About firehose connections

 

What are firehose connections?

AmazonGoogle, and Microsoft event queueing systems provide key capabilities such as many-to-many, asynchronous messaging that decouples senders and receivers. These systems all work at scale and can handle the potentially millions of events coming their way in real time via a BlueConic firehose streaming connection. 

BlueConic captures data and events at the customer level for known and anonymous users. Handing over this information to your cloud data store in real time gives your BI team complete freedom in how to process, enrich, and improve existing reports and applications and invent new ones. You can use BlueConic firehose connections to stream BlueConic profiles and event data to other systems in real time.

BlueConic firehose connections:

Use cases

BlueConic customers use big cloud vendor firehose technologies to:

  • Enrich profiles, for example using machine learning algorithms and then sending the data back to BlueConic in new profile properties for segmentation and targeting
  • Send data to data analytics tools for customer journey analytics, multi-touch attribution, and next-best actions (NBAs) 
  • Send profiles to data lakes and data warehouses such as Google BigQuery or Amazon Redshift for modeling and analysis

Customers use these tools to further enrich their customer data, gaining insight into customer segments, conversion events, and next-best actions (NBAs).

What kinds of data can you stream?

  • Entire profiles or only selected profile properties
  • Dialogue views, clicks, and conversions
  • Events from the BlueConic dialogues 'When' tab 
  • UTM parameters of landing pages
  • Record of all dialogues a customer saw before completing a specific conversion goal

Why use a firehose connection instead of batch uploads?

In a word, timeliness. Batch uploads run on a schedule, but firehose streaming connections run in real time, as events occur. When a visitor's profile is updated or an event is triggered, the firehose connection sends a message. 

Real-time customer data

The phrase "real time" can mean different things in different contexts, so let's be specific about the timing details:

  • Changes to profile property data (and therefore the segments the profiles are part of) are sent to the stream after being "in rest" for 2 minutes, meaning that no changes occurred in the past 2 minutes. 
  • Events are sent every 10 seconds.

Technology stack

An example technology stack might look like this: 

firehose-g.png

This example shows Google cloud systems as an example, but the logos and names can easily be exchanged for their Amazon or Microsoft counterparts. The event queueing system delivers low-latency, durable messaging that allows developers to quickly integrate systems for machine learning, analytics, and other applications. BlueConic provides the data collection and uses the firehose to send data to the cloud event queuing system.

When BlueConic firehose connections connect to your event queueing system, the data can be transformed by all sorts of systems into data that is usable for presentation and action.

Learn more

Setting up and running the connections