DOMO API - Upsert a Datasert

RGranada Contributor

We are on a process of starting to use DOMO's API to upload data to DOMO. Is there any way to Upsert a dataset using the API's?

Ricardo Granada 


**If the post solves your problem, mark it by clicking on "Accept as Solution"
**You can say "Thank you" by clicking the thumbs up in the post that helped you.


  • Rich
    Rich Domo Employee

    Upsert is a really useful approach to updating datasets. It is already feature of some datasets, but not all; and it currently varies connector by connector. We do not yet publish an upsert approach via API for Domo datasets, but when that feature becomes available you will find it here:



    We'd suggest you reach out to your Domo Account Executive and describe the specific upsert needs you have. From there, your Account Executive will help promote your need to the developer team and get you a roadmap for upsert on your specific dataset needs.

  • @Rich, is there an UPSERT methodology for SQL in the dataflows?  I haven't figured out how to save the resulting dataset nor can I get the SQL to actually resolve.    

  • Rich
    Rich Domo Employee

    Hi Matthew,


    The closest we've seen in SQL is to determine what makes a row unique and check against the existing dataset for matches. If you are able to find matches then you delete those rows from the existing dataset and replace those rows with their updated entries from the new dataset. Additionally, any net new rows not found in the existing dataset will also be appended. 


    An example of this could be if your file contained all entries for a day. You could check if your existing dataset has entries for that day (or days), delete those rows, and append the new dataset. This works if the new dataset contains all records for a day so you don't lose data. Combinations of entries can also be used if your data is more complicated than simply the date.


    Here is a link that may help you out:




  • Hi @Rich@RGranada

    I just posted about how I do UPSERTS in Redshift here.  

This discussion has been closed.