This was a concept we experimented with whilst I was at Redgate, where I was part of a small team housed in the "Ventures" internal incubator.
This started as a scratch your own itch problem. We found that many commercial analytics tools were geared towards collecting as much data as possible on the user, with little regard for their privacy. In these days of data leaks and privacy concerns, it felt one-sided. Additionally, collecting all this data without thinking about why you needed it led to issues later when you tried to make sense of it - could you really rely on what you were seeing, or was it a random correlation?
We made many research calls poking around the subject trying to understand if there really was a problem to be solved here. After some promising interviews, I put together this slide deck to pitch the idea to Redgate so that we might get further time to investigate:
We got the go-ahead to proceed, so I created the name, logo and the design for a product page that we could direct people to. This helped to give us an little credibility when we started approaching companies to see if we could get them to commit to using the (yet un-built) product.
The website was kept simple with a simple aim - to collect as many email addresses as we could to show a level of interest.
I went on to design the application itself after initially involving the whole team in sketching sessions and building many mockups with Balsamiq.
The main idea was that metrics were collected as part of an experiment - you needed to have a hypothesis you wanted to test and had to justify why you wanted to collect the data. This was because that reasoning would be exposed to the end user too, so if it looked like you were trying to collect some data that might be controversial, the end user would know about it.
Here's a screengrab of the application pulling back real data.
In the end, we decided against continuing with the tool - we felt like it was an idea slightly ahead of its time and maybe something to pick up later.