After evaluating MeetGeek's Salesforce integration for the past quarter, my overall assessment is that it delivers on core functionality but introduces significant cost and complexity overhead that must be carefully modeled. Our primary use case was automatic attachment of meeting summaries and transcripts to Opportunity and Contact records to enrich our sales team's context.
The integration works as advertised from a technical standpoint:
* Meeting recordings, AI summaries, and transcripts are pushed to Salesforce as Files.
* The object linking (Contact/Opportunity) is accurate when calendar events are properly tagged.
* The automated workflow eliminates manual uploads, which the sales team appreciates.
However, the operational and financial implications are non-trivial:
* **Storage Cost Amplification:** MeetGeek files are stored in Salesforce Files, consuming valuable Salesforce Data Storage. Our instance saw a 23% month-over-month increase in storage usage, pushing us closer to our limit and incurring additional cost.
* **Data Duplication:** The same file (recording, transcript, summary) is often attached to multiple related records (e.g., multiple Contacts on the same call), multiplying storage consumption.
* **Lack of Granular Control:** We could not configure the integration to attach *only* the summary, for example. It's an all-or-nothing package: video, audio, transcript, and summary. For many deals, only the summary is needed, making the other assets wasteful.
**Recommendation:** Before enabling this, calculate your projected Salesforce storage growth. Model the cost of additional Salesforce Data Storage against the productivity gain. For us, the ROI only became positive after we negotiated a higher storage tier with our Salesforce AE. For smaller teams or tighter contracts, this could be a deal-breaker.
Attached are screenshots of our storage analytics pre- and post-integration, and the configuration panel showing the lack of selective asset attachment.
—Jake
Show me the bill.
Interesting about the storage cost. Did you consider pushing only the transcript text as a note instead of the full file? That might cut down on the data duplication and storage hit, but I'm not sure if MeetGeek supports that.
Yeah, the storage cost angle is something we totally missed in our initial planning too. We got so focused on the automation benefit that we didn't model the downstream impact on our Salesforce instance limits.
A follow-up question, if you don't mind: what was your actual process for cleaning up old files? Did you have to write a bunch of scheduled jobs, or does MeetGeek offer any kind of lifecycle management for the files it creates? I'm nervous about starting this only to create a new maintenance headache.
One step at a time
Great question about cleanup. We didn't find any native lifecycle tools in MeetGeek, so we had to roll our own. Ended up writing a simple Kubernetes CronJob that runs weekly and calls the Salesforce REST API to purge files older than 90 days linked to specific record types. It's a bit of a maintenance burden, but it works.
You could also look at Salesforce's own data archiving policies, but they're a bit blunt for this.
yaml all the things