01What Is the Perplexity Computer Snowflake Connector?

The Perplexity Computer Snowflake connector lets you connect your Snowflake data warehouse to Perplexity's AI agent. This means you can ask questions about your live business data in plain English, and Perplexity will go into Snowflake, run the queries, and bring back the answer with citations.

Before this, getting data from your warehouse required knowing SQL or waiting for someone who did. Now, Perplexity removes that bottleneck. You just type a question, and the AI handles the rest.

It can pull data from your tables, apply filters, and show you exactly where the numbers come from.

This connector is available right now on Perplexity Pro, Perplexity Max, Enterprise Pro, and Enterprise Max plans. If you use Snowflake for business data, this update makes it much easier to get answers without waiting on reports.

02Key Features of the Snowflake Connector

The Snowflake connector comes with several features that make it stand out from traditional BI tools. Here is what you can do with it:

  • Plain-English queries: Ask questions like "show me month over month churn rate by pricing tier" and get the answer without writing SQL.
  • Data map generation: Perplexity learns your schema, column names, and relationships, so it knows where your data lives.
  • Citations: Every answer includes source tables, filters, and metrics, so you can verify the results.
  • Combines with web research: Pull internal data from Snowflake and compare it with industry benchmarks from the web in the same answer.
  • Slack integration: Query or schedule reports directly in Slack channels, and the answers appear in the thread.
  • Works with other connectors: Use it alongside Salesforce, HubSpot, Databricks, and more to cross-reference data.

These features let you have a real conversation with your data, instead of being stuck with pre-built dashboards.

03How to Set Up the Snowflake Connector

Setting up the Perplexity Computer Snowflake connector is straightforward. You will need admin access to your Snowflake account. Here are the steps:

  1. Log in to your Perplexity account and go to your connector settings.
  2. Look for the Snowflake integration in the list of available connectors.
  3. Click to start the setup. Perplexity will ask you to create a service account in Snowflake with the right permissions.
  4. Set up key pair authentication as described in Perplexity's help docs.
  5. Connect the service account to Perplexity by entering the necessary credentials.
  6. Once connected, trigger the data map generation. This lets Perplexity learn your schema and column names.
  7. Wait for the data map to finish building. This may take some time depending on the size of your warehouse.

After that, you are ready to start asking questions. The setup is a one-time process, and then the connector runs in the background.

04Using the Connector: Real-World Examples

Once the Snowflake connector is set up, using it is simple. You just type questions in plain English, and Perplexity does the work. In the episode, Julian showed several examples of what this looks like in practice.

  • Monthly churn rate: You can ask "show me month over month churn rate by pricing tier for the past 12 months." Perplexity will query your live Snowflake data and return the answer with source tables and filters.
  • NPS vs. industry benchmark: Pull your NPS scores from Snowflake and ask Perplexity to compare them against industry averages. It combines your internal data with real-time web research in one answer.
  • Cross-referencing pipeline data: Use the Salesforce connector to pull pipeline data, cross-reference it with warehouse data from Snowflake, and ask for a summary.
  • Scheduled Slack reports: Set up a recurring task that posts pipeline numbers to a Slack channel every Monday at 9am. Once set, it runs automatically with no manual work.

These examples show how the connector works with other tools to save time and deliver answers where your team already is.

05Results and Internal Testing Numbers

Perplexity tested the Snowflake connector (and the broader Computer for Enterprise) internally before launching it publicly. Here are the key numbers from that testing:

  • 300 employees used Computer inside their own company.
  • 4 weeks of internal testing with 16,000 queries.
  • Perplexity says Computer did what would have taken a full team 3.25 years of work.
  • The benchmarking standards were similar to those used by McKinsey, MIT, Harvard, and BCG.

Julian noted that these numbers come from Perplexity's internal benchmarks, so you might apply a "skeptic's haircut." But the direction is clear: the tool can handle a lot of queries quickly and accurately.

The data map is a big part of why it works. It learns your specific schema, so the SQL it writes matches your actual data structure. That reduces errors and gives you results you can trust.

06Tips, Limits, and Mistakes to Avoid

Getting the most out of the Snowflake connector means knowing its limits and avoiding common mistakes. Here are some tips from the episode:

  • You need admin access to Snowflake: Without it, you cannot set up the service account or key pair authentication. Make sure you have the right permissions.
  • Let the data map finish: Trigerring queries before the data map is complete may lead to errors. Wait for Perplexity to learn your schema.
  • Start with simple questions: Get comfortable asking about your data in plain English before adding complex queries.
  • Don't expect it to know undefined terms: If your data uses unusual column names, the data map helps, but you may need to clarify synonyms.
  • Combine connectors gradually: Start with Snowflake, then add Salesforce or HubSpot as you learn.
  • Use it for self-serve analytics: This is great for people who don't know SQL, but it is not a replacement for human judgment on what to measure.

Remember, the connector is only as good as the data in your warehouse. Keep your data clean and up to date for the best results.

07Who Should Use This Connector and Why It Matters

The Perplexity Computer Snowflake connector is for anyone who needs answers from business data without waiting on a data team. That includes sales managers checking pipeline health, marketing directors reviewing campaign performance, and small teams without a dedicated analyst.

Snowflake is no longer just for giant companies. Many growing businesses use it for product analytics, customer data, and sales data. If you are one of those teams, this connector gives you a data analyst "on call" for the cost of a Perplexity Pro subscription.

Why it matters: data has always been something businesses had but could not use fast enough. Reports were late, analysts were stretched, and dashboards only showed what someone decided to measure. Perplexity changes that by letting you ask any question in real time, in plain language.

This is not magic. It is about having answers when others are still waiting for the report. Teams that start using this now will move faster than the ones that do not.

08What’s Next: Personal Computer and More

Perplexity also announced a new product called Personal Computer, which is designed to run 24/7 on a dedicated Mac mini. It monitors triggers, runs tasks in the background, and carries work forward around the clock.

This means if your Snowflake data updates overnight, Personal Computer can run queries, process results, and have a report ready for you in the morning. You do not even need to be at your desk.

Personal Computer has not fully shipped yet, but it is coming. The Snowflake connector is clearly built to work with it. That opens up possibilities for automated reporting and continuous data monitoring.

The bigger picture: Perplexity is building a layer that sits on top of your data and makes it answerable in real time by anyone on your team. With more connectors and better data maps coming, the gap between teams using this and teams not using it will only grow.