How Dribbble uses Iteratively to create a single source of truth for their analytics.

With Charles Lariviere, Data Scientist

Dribbble is the go-to resource for discovering and connecting with designers and creative talent around the globe.
Social network

Dribbble is the community for designers and creatives who are looking to network, build and share their portfolio and find new work projects. At Dribbble, the data team is working closely with the product team to better understand and improve their product and user experience and they require high quality data to do so.


The data team at Dribbble was seeing challenges around their data collection strategy and tracking implementation which were impacting the quality of their data. The data team didn’t always have access to good data for their analysis in Mode and Amplitude and this impaired their reporting and analysis of new features.

Implementing tracking was often burdensome and overly complicated and new features would ship with event tracking that wasn’t implemented correctly and sometimes worse, without any tracking at all. Collaborating around new tracking was a pain and it was hard to get PMs, the data team and the implementation engineers aligned on what was needed.

"We would launch a new feature and two weeks later we’d realize the tracking wasn’t being triggered as expected. That caused us to lose confidence and trust in our data."

Data Scientist

The Dribbble team tried to manage their tracking plan using a spreadsheet and sprawling Jira tickets, but it always became a pain to maintain and they couldn’t ensure the tracking plan was being followed and the events implemented accordingly to the specs. They lacked a single source of truth for their data collection strategy that all the different stakeholders could refer back to.


As a part of a bigger initiative to rethink their analytics stack which included implementing Amplitude alongside Mode for easier product analysis and revamping data collection, improving their event tracking strategy and approach were finally prioritized.

Dribbble chose Iteratively to be their source of truth and the central tool that facilitates the creation and implementation of all of their event tracking. The data team moved the tracking plan into Iteratively which greatly improved the process and the ability to coordinate between teams.

“Going with Iteratively became a no brainer. It’s become the source of truth for our event collection and an integral part of our data stack.”

Data Scientist

While having a central tracking plan was a great improvement, the killer feature for Dribbble was the type-safe, auto-generated libraries for their developers to use when implementing new events. Now events are always implemented exactly according to the tracking plan and it’s easy to know which events have been implemented and which are still outstanding.


Dribbble is already seeing some great benefits and improvements after adopting Iteratively. The process of scoping out and implementing new events has become much easier. It has also made the process much faster: the time it takes to define a new event, scope it out, getting it implemented and seeing the data in Amplitude has been greatly reduced. And what’s more, it’s clear that Dribbble’s data quality has improved and forgotten tracking for new feature releases is a thing of the past.

“Enforcing the event naming and making sure that events are tracked exactly as scoped, we've seen great benefits from that and clear improvements in data quality.”

Data Scientist