Getting metrics from the outside (benchmarks, consultants) is like getting an organ transplant from another species. Some things might work, but it is likely to be rejected.

All these reasons are why the best approach, and the one that I use, is to have a process to create the important metrics at the top and translate these metrics down into the organization using the managers of each level to do that translation. Everyone should be able to make a pretty good guess about how what they do affects their manager’s metrics. And having them be asked and participate in answering that question gives them agency, making it far more likely they will use the metrics they were a part of creating. It is empowering to have a role in creating the metrics for which you are responsible. (You can see how that works in the video Hoshin Kanri - Making Organization-Wide Metrics.)

Being experts in their own job means that the metrics they create are far more likely to be useful. Of course, they won’t always be right, but the metrics will be theirs. And as they use them going forward, they will be the ones to recognize that a metric is bogus or that something is missing. Getting them into the habit of using metrics, and revising them when something new is learned, is a huge step towards being valuable as a manager.

And keep in mind the scalability of this approach. You don’t need one or a few individuals to figure out the whole business. Each manager has two questions to answer: “What should I measure to get my boss what they need?” and “What do my people propose to measure to get me what I need?” No matter the size of the business, these questions are constant for any individual.

The best way to generate the right metrics is to have each individual participate in creating their metrics, within a framework that insures they are proposing metrics that relate to the business objectives through a process of translation from top to bottom.

And that is how to get the right metrics everywhere without rejection.