The components of behavioral product analytics

This is the first of a series of 4 articles that will teach you the 3 components of behavioral product analytics. In this first article, we’ll introduce you to the two types of data you collect when analyzing the performance of a product and how they differ from each other. You’ll also learn what are the components of data. So let’s start by learning the types of data.

Types of data

When it comes to collecting data, it’s necessary to understand exactly what that means. Data is one of those generic words that’s easy to throw around because it can mean so many things. Data is information; it’s a description; it’s the 1s and 0s created by 1s and 0s, but also the original 1s and 0s.

In practice, and certainly in the context of product analytics, what we mean by data relates to the information that describes an interaction. There are two main types of data.

Static data

Static data is information that describes a participant in an interaction. This type of data doesn’t change; its meaning is constant and permanent or at the very least, it is rarely modified. When we describe a person, for example, we can use information such as a date of birth, the person’s gender, an email address, and even information like shoe size or number of toes if they’re considered necessary relative to our product.

Dynamic data

The second type of data is dynamic data. As you might have guessed, contrary to static data, dynamic data represents change. In fact, dynamic data describes the interaction itself. It is in constant flux and is generally based on behavior. You can think of this type of data as an action such as submitting a form, uploading a file or sending an email.

The reason why we generally view data in this way is to understand the relationship between a person engaging in an interaction and the interaction itself. In other words, our goal is to uncover relationships between people and the actions they undertake. Because static data is descriptive of an entity or a person, it can be compared to dynamic data, the description of an action or a set of actions.

Components of data

We’ve started to reference the concepts of people and interactions. These are the two main components of data capture insofar as they allow us to define the context for collecting information.

Let’s go ahead and break it down even more. As we previously noted, data is by definition a universal concept. As a result, it can be helpful to use the simplest of words to illustrate the most basic level of analysis we can obtain from data.

Things happen.

This statement is irrefutable. Unless you believe the world around you doesn’t change and that time is frozen, you must reasonably admit that, yes, things do in fact happen. Then what do these two words reveal more than you already know? Probably not much, until we extrapolate from them into the world of data collection.

Time is in fact not frozen. The clock ticks forward, so anything that does happen, occurs in time. One of the first descriptions of things happening is thus when they happen. When we talk about data, we must inherently refer to the construct of time.

Secondly, over time, things are happening. Things in this case refer to occurrences, events, interactions — all the aspects that make the world around us look, feel, smell, taste or otherwise appear differently.

Things can’t happen in isolation. Things are generally influenced by factors outside of their own existence. As such, things don’t happen in a bubble, but rather are initiated, affected or associated with their environments. In product analytics, the environment is active; that is people use products and make things happen. And sometimes (increasingly so), it’s not people that make things happen, but machines or programmed processes themselves that are responsible for making things happen.

If we translate the concepts that emerge from our shallow philosophical reflection, we can categorize them as the 3 components of behavioral analytics for products.

1-) The most basic concept is called an event. It is the the thing that is happening.

2-) The second concept is the user. It is the entity that influences or causes the thing that happens.

3-) And lastly, there is the session. A session is the time during which things happen.

Our end result looks something like this: users create events that happen within sessions.

To make things more tangible, let’s put this in the context of a real-world scenario, a football match. If something happens such as a player passing the ball, the user would be the player; the event would be that the ball was passed, and the session could be the 90-minute period during which the match takes place.

users create events that happen within sessions

User: Maradona. Event: scored a goal (USING HIS HAND!) Session: the 90-minute period of the match England vs. Argentina in 1986 World Cup Quarter Final.

In the example above, the match England vs. Argentina in the 1986 World Cup Quarter Final (the session), how do you think we would describe the fact that Maradona (the user), scored a goal (the event) using his hand? In this article, we’ll explain how to define an event and describe its properties.

Editor’s note: This is the first of a series of 4 articles that will teach you the 3 components of behavioral analytics for products.

To get notified when our next article is out, please sign up for our newsletter.


By | 2017-12-14T08:26:36+00:00 December 14th, 2017|product analytics, The 3 components of behavioral analytics for products|

About the Author:

Patrick is a rare MBA graduate with a background in economics and finance that turned away from traditional business roles to become a UX designer. This powerful combination has led him to focus heavily on data analytics applied to user experience design, product innovation and workflow optimization. Through metriq, Patrick helps his clients leverage the power of design methodologies, data collection and machine learning to enhance products and processes using recommendation engines, prediction models and automation. Patrick also currently serves as entrepreneur-in-residence at Five by Five. He was one of NUMA’s first entrepreneurs-in-residence and the 1st hire at Infinit (recently acquired by Docker).