Profitability within its First Year
Four of the top ten biggest Facebook Fan Pages in the world turn to PageLever to provide advanced Facebook analytics. In just one year, the company grew from a simple dashboard to an advanced tool used by international digital agencies and Fortune 500 brands.
PageLever Co-Founder David Turner knew the market for his product was big. As the sole developer for PageLever, he turned to Heroku so he could focus on building the features his clients wanted instead of bothering with systems administration.
Slow and Steady Releases Become Mission Critical Towards Success
“I love the simplicity of using Heroku,” says Turner. “With Git integration, I can push or change features in production 20 to 30 times a day.”
Heroku allows Turner to focus on building solid features one by one, instead of many features that may end up being unused by customers. Within minutes, he can push changes to his app to staging and testing environments that are identical to his production environment. Turner focuses on building every aspect of that one particular feature in his staging environment, tests it, and then pushes it to production. If the feature doesn’t work out for his customers the way he’d planned, Turner can instantly roll back to a previous version with Heroku Releases.
“By steadily releasing new features and measuring the impact they’re having with customers, we’ve learned which features to develop further and which ones to throw away instead of pouring days of work into them."
Enterprise Grade Performance Thanks to the Heroku Process Model
PageLever processes 500 million Facebook API requests each month, which are then stored in a MongoHQ database. If Turner had used a traditional web host, he would need to set up and manage a separate server in order to prevent overall application performance issues caused by this heavy processing.
The Heroku Process Model makes it much simpler to isolate these resource intensive types of processes within an application, so they operate efficiently without affecting an application’s overall performance. Turner pulls new data from the Facebook API every morning before his east coast customers would log in. This job is automated using Heroku Cron and is completely isolated from the rest of his application using Workers, which take just seconds to initially set up and scale. Turner’s customers then get the convenience of all the data PageLever provides without suffering from a lagging control panel.
I know how to manage my own servers—I just don’t have to now."
Heroku Offers Ultimate Flexibility for Developers without Sacrificing Convenience
Heroku supports many programming languages and tools, giving developers ultimate flexibility with their applications.
Turner currently utilizes two different types of databases for different tasks, which took just minutes to set up. PageLever uses MongoHQ for unstructured data it pulls from the Facebook API, and Heroku Postgres for user data and other critical records that are maintained over time.
Although Turner built PageLever in Ruby on Rails, the flexibility of the Heroku Cedar Stack allows him to scale with leaner code rather than with more hardware. He plans to open source a framework for PageLever built in Node.js. “This framework makes it easy for me to build performant, lean services for that don’t require a lot of batch processing,” says Turner. “Heroku’s flexibility means we can easily create many types of tools for our customers.”
Growth Never Inhibited by Infrastructure
Despite massively growing in 2011, the Heroku platform allowed PageLever to make sound architectural decisions, making it easy to scale to many more users later on. “2012 is going to be a year of of steady growth for us,” says Turner. “There’s a big market out there that wants to be getting more out of what they are doing on Facebook.”