Paul Godfrey is a senior developer at Zendesk. His team is responsible for building and optimizing Zendesk's marketing sites. The company is showing no signs of slowing down after their recent IPO. Heap helps drive their growth by providing actionable and comprehensive metrics on their marketing campaigns.
Problems with previous web analytics solutions
Zendesk runs a few hyper-focused marketing campaigns each year. In the beginning of campaign launches, oftentimes many variables are still in flux, so it's critical to act quickly on data. Google Analytics' and other vendors approaches to event tracking were frustrating and error-prone. "We tend to be pretty granular in what we want to track, so this ends up being a laborious process to roll out and maintain. Also, 90% of the time, after seeing some behavioral data you're going to wish you had some other data point you didn't track."
Paul and his team often found it difficult to iterate quickly on marketing campaigns. "When you're tweaking marketing campaigns you only have a few weeks to make it work. The only way to address this problem before was to add additional event tags, go through a round of QA, and deploy that change out to production. Once the changes were rolled out, you'd need to wait a few days to get a large enough sample to make a data-driven decision. The lack of flexibility in the tooling had a huge cost to our ability to iterate quickly."
They also felt frustrated with the lack of innovation happening in their previous toolset. Their prior vendor had chosen to branch out and start building tangential features rather than improving their event tracking system. Paul decided to try Heap's automatic and retroactive event tracking instead, saving hundreds of hours of developer time. Automatically tracking all client-side events also provided peace of mind that they weren't missing out on important metrics.
In addition to Heap's event tracking capabilities, Paul found a lot of value in Heap's user segmentation features. Zendesk has a lot of different types of people visiting their marketing pages. Brand-new prospects are learning about the product for the first time. People on their free trial want to explore the feature set before choosing to buy, and existing customers may be looking to upgrade their account. Users break down even further based on other dimensions: How big is their company? What industry are they in?
Paul uses Heap's advanced filtering capabilities to create user segments for every user type. Heap lets him define user segments based on the number of times a user has done a certain action in a certain time period, or based on user metadata that Heap tracks automatically, like browser and geographic location. Paul also sends additional data, such as company size or revenue, via Heap's robust APIs. By limiting a funnel to a specific user segment, he can measure how different portions of Zendesk's user base respond to their marketing campaigns.
Free Trial Users vs Direct Buyers
Optimizing a website for one particular user segment can have unintended negative effects for other types of users. That's why it's important to have detailed tracking and segmentation across the entire user base.
The majority of Zendesk's customers go through a free trial before buying. When Paul was manually instrumenting events in previous tools, it made sense to primarily focus on "trialers" and related events. But with Heap in place, Paul no longer has to pick and choose who they optimize for. In Zendesk's most recent marketing campaigns, Paul has been able to leverage Heap to ensure optimizations geared towards trialers don't have a negative impact on direct buyers.