When it Comes to Data Collection, Avoid MVP and Choose MDC Instead

20 years ago the term “minimum viable product” was coined and over the following decades the concept has taken on an almost religious following among digital product developers. Despite its many benefits, the MVP philosophy continually trips up organizations in one key area. If you don’t carve out an MVP exception for your data collection, you’re likely to find yourself trying to dig your way out of a hole pretty quickly.

Before going into the details of this problem, it’s good to look at what MVP is and why people use it. The core concept is that rather than try to build the “perfect” product with all of the features, bells and whistles you think it SHOULD have, you are better off identifying the minimum set of functionality that you need to present the product to the market for validation. Building a lighter version of the product and getting it to market faster (and cheaper) lets you validate the core assumptions behind your product plan and gather valuable data to figure out how to prioritize all of the other features you want to build. 

It’s that last part, “gather valuable data”, where countless companies trip themselves up with the MVP approach. When they apply the philosophy to ALL of the parts of the product development, they often include the data collection component as well. They end up capturing the primary KPIs that they BELIEVE will be important, but often fail to capture additional events, signals and attributes about the interactions with the product.

These gaps can severely limit the team’s ability to fully analyze user behavior and product performance. While product development is fundamentally about growing into the future, data collection is fundamentally about recording the past. You can always add new features and even create new reporting views and analyses, but you can never go back in time and add data that you didn’t collect in the first place. 

Ironically, successfully applying MVP to the data collection function makes the actual MVP process for the product less successful!

Here’s a real-world example: A media company decided to launch a new streaming service on a very accelerated timeline. They had no choice but to apply the MVP philosophy to the development process, and applied it to data collection as well. Three weeks after launch, as the leadership team was trying to evaluate the success of the service’s content presentation strategy, they asked a simple and critical question: “How important is the location of a content tile on the homepage (row and column) in terms of driving clicks and video viewing?”

After multiple attempts to answer this question, the ultimate response was “Unfortunately, we don’t know. We aren’t capturing position on the page with the click data.” This important attribute hadn’t made it into the MVP for data collection, leaving the team without key information to guide the next stage of product evolution. 

The solution to this problem is simple. When it comes to data collection, rather than MVP (minimum viable product), go for MDC (maximum data collection). By attempting to capture as much (and as detailed) information as possible right from the first release, you’ll minimize those “I really wish we had captured that data” moments and you will be much better prepared to drive the product evolution in the right direction. MDC will ultimately make the rest of your MVP process more efficient and more successful.