Hi everyone, I have a small question about normalization costs in BigQuery when back-filing. I want to get historical data from the last 3 years from Google Analytics, but the cost of the basic normalization would be too big, is there a way to reduce such costs? Thank you!

Hey Gabriel, just change your billing model to physical bytes. The expanded airbyte data takes a lot of space, but not compressed. Bigquery recently let you decide which type of storage you will be billed, physical or logical https://cloud.google.com/bigquery/docs/datasets-intro#dataset_storage_billing_models

Thank you for you answer, but my question was actually related to the normalization cost and not the storage cost. When Airbyte transforms the JSON data to columns, it does so with queries that would cost too much if I pull all 3 years of data

BigQuery bills you for bytes scanned. If you change your billing to physical bytes, it will bill your scanned bytes for physical bytes, which can reduce more than 80% of the cost for normalization

I didn't know that, thank you!

I just double checked that information and I may be wrong on this. But I think is worth to check

Hello Gabriel Barbutti, it's been a while without an update from us. Are you still having problems or did you find a solution?