Metrics are a must when you are trying to achieve goals. They let you know what to focus on and how much progress you're making (or not making). You can use details to put more resources into activities that are helping you achieve your goals and determine what needs improvement. But sometimes, metrics get in the way. The wrong metrics don't help you measure anything or lead to the wrong conclusions. And sometimes you're not ready to use the metrics you have in mind. To do well with metrics, you need to set yourself up for success.
What kinds of problems can metrics have?
There are a few really common ways metrics can trip you up. Here's some common ones:
- Not ready - You aren't ready to track the metrics you want to use
- The wrong metric - You're measuring something that doesn't help you achieve company goals
- Bad analysis - You're not analyzing your metrics correctly
Not ready for metrics sounds weird, but it happens more often than you think, even in day-to-day life. Imagine you want to run sprints twice a week and measure how fast you are and how fast you become over time. You set up the most obvious metric that will help you with this - recording the time it takes you to run each sprint. But then you sprain your ankle. Imagine if you kept trying to measure your sprints. That's clearly not going to help, and you're going to make your ankle worse. This is bad for morale - every day you will feel that you are not achieving the goals you set for yourself.
Instead of measuring your speed at sprints, once you have a sprained ankle, you need to use DIFFERENT metrics. You need to set up a plan that includes healing and tracking progress to get your ankle back to normal. This is better for you (or a team) mentally, because you're doing something that needs to be done so you can be a success later when you are able to start recording sprint times.
It's easy to do something like this with a business. One example might be tracking how successful a blog is before you have time to post to it regularly. You can't really determine blog success if you don't post with a regular cadence. Another example might be measuring product sales, but you don't yet offer a product demo or trial.
Often the metrics you want to use are great metrics, you just need to be ready to use them. And it's okay to not be ready. When you aren't ready, you just work on what it takes to be ready. There is merit and progress to be made in every step it takes to reach success.
The wrong metric
In the last section I talked about how metrics can be wrong because you're not ready for them yet. I'm not going to include that kind of metric in this section. When I say 'the wrong metric' here, I mean you're just tracking something that doesn't help you vs. its not the right time for it. For example, say you create metrics for technical documentation. This is an excellent idea. But perhaps you choose a metric like 'pages of text created' or 'number of words written.'
These types of metrics are well meant, you're trying to measure output. However, if this is what you measure, people will start to try to write longer pages and more words to show higher output. What you really want for technical content is to measure impact - how many people are reading it? How often does someone say technical content was helpful? How long do people spend reading content? Do people convert to a sale after reading particular kinds of content? Is a particular language or tool most popular? These are the questions you want answers to, and metrics like pages or word count won't answer them.
Bad analysis happens when you don't have enough information to decide something or you don't understand the metrics you're using. For example, a popular metric that gets analyzed poorly is bounce rate. Sometimes people will see a high bounce rate and assume there's a problem, but bounce rate is about context. If you have a high bounce rate on a tutorial for example, but someone spent 10 minutes, it's likely they read and used the tutorial. On the other hand, if you had a landing page with a high bounce rate, now it's bad because on a landing page you want people to choose something else from your site to read. For contextual metrics like bounce rate, make sure everyone understands how they are used and how they should be analyzed. If you have disagreements about how to analyze a particular metric that requires context, come up with ways to get the data you need to answer your questions.
Set yourself up for success
You can set yourself up for success in a lot of different ways:
- Assess whether something stands in the way of you measuring with the metrics you want to use, then create a plan that helps you reach a point where you can use those metrics. Your plan should include replacement metrics that help you measure the success of the plan until you're ready to switch to your preferred metrics.
- Choose metrics that are about measuring impact. When you want to measure something, ask yourself why? What do you get out of a particular metric? How does that metric help you explain how you're achieving a company goal?
- Assess and adjust broken metrics. It's okay to measure something that's not as helpful as you thought it would be. Just learn from your choice and replace your choice with something better.
- Come up with checklists for what makes you ready to start measuring results of work you complete. What needs to be in place for you to see if your project is a success? Can it be measured from the start? The middle? Figure that out ahead of time and get buy in from key stakeholders. This can help you avoid situations where project success is measured when the project is only half complete.