For AOL Paid Services, Optimization Is the Name of the Game
For AOL Paid Services, Optimization is the Name of the Game
In marketing, as in life, perfection is unlikely. Brands can always strive for more optimization, personalization, and segmentation to deliver messages that feel like tailored experiences instead of sales pitches.
“You want to be able to customize as many pieces of the experience as possible,” Tom Wyland, program director for AOL Paid Services, said at the Direct Marketing News 2014 Marketing&Tech Partnership Summit. “You want to let the data drive you to the best experience.”
To help its customers receive offers that are relevant to them, AOL Paid Services decided to implement the CRM solution Infor Epiphany Interaction Advisor (IA) across its online, call center, and e-mail channels. The solution would leverage customer data across channels in real time to deliver targeted offers, Wyland explained. AOL Paid Services decided to test the technology by running a pilot. But if this pilot was going to be successful, the organization would have to overcome siloed data. So, the pilot turned into an all-hands-on-deck, company-wide effort.
“All organizations within the company have to work on this,” Wyland said. “You have to think big when you do an implementation.”
And before AOL could think big, it would have to start small. So the company had to define its user types, evaluate what data the organization had, determine how the different data points would work together, and consider how the different channels would align.
“If I send an email to you and you didn't open the email yet, but you called our support [team] four times,” Wyland explained, “by the time you open that email, it's going to know that you called the support center four times.”
Here's how the solution works: When a customer goes to a Web page, such as the “My Account” page, the page initiates a load. Before the page completes loading, a request for an offer is sent to the IA solution. IA then takes what it knows about the customer's current state, such as what device the customer is using, and pairs that insight with additional customer data, such as the customer's browsing preferences, history, or past purchase transactions. IA then chooses the best offer for that user and retrieves the best offer before the page finishes loading.
“Everything needs to happen in real time,” Wyland said.
To ensure that all of the different data points work in tandem, AOL started to build a more robust customer API around the same time it piloted IA. Wyland said that it was important for AOL to develop an API that the entire company could use.
Like with IA, AOL decided to roll out the development of the API in phases. First AOL had to enable users to pass data to IA. This capability only worked where they had access to data, which provided a limited scope. AOL also had to make sure that the API was “extendible.” For instance, if AOL has eight demographic elements now and receives 23 elements later, it can integrate the new data points, Wyland explained. Fortunately for AOL, once the company put IA in place, the solution was able to handle data from both places of development.
But was the strain worth the gain? Since implementing IA and the API, AOL has been able to integrate the best customer offers into its call center, website, and email, Wyland said. He also noted that the customer data API has turned into a product of its own. In addition, AOL has experienced increases in click-through rates, as well as lifts from smart targeting. In fact, Wyland said that AOL has achieved a 30 to 40% lift from targeting the right people. The development also enables AOL to test and segment simultaneously—such as by seeing what offers people click on the most—so that it can continue to optimize.
And so the never-ending test-and-learn journey continues.