Customer Success Story: FlowBoost from Etumos
- Conditionally bumping a person’s behavioral score based on their existing score, a score cap and other conditions
- Complex datetime and interval calculations
- Intelligent phone number formatting
- Using regular expressions to normalize data values against a set dictionary
- Determining which percentile a given lead is in for form fill-outs, content downloads, video plays, and so on
- While they already offered an in-memory database (“FBStorage”) which FlowBoost users can use to, for example, calculate percentiles across all the *current* values for a given field, they didn’t have a way to do this same calculation over time series data
In particular, a client asked if they could get percentiles using the count of a certain type of activity, by each person in their database, that occurred in the past 30, 60, and 90 days — this clearly called for a time series database (TSDB) with varying retention periods and Continuous Queries (CQs).
After a (blissfully!) brief evaluation and finding a hosting partner, Etumos was very happy to choose InfluxData’s InfluxDB as the backbone of “FBTimeSeries” — their time series alternative to FBStorage.
With InfluxDB backing their martech platform, they can do resource-efficient longitudinal modeling of lead behavior that other platforms only dream of.