What is Analytics?

"I talk to people about numbers."

When I need to explain my job in three seconds, this is what I say.

In reality, I usually get more than three seconds, but not much more, before my discussion partner’s attention starts to drift. In that short window of time, I try to describe not only what I do but also what the professional discipline of analytics actually involves.

What is analytics? How would you answer that question?

There are concepts so fundamental to my worldview that I assume everyone understands them the same way I do. “Analytics” is one of those terms. Surely, we all share the same understanding of what it means, right? Maybe not.

Recently, in a discussion with John Lovett and his data team at Seer Interactive , he asked me how I would define analytics. Here’s how I responded to John’s question.

When people say “analytics,” they either mean it in the broad sense or the narrow sense. In the broadest possible sense, analytics encompasses all the ways a business collects, manages, and uses data to drive better outcomes - making money, saving money, or keeping customers happy.

As an umbrella term, analytics spans the entire data lifecycle:

1. KPI definition: Helping the business figure out what is most important to measure.
2. Data collection: Creating instrumentation to capture meaningful data, such as customer behavior signals, or sourcing the data the business needs.
3. Data management: Aggregating and organizing data to make it useful for the business, while ensuring it is secure, compliant, and well-governed.
4. Self-service enablement: Building or buying, then implementing and maintaining, self-service tools so people throughout the company can access relevant data and use it themselves.
5. Reporting: Creating dashboards, reports, and visualizations to highlight important cuts of data.
6. Data analysis: Generating insights and recommendations that align with business objectives, often shared through presentations or documents.
7. Experimentation: Proposing ideas for experiments, contributing to their design, and analyzing the results.
8. Model-building: Developing predictive models to optimize personalization, recommendations, and other business processes.

You’ll notice I haven’t mentioned any software products by name. That’s intentional. While software is important, it is not the end-all be-all. The activities in this list should apply no matter which vendors you happen to use.

In the end, analytics isn’t just about the behind-the-scenes work of gathering and managing data. It is also about delivering visible, actionable insights and recommendations that influence decisions and drive change. The whole enchilada.

That’s how I define analytics. What about you?