Between getting really good birthday gift suggestions from your favorite sporting goods outlet and seeing promoted Twitter posts about how a favorite author’s newest book is on sale, the ecosystem of data around contemporary consumer products makes it easy to believe that sellers know everything they need to about their customers to make the most of them. For most of us, that’s only half true. Usually, what the sellers have instead is the potential to know everything they need to know. The trick is getting from that raw state to actionable business intelligence.

Say you work for a company that manufactures durable medical equipment. You have one key product. Now say that a hundred thousand people use that product half a dozen times a day. Every time someone uses the product, information about that use case is associated to a customer ID and sent back to you. If the information sent back includes only five metrics: time of use, duration of use, type of product used, volume of product used, and whether or not the use was scheduled or unscheduled, that’s 600,000 records created every day for every user.

Very quickly you’re going to amass a comprehensive and totally useless profile of your customers. Your first problem is that the profile is disconnected. While the basic, demographic information about the customer and his or her buying patterns resides in Salesforce, more detailed data about device usage doesn’t.

If you’re using a CRM solution like Salesforce, your device information is probably displayed on a custom object related to the customer object. And a new record is created in that custom object every day for every customer via a batch job that pulls from a staging table. That staging table is probably part of a custom application that lives outside of Salesforce and that was designed specifically to capture device data.

The second problem is that even if you were to bring all of that information into Salesforce, you’d end up with an overwhelming level of detail.

What you really need is something that can get that information into the hands of your sales reps, customer service reps, and marketers in a useful format.

Enter Heroku. Heroku Connect syncs data between Salesforce and a Heroku PostgreSQL database, meaning that Heroku can combine data from Salesforce and external applications. Via Heroku Connect’s External Object feature, users can gain visibility into external data sources. To create the connection select the relevant table, and using the ID field on the device records, create an indirect lookup between the device readings and the associated customer. Your medical device data is now connected to your customer data.

More importantly the data is actionable. You have metrics around product usage rates, and it’s safe to assume that you know how much product ships with each device. The obvious thing is to use that information to predict when a customer is likely to need a reorder. Because Heroku is a relational database, you can query against it, so you don’t have to wait until your data is in Salesforce, which was never meant to house that level of detail anyway, to start preforming analysis. Instead, you can create a new metric inside Heroku based on the raw data and display the results in a way that’s much more in keeping with Salesforce’s intended use as a CRM tool.

For the true data gurus, Heroku Connect also includes a tool to publish data directly to the Analytics Cloud, Salesforce’s new BI offering. Import data from your Salesforce and Heroku instances into Analytics Cloud and leverage the more robust business analysis and data visualization capabilities to make your sales rep’s lives very easy.

Or, if Salesforce isn’t the best place for your newly analyzed and connected data, Heroku provides options to display that data in a custom web application, meaning you can put the information your business users need to make the most of their customer relationships wherever those users are most likely to see it.

And you get to look that little bit more omniscient to your customers.

Interested in learning more? Join our mailing list!

AndyWhitehouseAuthor: Andy Whitehouse

Leave a Reply

Your email address will not be published. Required fields are marked *