Convert
New-line launch outreach to past buyers
A new range just dropped, and the customer who bought the boots last season is exactly who it was made for.
What it does
A brand launches a new product line and defaults to a generic blast to the whole list, so the people whose past purchases actually pair with it get the same message as everyone else, and the launch week underperforms. The agent does the opposite: when the new line goes live, it tells each past buyer whose history genuinely matches ("the new trail range would go with the boots you bought last season"), with the reason in plain words. The launch lands as a relevant note from a brand that remembers you, not a campaign — and first-week conversion comes from the warm base instead of paid reach.
How it works
- 1Trigger. a new SKU or product line publishes in the catalog; the agent receives the launch event with the line's attributes and pairing logic.
- 2Decision. for each past buyer, the agent checks purchase memory for a genuine pairing — a complementary category, a matching use case, a size or style continuation — and weighs recency, consent, and frequency caps so only the customers it actually fits get contacted. A judge gates each message and its claimed reason.
- 3Action. a personalised note on the preferred channel — chat, push, or email — naming the specific prior purchase and why the new line pairs with it, with a one-tap path to the product.
- 4Follow-through. if the customer buys, browses, or opts out, pending sends are cancelled rather than left to go stale; conversions are attributed to the launch and measured against a holdout, so the warm-base lift is proven rather than assumed.
Configuration
How the agent is wired for this use case.
product_line_launched event from the catalog / merchandising system, carrying the new line's attributes and pairing rules.- Catalog / merchandising system · read the new line's attributes, stock, and the categories it pairs with.
- Customer memory · match each past buyer's purchase history against the pairing rules and rank by fit and recency.
- CRM · apply consent and frequency-cap checks; record who was contacted, who converted, and the holdout split.
- Order management system · confirm size/variant availability before naming a specific item.
- Messaging channel · send the personalised launch note and run any follow-up on the customer's preferred channel.
What you need
The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.
Signals
catalog launch event (product_line_launched) with line attributes and pairing logic.
Data
per-customer purchase history and memory, product pairing rules, consent state, contactability and channel preference.
Guardrails
consent and marketing-permission state, frequency caps so launch outreach doesn't stack with other sends, judge review of every unprompted message and its claimed reason, holdout measurement, the client keeps owning the send channels.
Metrics it moves
- conversion-rateup: the launch reaches the buyers it actually fits, with a specific reason, instead of a generic blast.
- attach-rateup: the new line is positioned against a complementary prior purchase, so it lands as an addition rather than a cold pitch.
- revenue-per-conversationup: warm-base launch notes convert at a higher value than untargeted reach, attributed against a holdout.
Related use cases
See it on your own customer journey
Bring one drop-off, one churn cliff, or one silent segment. We will show you what a proactive agent with memory and judgment does with it.
Book a demo