Register a partner webhook subscription via the Toast Developer Portal specifying the ORDER_CLOSED and CHECK_CLOSED event types for dine-in reconciliation
In your webhook handler, parse the diningOption field to separate dine-in covers from delivery and to-go orders
Aggregate cover counts per time period using the openedDate and closedDate timestamps on each check
Pull delivery order counts from the Toast Orders API for the same time windows using the serviceArea filter set to third-party delivery channels
Calculate the ratio of delivery volume to total covers to assess delivery dependency and identify peak periods where delivery backfills low dine-in occupancy
Publish the combined dataset to a BI dashboard refreshed on each ORDER_CLOSED event
Known gotchas
Toast webhook payloads may arrive out of order during high-volume periods; use the modifiedDate field rather than receipt order for temporal sorting
Checks with multiple diningOption values (split dine-in and takeout on one check) must be decomposed by line item, not by check, to avoid double-counting covers
Toast partner webhook subscriptions require explicit restaurant opt-in via the partner activation flow; webhooks will not fire for restaurants that have not completed activation even if the subscription is registered
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp