How Postmates Optimizes Campaigns on the Fly with Real-Time Data
Lead Product Manager
CEO and Co-Founder
When delivery service Postmates rolls out a national campaign, real-time data is not a nice-to-have—it’s a must-have. The company doesn’t wait for a campaign to be over before assessing learnings and making changes, and instead prefers to make changes while campaigns are live. Tanner McGrath, lead product manager at Postmates, and Bill Magnuson, CEO and co-founder of Appboy, spoke with eMarketer’s Maria Minsker about how Postmates chose the vendor to help achieve that level of agility and enable real-time data to flow effectively through its technology stack.
eMarketer: How important is real-time data insight for Postmates’ marketing campaigns?
Tanner McGrath: It’s very important for our organization, especially when we’re running large national campaigns, because it’s difficult to measure the effect of massive campaigns on our user base. It’s extremely important that our marketing stack is connected to our product, so that in-app behaviors can quickly be tied back to marketing communications.
For example, we built a unique in-app experience to coincide with the premiere of “The Bachelorette” on ABC, and we wanted to test how that experience was received from the moment we shared it in our marketing efforts. We rolled the campaign out exponentially, meaning we tested it on a smaller group initially and optimized it for larger groups as time went on. We like to improve engagement and alleviate pain points in real-time, but it was difficult without the right technology.
“We like to improve engagement and alleviate pain points in real-time, but it was difficult without the right technology.”
eMarketer: How did you go about selecting a vendor that would provide the real-time insight you needed?
McGrath: We tested a number of vendors and settled on the Currents [real-time data management] tool from Appboy. We liked the scale and speed it provided, plus it instantly hooked up to our analytics stack, which was critical.
eMarketer: How does the Currents tool work? How does it provide actionable data insight on campaigns in real time?
Bill Magnuson: Currents is a tech stack connector that allows us to stream data insights into other parts of our customers’ stacks, [such as their email marketing tools.] Our product strategy has always been focused on CRM [customer relationship management], but with Currents, we now have an ability to export all of the data that’s flowing through a tech stack in real time. Currents creates an entirely real-time data stream, so when an event [such as the opening of an email] becomes known to us, we immediately send it over to Postmates.
eMarketer: Why should brands consider using a tech stack connector?
Magnuson: Typically, when CRM vendors offer flexible APIs [application programming interfaces], data can theoretically flow freely and inform the rest of the stack, but when users hit the realities of coordinating with an engineering team to connect a data stack, what should be possible in theory breaks down in practice. That’s why our goal was to build a connector that not only delivered insight in real time, but was also turnkey and easy to use.
“It wasn’t just about seeing which email had the highest open rate, but also about determining which copy enabled a customer to take a specific action.”
eMarketer: Can you share some examples of changes you might make to a campaign based on real-time insight?
McGrath: With the “Bachelorette” campaign, we were able to change in-app messages and emails related to the in-app experience built for that promotion. It wasn’t just about seeing which email had the highest open rate, but also about determining which copy enabled a customer to take a specific action, such as sharing something on social media or actually taking advantage of free delivery offers.
eMarketer: What are some other ways that Postmates plans to leverage real-time data in the future?
McGrath: One goal we have for our data science team is to apply machine learning to real-time data to inform future campaigns. We want to enable our data scientists to leverage the data available and create better campaign recommendation engines. We also want to have better targeting, and more feedback for the marketing team so that they can synthesize learnings quicker than has historically been possible.