The solution is mainly for bringing in a large amount of data. For example, let's say you have a retailer and they have various types of sales. They have stores both online and brick and mortar and they have sales happening in both places. What you're trying to do is decide all kinds of information based on the store versus online. Stores have different numbers of square feet and carry different types of merchandise depending on how they rank the store in different cities. If, for example, in Columbus, Ohio, if there are three stores, exactly the same store, they may be ranked differently based on the monetary intake that they have. Then there's the online information that they're pulling in, and data is being collected around who's ordering online and if they ordering versus going into the store, etc. All that data is pooled from the credit card information and it's cataloged. Trifacta allows you to write a code to bring that information together so that you can manipulate it at the end.

Once the information is collected, a data scientist can actually begin giving VPs in their departments the information that they need on the spot to make decisions about products. They can assess the information that the AI and machine learning is putting out and they can look at it and go, "Okay, you don't even need store C, so make store B larger, combine those two stores, give them different clothing in store B and they'll start to compete better in a market". It's amazing how much detail they can get in order to help make sales more efficient.