Review the output, then ship
See exactly what one guest receives, then push the file to your email platform and schedule per-guest sends.
- Subject
- Marcus, your $5 off your next $20+ order is waiting at Verde Cantina
- Send
- Tue, Jun 9, 2026 · 5:00 PM (local)
Hi Marcus,
Taco Tuesday hasn’t been the same without you and the crew — so here’s $5 off your next $20+ order at Verde Cantina, this week only.
[ Order now → ]
Your reward is already loaded in the Verde Rewards app. See you soon.
— The Verde Cantina Team
Offer valid through Sunday. Minimum $20 subtotal. One redemption per member.
Tuesday tickets run 3–4 taco plates in a single order — clearly feeding a group. The hook leans on that standing Tuesday taco run, not the discount.
Those Tuesday orders land ~6:30 PM, so 5:00 PM reaches them while the group is still deciding where dinner’s coming from.
Every row carries both rationales, citing the guest’s actual orders — so any send is auditable, never a black box. Skim a handful before you ship.
The upload CSV drops into any customer-engagement / email platform — Klaviyo, Braze, and the like. The steps are the same everywhere:
- 01
Import & map
Create a list or segment, upload the CSV, and map the columns: email and first_name to their standard fields, and personalized_message_part, send_date, send_time (plus the rationales) to custom profile properties.
- 02
Wire the message with a fallback
Drop the personalized field into the email body with a safe default, so anyone missing a value still gets a clean line.
- 03
Schedule per-guest sends
Send each guest at their own send time. Best: a flow / Canvas that waits until each profile’s send_datetime_local, then sends. If your platform doesn’t support that, segment by send slot and schedule a send for each.
You just turned one offer into twenty-five thoughtful, on-brand messages — each timed to the guest, each drawn from their real ordering history — while only ever approving the rules, not every message. Do it continuously, across every channel and every guest, and that’s the whole idea behind a CDP + AI.
