Why Us
Build Conversation Apps on Rippit over Snowflake
Crocodiles are easy. They try to kill and eat you. People are harder.
Theme 1
Conversation Processing Abilities
We use Snowflake as a proxy for data warehouses, but Snowflake is optimized for numbers and structured data that can be analyzed with SQL. Conversation data is unstructured and requires AI to be analyzed.
When you run Claude on top of a data warehouse like Snowflake versus Rippit to query conversation data, you will be making material quality, performance, and cost tradeoffs.
Below are 2 research reports on this topic.
The first is analyzing Conversation Data in Snowflake leveraging Snowflake’s native Cortex for transcript processing.
The second is analyzing Conversation Data in Snowflake but replacing Cortex with Claude for Transcript processing.
Snowflake is an amazing solution for traditional business intelligence. However, the way DataDog is optimized for observability - Rippit is optimized for conversation data.
The best way to see this for yourself is to do a side-by-side comparison where you ask the same set of questions on the same conversation data and compare the experience side-by-side.
Reach out if you want help designing the experiment.
Theme 2
Actionability
The user experience to access this data via Claude and other AI Agents is the same for any single individual. Things break down with use cases that involve anything real-time, coordination across large groups of people, or a continuous connection with another AI agent or software.
For example, there are use cases that require employees having access to a specific slice of conversation data accessible to them and constantly updating.
There are use cases that require continuous monitoring of conversation data for incident detection to continuous data enrichment on 100% of conversations.
These are both examples of actionability that will be difficult to replicate on Snowflake.
Our recommendation
It’s not Rippit or Snowflake (or any internal data lake). It’s a better-together story.
Rippit ingests structured data from Snowflake often to enrich conversation data. Rippit pushes structured data derived from conversation data back into Snowflake.
Where conversations become
insights
actionable data
business intelligence
enterprise visibility
insights
I fear not the man who has practiced 10,000 kicks once, but I fear the man who has practiced one kick 10,000 times
