Does Your Data Visualization have a Takeaway?
03/30/2015 by Jaime D’Agord Data Communication
All the nice data visualizations in the world are worthless without a takeaway. The takeaway may be an insight about a process, a confirmation of what you thought, or lead the viewer to a way to improve a situation.
In the case study from the paper, our imaginary customer, a small Texas-based retailer looking to make an impact in the digital world is examining their customer’s online and in-store behavior.
The analyst notices customers who use the website put items in their cart without purchasing them – yes they abandon their carts. Since the goal is to have the customer make the purchase – what should a marketer do?
One theory is that you aren’t really losing the customer, they just go to the store to make the purchase. But how do you entice them to purchase? Ahh, every marketer likes to think about that!
One item we explored was how long do the abandon cart customers spend online – we are wondering if maybe we aren’t exciting enough. Is the website able to generate sales, inform users? We categorized the session times into buckets based on time and put the users into two categories: those who purchased and those who didn’t complete the checkout.
We can verify that customers spent time looking around the site and did complete purchases. If the site were poorly designed, you might expect that they would click away fairly quickly. We didn’t think the website itself was the since customers were able to put items in their carts and many did make purchases. Our theory was confirmed.
When we focus on those who put an item in the cart but didn’t check out – we can see only 14% have a less than 1-minute session duration – maybe they were just price checking an item?
Since these are loyalty card members, we can link the loyalty card ID on the website to the in-store purchase at the cash register. This is when we discovered that the abandon cart shoppers went to the store either the same day or within 2 days and made a purchase. Ah – we now had insight that these customers who purchased in-store had been investigating online and even put an item in their cart.
In this case, notice that the TV/Electronics button is selected so we are specifically looking at items from that department. It’s easy to understand why someone might spend a lot of time reading about an appliance before purchasing. SAS Visual Analytics makes it easy to create multiple views of the data because of its interactiveness. Imagine if the marketer had put this data into Microsoft PowerPoint – it would have taken at least 7 slides to offer the same information.
Further, we can see that most of them live close to the store. So you can imagine checking on larger ticket items like a TV or refrigerator and then going to the store to make the purchase. Perhaps the customer wants to avoid a shipping charge or needs the appliance that day. Using the SAS Visual Analytics maps and the customer zip code, it is a handy way to create a map. [More SAS Visual Analytics Map Tips]
Now we can improve the situation. While there were people who did complete the purchase quickly, we also had those who waited a while. Maybe we could offer them an incentive to shop sooner? This could pull the shopper toward our store and thus improve our bottom line.
With the information in hand – we have a better idea of how to generate leads. And I suspect there are more insights to be gained in the data.