Descriptive analytics

Descriptive analytics

Quick definition: Descriptive analytics is the practice of gaining insights from existing historical data.

Key takeaways:

The following information was provided during an interview with Rohit Gossain, product manager for Adobe Analytics with a focus on Customer Journey Analytics.

What is descriptive analytics?
Why is descriptive analytics important?
How does descriptive analytics differ from predictive and prescriptive analytics?
What are the different categories of descriptive analytics?
What tools are necessary for descriptive analytics?
How does a company ensure they are using descriptive analytics strategically?

What is descriptive analytics?

Descriptive analytics is the utilization of data to identify patterns, relationships, and trends. Through data analytics, data describes what is happening at a given moment or during a period of time.

Why is descriptive analytics important

Descriptive analytics is the foundation of all business analytics. To achieve a goal, like improving your business metrics, and before you can give any kind of prescription or recommendation, you need to learn about the health of your business. Descriptive analytics is vital to understanding the health of your business.

Once you have the descriptive analytics in place and you know what's really happening, then you can go on to talk about optimization and improvements. After that, you can add the predictive analytics component, which allows you to forecast the future of your business. Predictive analytics is highly dependent on the data quality of your descriptive analytics. If your descriptive analytics aren’t accurate, then they will not work at all. And to get high-quality descriptive analytics, you need to have good data, methodology, and KPI definitions.

How does descriptive analytics differ from predictive and prescriptive analytics?

Here is how the breakout of the three analytics types looks:

What are the different categories of descriptive analytics?

The two main categories of descriptive analytics are operational intelligence and business intelligence.

What tools are necessary for descriptive analytics?

Descriptive analytics requires a layered stack of different solutions working together. To start with, you need to collect the data. You might get data from surveys, data-capturing solutions like Google Analytics or Adobe Analytics, or third-party vendors like Nielsen and ComScore. Once the data has been collected, it needs to be stored in a database, data warehouse, or data lake. At that point, you need a solution that processes the data, like Hadoop.

After performing analytics and processing the data, the final step is to present the data in a format that can be easily understood. Many companies still use Microsoft Excel to report on the data, but there are other data visualization solutions available, like Tableau or the visualization feature of Adobe Analytics.

How does a company ensure they are using descriptive analytics strategically?

It’s important to make sure you are looking beyond vanity metrics, which are numbers or factors that don’t have an impact on the success or health of your business. A metric like visits to the website can provide information about how much traffic is coming to your website, but it doesn’t tell you what to do next or indicate how many of those visits resulted in conversions.

When looking at descriptive analytics, pay attention to metrics that reflect how well your business is performing. Conversion rate, for example, provides information about how many visitors went through the entire customer journey and completed the goal action, whether that’s making a purchase, becoming a subscriber, or something else. By analyzing the data surrounding conversion rate, you can see how many customers convert over time and evaluate how well your marketing strategies are resonating with customers.

Paying undue attention to vanity metrics can waste an organization’s time, and not provide any helpful information that can inform next steps. With descriptive analytics, make sure you are analyzing data that you can use and build on.

Companies also need to focus on data quality and data governance. To maintain high-quality data, you should both invest in the right testing tools for your organization and encourage communication between your analytics, products, business functions, and engineering teams. Each owns a different part of the descriptive analytics process. If there is poor communication, the risk of mistakes and bad data quality increases. But by paying attention to data governance, which deals with the organization and restrictions of data, the data is easy to access and use properly.

Another important factor in using descriptive analytics strategically is being skeptical about the data. Don't just absorb what is being given to you and take it as fact. Become more inquisitive and ask more questions about the quality and numbers.

People also view