Skip to main content
Docs
Feature Requests Onboarding Trust Center Status Page

Community

Connector Improvement: AWS Cost Explorer improvement

Answered
Edgardo Mejia Martinez User

Hey Erin, Fivetran Team! How are you!

Thanks for allowing us to use it in preview mode.

I’ve been analyzing the tables and I have a couple of doubts/comments:

  • We pivot groups between 4-5 options:  Service, Region, Linked Account, Charge Type, and Billing Entity.  How can it be possible to extract data with more granularity? We solve the issue by manually filtering by one dimension and exporting data using another dimension.  We have been also analyzing the API which could allow groups by two dimensions.
  • We extract from 5 different accounts, are you planning on adding some sort of consolidation?  
  • We filter out and group by data using Tags, is it possible to know a little bit more of what’s coming for that Dimension?  

Different than the initial instructions, I was able to configure the connector without granting admin privileges to a role, this is very cool since getting an admin role could be hard. Daily granularity is great!

We’re eager to see the evolution of this connector, if you guys and gals want to jump on a meeting to know more about what we do, I’m available.

 

Best Regards,

 

Edgardo

Please sign in to leave a comment.

Comments

2 comments

Date Votes
  • Official comment
    Frank Fivetranner
    • Edited

    Hi Edgardo, 

    I'm the new product manager for this connector so apologies for the late response. We are actively investigating options to improve our connector so I unfortunately do not have any answers at this time, but feedback like this is very helpful.

    I appreciate your offer to connect and if we have any future need for customer input we will reach out. 

    Thanks, 

    Frank

  • Joshua Trampier User
    • Edited

    As I stated in a comment on the GCP Billing connector, it's vital for companies operating a SaaS in the cloud to have a Fivetran-aided, frictionless way to quantify per-client costs.

    The current beta connector as of Feb. 26, 2025 has a fair amount of friction. The Fivetran instructions require its customer to 

    1. Create a standard report following the AWS instructions
    2. Under Data table configurations, select columns and granularity
    3. Under File versioning, select the Overwrite existing data export file delivery option.

    These steps can and should be automated. Like its other connectors, I would rather Fivetran tell the customer what read permissions to add to an IAM service role and which AWS account's billing API to query. This would make the experience much more straightforward and not require using a bucket as a drop-zone. If I am already doing that much, I can write a simple script for my data warehouse or lakehouse to pick up daily report drops to a bucket.

    In short, the current setup for this connector is a relatively high level of effort compared to other connectors, which are plug-and-play once credentials and permissions are supplied.