The second box that never ships

The situation

A scaffold, not a teardown. The brand has not been selected and nothing has been ordered yet, so this section is the set of facts the finished piece has to establish in 150–250 words — all of them observable from the storefront and a real customer account.

• Chilled, frozen or shelf-stable, and what does that mean for how the first box arrives — and for what has to be at home when it does?
• What does the onboarding quiz ask, and does the plan it recommends line up with the servings actually printed on the pack?
• What is the first-order discount, and what does the price step up to on the second delivery?
• What cadence is the default, how much freezer or fridge space does one delivery need, and can that cadence be changed before the second box ships?
• Which ecommerce platform, which subscription app, which email platform — and who carries the cold chain?

What we found

The diagnostic section. Nothing goes in it without a timestamp and a screenshot from the account. The questions it has to answer:

• How many messages arrive between the first delivery and the second, and on which days?
• Is there anything covering the transition period — the two to four weeks when a household is fitting the new plan into its week and the portion size, the variety and the customer’s confidence in the whole idea are all unsettled?
• Does anyone check in after the first meal? A first-order check-in is the one message that catches a portioning or taste problem while it is still fixable.
• Does the second charge arrive before or after the first delivery has been eaten, and does the customer get any warning of it?
• What happens when a perishable shipment misses its window? A late or warm box is a support ticket, a refund and usually a cancellation — and the WISMO volume spikes on exactly the days a renewal batch ships.
• What does the cancel flow offer when the reason is “I have too much food” — which is a cadence problem wearing a cancellation’s clothes?
• Is a smaller box, a longer interval or a skip visible on that screen at all, or is the only visible control Cancel?
• If the second charge declines, what arrives, and how often?

What we’d build

A menu of the candidates. Which of them matters here is exactly what the observations above decide.

• A transition-period sequence timed to the first three weeks, written for the customer standing in front of an unfamiliar freezer drawer rather than for the brand.
• A first-order check-in that asks one answerable question and routes a bad answer to a portion or menu change instead of to support.
• Reorder timing modelled from servings per box, household size and delivery volume, so the prompt lands before the freezer is empty, not after.
• Proactive delivery messaging on the perishable window, so a cold-chain exception reaches the customer before they open the box and before they open a ticket.
• Cancel-flow alternatives that name the real problem: too much food, wrong size, wrong menu, going away.
• A retry ladder and dunning sequence for the second charge, which is where involuntary churn concentrates on a stepped-up price.

What has to happen before this publishes

1. Select a real, currently trading chilled or frozen meal brand with a public subscription.
2. Order, receive at least two deliveries, and hold the subscription long enough to see the second charge clear.
3. Log every message with its date, and record the cancel flow screen by screen.
4. Anything not observable from a customer account is marked unknown.
5. Right of reply to the brand before publication.