Many startups generate large amounts of data as part of their normal operation. It seems obvious that all that data is valuable, but where does a resource-strapped startup begin? Here are 5 ways startups can leverage big data now (without trying to turn it into a revenue stream):
1. Improve Onboarding
The onboarding experience sets the foundation for a new user’s success and is the single biggest factor in determining whether they will become an active user and a paying customer. For many startups, the onboarding process is designed with little effort and even less research, making it an easy target for improvement using data.
Analytics services like Mixpanel make it easy to see how a user’s progress through the onboarding flow impacts their retention. A/B testing tools like Optimizely provide a point-and-click interface for testing changes to a page. By combining the two (both of which offer free starter plans), you can identify weak points in your onboarding flow, quickly and easily test improvements, and see exactly how changes affect user retention.
2. Create Recommendation Engines
Using data to generate recommendations is one of the most visible uses of big data today. The most well-known examples, from Amazon to Netflix, are driven by powerful recommendation engines that took years to develop, but they didn’t start off that way. LinkedIn’s revolutionary “People You May Know” feature, for example, began as an experiment where ads were placed in a user’s sidebar based on recommendations from a manually-run script.
Using simple heuristics combined with manual data analytics is an easy way for startups to try incorporating recommendations into their product. At DataHero, we built an advanced recommendation engine that suggested visualizations when a new dataset was imported. The first version of this feature was a hard-coded decision tree based on an analysis of our early users’ behavior. It took less than a week to create.
3. Customize the User Experience
Another surprisingly easy way for startups to leverage big data is to customize the user experience, the way social networks do with content curation (showing content based on who you interact with). Just as we can create simple recommendations using heuristics, other aspects of the user experience can be customized in the same way.
At DataHero, we saw early on that users had strong preferences when it came to what type of chart they created. Our recommendation engine rarely suggested pie charts, but we observed that certain users (particularly those in sales) would consistently change line charts into pie charts. Our solution? A few edits to our recommendation engine to create pie charts instead of line charts when the data came from Salesforce.
4. Identify Targets for Product Improvement
Tracking how users interact with your product can provide valuable insights and help you prioritize product improvements. Even if you think you don't need the data today, integrating an analytics tool like Mixpanel early on ensures that you have the data when you do, so that you can make informed decisions.
One design decision shared by almost every software company is how many items do you show in the interface (how many albums do you see in iTunes, photos in Facebook or charts in DataHero?). Most startups make this decision based on the default dimensions of their grid or the number of items developers have in their test systems, rather than looking at the statistics of actual users. Optimizing for the behavior of active users, rather than the internal team, can have a huge impact on usability for customers (hint: most users have significantly fewer items than the engineers do!).
5. Inbound Marketing
One valuable yet frequently overlooked way to leverage big data is as a source of marketing content. With the growth of inbound marketing, finding interesting insights in user data can be a quick win for startups. The key is to find topics that are both unique and appeal to your target customer (bonus points if you can piggyback on a major holiday or world event).
OkCupid was one of the first companies to successfully leverage big data for marketing with their OkTrends Blog, which covered everything from humorous TMI topics to deeper questions on race and relationships. The success of their blog drove massive growth, contributing to their acquisition by Match.com in 2011.
Want more tips? See how easy it is to use your data for inbound marketing.