Hey everyone! I've been using MeetGeek for a few months to record and transcribe my team's meetings, and I love the automatic summaries. But I got a bit of a shock last month when I saw the billed minutes on my invoice—it seemed way higher than I expected.
So, I decided to write a little Python script to audit it. I'm pretty new to data engineering, but I figured I could use the API to pull my usage data and compare it to my own meeting logs. The idea is to check if the billed minutes line up with the actual meeting durations I have in my calendar.
Here's what I did:
- Used the MeetGeek API to get a list of all processed meetings for the billing period, along with their durations.
- Fetched the original calendar events from Google Calendar API for the same dates.
- Compared the durations. I'm summing up the MeetGeek "processed seconds" and converting to minutes for billing.
I ran it for last month and found a discrepancy of about 15%! My script says we should have been billed for about 300 minutes, but the invoice was for 350. I'm not sure if I'm missing something—like does MeetGeek count time for failed processing attempts? Or maybe my script's logic is just off.
Has anyone else tried something like this? I'd love some feedback on my approach, especially if there are nuances in how MeetGeek calculates billable time that I'm not aware of. Also, if you have tips on making this into a more automated check (maybe with Airflow?), I'm all ears!
-- rookie
rookie