Real Time Data Controls
Configure which and how much data travels through each of your data pipes, which data to combine it with, and how to transform it as needed for your data tools.
From data pipe to data pipe, choose which data points you want to collect from which Collectors.
1. Choose data points
2. Optimize your data volume
Sampling: specify the percentage of sessions to sample. Example: I want 5% of sessions on my QoE data pipes and 100% of sessions on my ad data pipes.
Filtering: collect only events with certain metadata or remove metadata from certain events. Example: I only want to collect geographic and device metadata on the first event of the session.
Enrichment: add information from backend systems to event streams. Example: I want CPMs from my Google Ad Manager report correlated to my ad engagement events.
3. Avoid post processing
Transformations: change any event or dimension name or its value based on logical operators. Example: when Milestone event has a Milestone Percent dimension that equals ‘25’, change the event name to First Quartile.
An Excellence in Data Quality
Guaranteed under SLA
Hundreds of validations and alerts ensure that expected messages are received in the expected time, format and location.
Expected number of messages (prevent loss)
Expected message types (selected events, metadata, fluxdata)
Expected enrichment source data is updated and enriching messages
Remove duplicate messages
Failed delivery to connector is retried
Transformation rules are applied
Event message chronology
Data type validation
Ingress and egress lag does not exceed max threshold
Errors (4XX, 5XX) do not exceed max threshold
Expected infrastructure capacity for load trajectories