Metrics from the bottom up.

Zooming into your users viewpoint with micro-metrics.

Had an interesting shift in my internal analytics paradigm today. It’s interesting because I just made a list of business critical metrics for our SaaS startup. On it I had the usual suspects like churn rate, LTV, ARPU, along with a few more interesting ones such as the acquired users and revenue coming from various campaigns based on data from google analytics API.

I could see the growth rate, and we were happy and content. Until today when I was asked how many APIs on average does each of our users subscribe to. I had not thought about that. Something inside said I’m doing this wrong.

Not thinking about “today’s” numbers, but rather “Sanjay’s” numbers.

So I knew how many sign ups we had, subscriptions were canceled, and number of searches we had for each day, week, and month but only in aggregate totals. Looking at total numbers doesn’t tell you the full story. It might tell you 120 APIs were consumed today, but are those today’s new users, or people from a year ago, or one geek hyped up on 5hour energy?

Changing the way we track metrics, could also change how we act on them.

By tracking actions per user instead of actions by users you could for instance see each of your signups as a person along with the number of previous visits and how long they spent on your signup page. That would tell me wether I need a clearer value proposition or shorter forms. The same could be done with products purchased, times searched, or whatever is important to your business.

Folks were tweeting 5,000 times a day in 2007. By 2008, that number was 300,000, and by 2009 it had grown to 2.5 million per day. Tweets grew 1,400% last year to 35 million per day. Today, we are seeing 50 million tweets per day—that’s an average of 600 tweets per second. (Yes, we have TPS reports.)

The best part is that by simply adding all the individual numbers together we get the old macro level view that we have come to know and love, because let’s be real, who doesn’t like big numbers.