Vantage Launches AI-Powered Custom Providers: Upload Costs in Any Format

by Vantage Team


Vantage Launches AI-Powered Custom Providers: Upload Costs in Any Format

Today, Vantage is enhancing its Custom Provider integration to ingest billing CSVs in their existing format without the need to reformat to the FinOps FOCUS format. When customers upload costs in a format other than FOCUS, Vantage will use AI to automatically map the data to the FOCUS format and ingest for use throughout the console.

Vantage can now ingest CSVs from providers that do not support FOCUS exports, and will convert the data to FOCUS format automatically.

Previously, customers could add costs from providers that don’t expose billing APIs, aren’t yet supported by Vantage, or originate from custom billing systems through Custom Providers using the open-source FinOps FOCUS standard. However, adding providers that don’t support FOCUS exports required manually reformatting each file to match the format. This process was error-prone, particularly for non-technical teams or organizations managing multiple vendor formats.

Now, with this enhancement to Custom Providers, you can upload cost files in their original format, and Vantage extracts the relevant columns, maps them to the FOCUS format, and brings them into your Vantage dataset for cost reporting, virtual tagging, or cost allocation. Like before, you can navigate to the Custom Provider integration within the Settings screen, and upload your file. If it is not already in the FOCUS format, Vantage will inspect the headers, and let you know if the file was successfully mapped. If the file is valid, you’ll see a preview of how it will be mapped before it’s ingested, so you can confirm everything looks correct. This simplifies cost reporting for vendors that don’t offer billing APIs but do provide CSV invoices.

This enhancement to Custom Providers is now available to all Vantage customers. To get started, navigate to the Integrations screen in the console and create a new Custom Provider integration. To learn more about how to integrate a Custom Provider, see the documentation.

Frequently Asked Questions

1. What is being launched today?

Vantage is launching an enhancement to Custom Providers that uses AI to automatically map CSV columns to FOCUS columns, eliminating the need for customers to do manual data transformations prior to uploading their costs.

2. Who is the customer?

Any Vantage customer who uses Custom Providers to bring in cost data from unsupported vendors or platforms.

3. How much does this cost?

There is no additional cost for Custom Providers support; however, costs from Custom Providers will be included in quota tier enforcement. In the event that your custom costs push you over your current tier limit, you may be prompted to upgrade. To see more details on pricing, please refer to the Pricing page.

4. How does the new Custom Providers formatting work?

When a user uploads a file to the Custom Providers interface or the /integrations endpoint, Vantage uses AI to detect the structure of the document, identify columns for services, usage, cost, and other FOCUS metadata columns, and map these to the FOCUS format. Columns that do not match the FOCUS format will be discarded. Once this is validated, customers can preview the mapped data, assign the integration a name and a description, and upload the data into the Vantage console.

5. What is the minimum amount of information that the uploaded CSV has to contain?

Vantage requires the equivalent of the following FOCUS columns to create a cost entry:

  • BilledCost
  • ServiceName
  • ChargePeriodStart
  • ChargeCategory

6. Can I preview the transformed data to ensure it has been mapped correctly?

Yes, once a success message is shown that the CSV can be mapped, users can click the “Review” link to generate a preview of the data that will be ingested. If this mapping looks incorrect, you can rename the column headers in your CSV to match the FOCUS format and re-upload the file

7. What happens if Vantage cannot parse the CSV?

If the CSV cannot be mapped, an error will be returned. It is recommended to perform your own manual mapping to the FinOps FOCUS specification if this is the case. For more information about errors, see our troubleshooting documentation.

8. What file types are supported?

Costs can be uploaded via a CSV within the console or via the API.

9. How do I create a Custom Provider in the API?

  • Send a POST request to the /integrations/custom_providers endpoint to create the Custom Provider Integration
  • A Custom Provider integration token is returned in a successful 200 response (e.g., accss_crdntl_1a234bc5def678) auto_transform parameter set to true
  • Use the token to upload a CSV file via the /integrations/{integration_token}/costs.csv endpoint.

See the documentation for a detailed walkthrough of how to use these API endpoints.

10. How often can I upload Custom Provider data?

Custom Providers can be uploaded at any time. Vantage will use AI to detect the equivalent of a ChargePeriodStart column from the FOCUS Format, so Vantage can place the data in the correct relationship to other costs under a given Custom Provider. Note, however, if you upload a file with the same name as an existing import, your new file will overwrite the original file. Ensure you have different names for each file upload (e.g., cloudflare_costs_8_2024cloudflare_costs_9_2024). You can view the list of imported filenames on the Manage tab for each Custom Provider integration.

11. What permission level is needed to upload data?

Users must have Owner or Integration Owner permissions in Vantage.

12. Are there any limits to how much data I can upload for a Custom Provider?

Currently, the largest file you can upload is 10,000 row / 2MB file size limit. If you need higher upload limits, contact support@vantage.sh

13. What date granularity is supported?

Vantage aggregates uploaded custom costs to a daily level.

14. Can I use the Vantage Terraform provider?

No, there is no Terraform support for uploading custom providers.

15. What AI is being used to process my data?

We are using OpenAI to process this data.