“Bookie” Bank Report
Problem Statement
Checking data is the backbone of financial insights for small businesses, and Nav cannot deliver intelligent financing recommendations without it.
Background
The small business financing market is flooded with competitors that essentially act as lead-generation for alternative lenders. One of the main components of any SMB loan underwriting process is verification of revenue. Lenders only want to lend to those who are already making money. Many recommendation engines in online marketplaces can discern business revenue from business checking data. Therefore, SMBs on the hunt for financing are often prompted for either self-reported revenue numbers, or better yet, to connect their checking accounts in order to be “pre-qualified” for financing offers.
Nav understood the need for business checking data to make the best financing recommendations. But, we wanted to go further. In partnering with local bank data aggregator, Finicity, we saw two strategic advantages:
Use customer bank data to make accurate financing recommendations and pre-qualifications
Apply machine learning to customer bank data to deliver financial insights that equip a customer to make smarter decisions regarding their overall business health
Research & Learnings
In discovery interviews, several users reported strong interest in "business banking health" reports. Our qualitative and quantitative analysis indicated that users would be especially likely to engage with this product on a mobile device, given that their routine check-ins with their own personal bank or financial monitoring tools happens primarily on a phone or tablet.
Initial Solution and Results
“Bank Report” (Internally and affectionately named “Bookie”) was implemented on both the Nav web and mobile apps. This first iteration includes several rudimentary insights and metrics generated from a user’s business checking data.
When compared the web app version of bank report, the mobile app conversion for connecting a business checking account was 14x better, validating the assumption that the mobile app was a prime location for implementation of this new experimental feature set. Overall engagement amongst connected users spiked, as they were now using a feature set with data that actually changed daily rather than monthly (credit reports).
Iterative Learning
In future iterations, we plan to implement the following:
Machine-learning driven predictions, allowing a user to see the potential of suffering a negative bank balance in a given week or month based on historical trends.
Expenses categorization and revenue analysis based on transaction history.
Ability to connect multiple accounts for a more wholistic view of business finances.