This application exclusively depends on job metadata from BigQuery for the cost analysis. Therefore, costs incurred by e.g. storage space are not included in our analysis. This may lead to a slight mismatch in total costs between the two systems.

Google rounds the amount of slot hours up to a set T-shirt size when billing based on the capacity price model. This rounding is yet to be implemented in BID. The current prices are however close enough to the actual cost to be useful.

The main goal of this application is to enligthen users as to what queries generate the main cost for their organization. We have therefore decided to ignore the free 1TB / month granted by BigQuery in order to not obscure which queries processed the most bytes. Assuming your organization is using on-demand pricing, a good rule of thumb is simply removing 5$ from the total cost each month to correct for this. Our prices are calculated by simply converting the totalBytesBilled from each job to dollars, and should therefore be thought of as a representation of bytes billed as opposed to actual cost.

Storage prices are not considered at all in this application, so the discount naturally is not included either.

In order to save costs for our clients we utilize a cache to store the analysis results. This cache resets every whole hour, so any users from 12:00 to 12:59 will get the same results, until the system updates at 13:00. If your organization sees value in getting more up-to-date data, feel free to contact us at:
bigquery-insight-devs@computas.com.