Source: https://joinfinn.com/integrations/snowflake

# Snowflake

## Snowflake + Finn.

Snowflake integration with Finn. Snowflake stores and processes business data for analytics, applications, and AI workloads.

Business function: Engineering and data.

## What you can do with Snowflake.

Snowflake stores and processes business data for analytics, applications, and AI workloads.

### Publish the weekly business numbers

Finn queries the agreed Snowflake datasets, explains what changed, and delivers a completed report to the business team.

**Typically handled by:** Business intelligence analyst

**Starts when:** The team's recurring reporting deadline arrives.

#### What Finn handles

1. Read the report definition, measurement rules, audience, and reporting dates.
2. Execute the required queries and collect their completed results with the source freshness information.
3. Calculate the agreed comparisons and write the explanation, identifying missing data or changes in definitions.
4. Save the finished report in connected SharePoint and distribute it through Outlook or Slack.

**The result:** The team receives its completed report with traceable figures, definitions, and reporting dates.

### Investigate a business performance question

Finn answers a specific operating question by examining the records and explaining the evidence.

**Typically handled by:** Data analyst

**Starts when:** A manager asks about a change in sales, service demand, product usage, or another recorded measure.

#### What Finn handles

1. Confirm the question, metric definition, comparison period, and permitted customer or business scope.
2. Query Snowflake for the totals and underlying groups or records.
3. Investigate the likely explanation against the data, separating observations from causes the data cannot establish.
4. Deliver the analysis with the supporting table, query references, and clearly identified unanswered questions.

**The result:** The manager receives a documented answer based on the available data.

### Complete a recurring file import

Finn takes a delivered source file through its Snowflake load and finishes the data delivery record.

**Typically handled by:** Data operations specialist

**Starts when:** A partner's or business system's recurring file reaches the designated storage stage.

#### What Finn handles

1. Read the file's reporting period, accepted format, and destination table.
2. Run the existing load process or the accepted COPY operation for the staged file.
3. Read the load results, resolve format issues with the source owner, and complete the accepted correction or resubmission.
4. Check the loaded period and counts, then publish the import result and remaining exceptions in the delivery record.

**The result:** The accepted source file is loaded and the delivery record accounts for successful and rejected data.

### Repair a failed scheduled data task

Finn investigates a failed Snowflake task and carries the accepted correction through a successful execution.

**Typically handled by:** Data engineer

**Starts when:** A scheduled task fails or the expected dataset is not refreshed.

#### What Finn handles

1. Read the task result, affected query or procedure, and the upstream input state.
2. Correct the source-input problem or submit the owning SQL or procedure change through connected GitHub review.
3. Apply the accepted production change and execute the affected task through its established run procedure.
4. Check the completed result and dataset freshness, then update the incident and notify the report owner.

**The result:** The scheduled data process has a verified repair and the dependent team knows the refresh result.

### Resolve a data reconciliation discrepancy

Finn traces a mismatch between a business record and a warehouse report, then completes the correction at its real source.

**Typically handled by:** Analytics engineer

**Starts when:** A business owner reports that a Snowflake total does not match the source system.

#### What Finn handles

1. Read the reported period, relevant source records, and the warehouse transformation logic.
2. Identify whether the mismatch comes from timing, a definition, an upstream record, or a transformation error.
3. Get the business owner's decision for record corrections and complete the accepted source or reviewed transformation change.
4. Refresh the affected processing, query the corrected comparison, and document the resolution for the report owner.

**The result:** The discrepancy has a verified explanation and any necessary correction is completed at its owner.

### Fulfill a dataset access request

Finn records the data owner's decision and completes the specific Snowflake permission change the employee needs.

**Typically handled by:** Data access administrator

**Starts when:** An employee requests a dataset for an accepted business task.

#### What Finn handles

1. Read the requested tables, purpose, employee identity, and existing role access.
2. Get the data owner's decision on the exact dataset and permission scope.
3. Apply the accepted role or object grants through Snowflake and inspect the resulting permissions.
4. Send the dataset location and usage instructions to the employee and close the connected service request with the decision and result.

**The result:** The requested access is fulfilled within the accepted scope and recorded in the service request.

### Complete an accepted warehouse usage change

Finn investigates a warehouse operating issue, applies the owner's chosen setting, and checks the effect on the intended workload.

**Typically handled by:** Data platform administrator

**Starts when:** A platform owner requests an adjustment to warehouse availability or operating settings.

#### What Finn handles

1. Read the warehouse's state, usage evidence, workload requirements, and the specific requested change.
2. Present the workload impact and get the platform owner's decision on the setting or suspension window.
3. Apply the accepted warehouse change through Snowflake and observe the affected query or scheduled workload.
4. Record the resulting setting and workload outcome, then notify the teams using the warehouse.

**The result:** The warehouse change is completed with an owner decision and a recorded operating result.

### Prepare a customer account review

Finn combines the account's warehouse facts with its customer record and completes the team's review briefing.

**Typically handled by:** Customer success operations analyst

**Starts when:** A scheduled account review is approaching.

#### What Finn handles

1. Read the customer account in connected Salesforce and the review's agreed questions.
2. Query Snowflake for the mapped account's usage, order, or service facts over the review period.
3. Write the briefing with source dates and explain observed changes without inventing customer intent or unsupported health scores.
4. Save the briefing to the customer record and create the agreed follow-up tasks for the account owner.

**The result:** The account team has a completed factual review and the follow-up work recorded against the customer.

## Connection setup

Contact Finn to discuss the records your workflow needs and the actions you want it to perform.

Your team chooses what Finn can access, what agents can change, and when a person must approve the work. The connection needs the appropriate account access and permissions from Snowflake.

## Explore Finn

- [Agents](https://joinfinn.com/agents): Help customers. Work with your team. Finish the job.
- [Systems](https://joinfinn.com/systems): Business apps for the work behind the conversation.
- [AI Gateway](https://joinfinn.com/ai-gateway): Control connected AI apps, models, access, and spending.

[All integrations](https://joinfinn.com/integrations)

[Discuss this integration](https://joinfinn.com/contact)

[Discuss your workflow](https://joinfinn.com/contact) · [How it works](https://joinfinn.com/demo/app/#/w/borough/overview) · [Get started](https://joinfinn.com/app/#/signup)