article: tracking net (and software) usage

In an ecommercetimes.com article from this morning titled "How to Count Clicks on the Net", reporter Sarah Lacy describes a project called "the Nomenclature Project" being conducted by the Interactive Advertising Bureau (IAB), a "trade group that sets guidelines and standards for online advertising".

The goal of this project is to establish standards for determining exactly how many people visit a single website.

A quick leap over to the IAM's homepage reveals:

IAB General Membership is open to companies that sell any form of interactive advertising, including online, email, wireless, interactive television or other emerging platforms.

IAB Associate Members are those companies whose business supports the sale of interactive advertising and marketing. Associate Members enjoy similar benefits to IAB General Members, including involvement in committees, research and events.

Makes sense to me. IAB's members are the people who are spending money. They want/need some standards around how to define the key metric used to set pricing: discrete clicks.

To its credit, the Nomenclature Project seeks to establish "who is looking at what", which is different than just counting hits.

This is, obviously, a critical piece of intelligence, complicated by the wide variety of ways in which content is now being distributed.

I'm asking myself: if this is possible for web content, why can't I get the same information about my software? Replace the word "content" with "software" in these articles, and you get an interesting perspective.

Who wouldn't love to get detailed metrics on what elements of your software product are actually being used? How about details on what versions are active in the field, and how often they are used? Wouldn't it be cool to get details on what parts of your help system are hit most?

Once you had this data, think of all the good you could do with it. You could actually focus on the features your customers value, the use cases that reflect their real-world needs, and focus less on the "bloatware" features that look good in the requirements but which don't translate into real customer value. You could document the adoption rates of new products, and help fine-tune end of life planning based on actual utilization numbers.

And perhaps most importantly to the product marketing guy, you could correlate outbound marketing to actual product usage in order to see if your messaging works (or doesn't).

Tracking usage using a consistent set of definitions, and reporting on them in a consistent way, is the first step. I'm hopeful that the Nomenclature Project can get it right.