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

Community

Connector Improvement: Fix pendo data history tracking

Planned
Brendan Hawk User

Using the Pendo connector (API) the data syncing to our destination is being partially treated as slowly changing by duplicating a new row with the `_fivetran_synced` timestamp being the only way to differentiate between rows. If this is intentional, then it is only implemented and should also include some kind of column denoting which rows are active (per fivetran's own documentation, something like a `_fivetran_active` column).

Currently, this is causing a headache where the only options to differentiate between rows downstream is either an expensive window function in an inner query (see: https://fivetran.github.io/dbt_pendo/#!/model/model.pendo.int_pendo__latest_account#code) or to join my operational data to the `fivetran_metadata` table so that i can do a partial match against the connector's last sync time.

Otherwise, this should be configurable on the connector like any other connector - allow me to choose to update the rows OR duplicate them with a properly implemented type 2 slowly changing dimension as it is implemented on other connectors.

Please sign in to leave a comment.

Comments

2 comments

Date Votes
  • Official comment
    Lucas Alexander Fivetranner
    • Edited

    Hi Brendan,

    Thank you for taking the time to share your detailed feature request regarding the Pendo connector. We truly appreciate your input and are grateful for your engagement with our community.

    I've added your request to our backlog. We will prioritize this enhancement based on the overall demand and impact it has on our user community. We recognize the significance of this improvement and will consider it carefully during our planning process.

    We will keep you updated on any developments related to this feature. Your feedback is invaluable to us, and we are committed to improving your experience with Fivetran.

    Best,  
    Luke

  • Brendan Hawk User

    After further searching, it appears that joining to the metadata table doesn't actually fix this issue, since not all rows are updated on sync. So the only option to solve this at the moment is the expensive window functions on every table i'm using in a query that has an ID and the half-implemented SDC column.