Field marketing is account-based or it is nothing. Begin with the twenty to fifty accounts the revenue team most wants to move this year, and treat every event decision as a question about those accounts.
This inverts the usual order. Most programs choose an event, then look for accounts to invite. That is how you end up in a city where none of your targets are.
Cluster the account list by geography and look for density. Three targets in a city is not a dinner. Twelve is.
Two data sources matter and most teams only use the first:
| Format | Seats | Best for | Watch out for |
|---|---|---|---|
| Executive dinner | 10-14 | Breaking into named accounts | Expensive per head; dies without warm invites |
| Roundtable | 10-20 | Category education, peer credibility | Becomes a webinar if anyone presents |
| Sports suite or experience | 15-40 | Existing relationships, expansion | Weak for cold targets; hard to run a conversation |
| Conference side event | 20-60 | Efficiency when targets already travel | You compete with the conference itself |
Pick one. Running three formats once each teaches you nothing; running one format three times teaches you how to run it.
This is the part that decides whether seats fill, and it is operational rather than creative:
Record the account list and its state on the day. Re-measure a quarter later against comparable accounts that did not attend. Report new stakeholder engagement, restarted deals and reopened closed-lost accounts - not headcount and photographs.
The honest difficulty of this program is step 4. Knowing which of two hundred colleagues has the warmest tie to each of forty guests - and then tracking who has asked whom - is the work, and it is why most programs collapse back into one marketer sending one email. Orbb's Events Agent exists to run that step: it picks the city from the relationship graph, builds the segmented list, assigns each invite to the closest colleague, and sources the +1s.