I'd love to hear how you all are sending product usage data from your apps to Marketo to use for campaign targeting.
We're tracking product usage data with Segment.io. From Segment, the data goes into our warehouse, from the warehouse to some other tools, and then eventually gets dumped into to a custom object in Salesforce that then syncs to Marketo.
Because only 10K objects sync per cycle, our biggest challenge has been dealing with some pretty significant sync backlogs. Just last week, there were nearly 355K items backlogged, and we've only gotten it down to 195K.
We've considered going directly from Segment to Marketo, but are concerned that we'll blow through our API calls.
I'd be grateful to anyone who'd be willing to share how you've handled sending large amounts of product usage data to Marketo.
Lots of ways to do this with diff databases.
From my experience with Segment, it can certainly do this, however, API calls seem very high and controls on entry to Marketo are non-existent.
I would consider another DB with an API connection you can build yourself into Marketo Custom Objects or fields. Not sure of your particular needs, so could be other methods or solutions.
As you've noticed already, creating a CO record for individual product actions will swamp your API calls, even if the insert is bulkified.
Like Josh says, bulk-loading Marketo COs instead of SFDC COs will help a lot with the overhead, since you'll be updating out-of-band from the SFDC sync. You can have an unsynced SFDC object that you maintain as well as the Marketo CO, so both sides have the data (eventually!).
Still, there's a ceiling to how much you can pack into individual Marketo COs in a given time period. You might think about creating "roll-up" objects that encapsulate the last N actions in one CO. Of course I don't know anything about the level of detail you're logging for each action. If you have 10 fields for each action then you can only put 4 actions in one CO.