There isn't enough information to answer that. Some point that can help you:

1) I'd try getting the RAW SQL query and investigating it:

• How many indexes do exist in each table? Too many indexes may incur a lot of indexes to update, that can overload system resources
• Is there any trigger in this INSERT? What does it do? Probably you're supposed to look into each trigger and understand it as well
• When you say it takes 2 minutes to run, how were systems resources before it hit the server?

Do you have any replicas? If not, perhaps you can create replicas and segregate a part of this load

3) Besides that, I'm wondering if you're using the right service for this job. Actually, that sounds not, so I suggest you to take a look at GCP BigQuery, Dataflow, and Dataproc services. I think you can improve your architecture using Google's AI and ML services.