Create a Data Dictionary for your company Most companies, even the most mature, do not have good documentation about what is the database. Creating a living document (to be updated when there are changes to the database) to define every (or at least the most commonly used) database, table and column can be incredibly useful. Explain how each row of data is created, what event in the product triggers the data to be recorded.
User Persona Study Use methods like a K-means cluster analysis to group users based on their similarities to each other. Then, once you know which cluster each user falls into, run all of the data through a Random Forest to see which features are most predictive of any one user being assigned to a particular cluster. In the end, you will have a list of features that are most important for figuring out who belongs in what cluster. In the end, this research can turn into a recommendation in the product, as well as inform targeted marketing and customer success strategies .... more on this coming soon. I have a presentation about this and I will try to go into more detail here as soon as I have time!
Product Funnel Analysis If you have some event data about how users move through the product, you can see where people fall out of the funnel ('exit pages'). Also, check out page load times (known to be of huge importance to the bottom line at most companies). Quantitative analysis of the product funnel, paired with qualitative usability research can lead to some great product feature innovations.
Marketing Channel Attribution Analysis