This year, I have been experimenting with a new way of doing things. Every year, at the start of the year, a lot of people end up making New Year's resolutions but a vast majority don't manage to keep it up and follow through for long. And these resolutions generally focus on one activity or a goal. At the start of this year, I decided to do something different. Nine and a half months later, at least the approach has been fairly successful.

And as I end week 41 today, looking back upon how the last 41 weeks have progressed, I see something that we often fail to in our products. I'll hold on the details of how I've done to the end of the year, and instead, focus on what I've learnt about metrics through this experiment.

If you work at a product company, your work is very likely driven by metrics. Most of the decisions you take, the features you ship, the designs you reject, are all influenced by the metrics they are expected to move (or not move).

When this is the case, it becomes very important how you define your metrics and what metrics you choose to move (and in what order of priority). Otherwise, all those decisions you are taking are not aligned with where you want your product to head.

Coming back to my experiment, I decided to take a far out view (five years into the future) and defined a few things I'd like to be, have and have done by that time. Then, I outlined what habits I needed to cultivate in order to get there. Then, I defined the metrics I would track on a weekly basis to help give me a picture of how I'm progressing towards these goals.

Even though I've been measuring these metrics week on week, I have actively not pursued ways to optimize on any of these metrics week on week. For example, if I failed to run on Monday and Tuesday, I wouldn't run twice the original plan on the weekend to make up for what I missed and hit the weekly target. If I did, I would be too tired the next Monday and probably miss the next Monday and Tuesday as well and turn this into a pattern, which is undesirable to me.

A lot of us make this mistake with our products though. When we take on 1-month, 3-month, 6-month or 12-month targets for our metrics, we tend to make up for what we failed to achieve in the initial weeks or months of the target period by shipping features or launching marketing campaigns that are aimed at increasing the metrics towards the end of the time period. While this might still help us hit the target we have taken, it leaves us fatigued for the start of the next cycle and turns this into a pattern. This is especially dangerous if you're leading a team as your team will imbibe this as the culture.

While it is important to hit the metric targets, it serves us well to understand the reason for having them in the first place. Metrics are milestones that define the trajectory our product will take. They are only milestones. Sometimes, it is alright if you don't hit all the milestones at expected times, as long as you hit the ultimate destination (the 5-year outlook in my experiment).

It is far more important to set the right culture than to hit every milestone, because if the culture turns to one of hitting milestones (which are usually in terms of revenues), we end up losing sight of what's best for the user. And that makes it all the more harder to hit future milestones.