All use cases

Win back

Lapsed-traveller reactivation

They booked with you last year, browsed a city twice this spring, and have a points balance about to expire — and all they've had since is the same weekly fare blast everyone gets.

Get aheadTime-based triggerBehavioral triggerEmailPushWhatsAppChatTravel

What it does

A traveller who hasn't booked in a while still leaves a trail of intent: a destination they searched, a route they priced and abandoned, loyalty points quietly heading toward expiry, a season that always pulls them somewhere. Most travel win-back ignores all of it and sends one mass fare email. The agent reactivates each dormant traveller from their own history — the destination they browsed, a fare drop on a route they searched, the points about to lapse, the time of year they tend to travel — and reaches them on the channel they actually open. A blank "miss you" becomes a specific, timely reason to book the next trip.

How it works

  1. 1
    Trigger. a dormancy condition matures — no booking past the traveller's typical re-book window — or a behavioral hook fires (a fresh fare drop on a previously searched route, points nearing expiry, a seasonal window opening for a destination they have shown interest in), and the agent times the outreach to the hook rather than a calendar blast.
  2. 2
    Decision. the agent reads memory for the strongest reactivation angle — browsed destinations, abandoned route searches, points balance and expiry, prior trip pattern, channel preference — and a judge gates the drafted message for relevance and tone before anything sends.
  3. 3
    Action. a specific message goes out on the traveller's preferred channel — email, push, WhatsApp, or chat — leading with the personal reason (the route, the fare, the expiring points) and opening a conversation rather than dropping them on a generic results page.
  4. 4
    Follow-through. replies are handled in-dialog through to a booking hand-off in the operator's own flow; the outcome writes back to memory; if the traveller books on their own first, the pending outreach is cancelled as stale. Reactivation and recovered booking revenue are measured per cohort against a holdout.
Example
AgentHi Priya — you looked at Lisbon a couple of times back in March. Fares on that route have just dropped for early October, which is the kind of timing you've booked before. Want me to pull a few options?
CustomerMaybe. I've got about 8,000 points sitting somewhere too, do those still work?
AgentThey do, and 4,000 of them expire in November, so this is a good moment to use them. I'll put together a couple of early-October options that apply your points and send them here to compare.

Configuration

How the agent is wired for this use case.

Triggera dormant_traveller time condition from the booking platform, or a behavioral hook — fare_drop on a searched route, points_expiring, or a seasonal-window match — from the fare and loyalty systems.
Tools & actions
  • Booking platform · read trip history, last booking date, and typical re-book cadence
  • Customer memory · retrieve browsed destinations, abandoned route searches, channel preference
  • Fare/availability engine · pull live fares and options for the matched route and dates
  • Loyalty system · read points balance and expiry to fold into the reason and offer
  • Messaging channel · send the quality-gated outreach and run the reply dialog to the booking hand-off
Autonomyunattended, judge-gated for outreach and the conversation; the booking itself is completed in the operator's own flow, and any payment is a customer-completed step the agent never takes in chat.
Channelsemail · push · whatsapp · chat
Escalationa complex multi-leg or special-assistance request, a complaint about a prior trip, or a fares/policy question beyond the playbook hands off to a human travel agent with the history attached.

What you need

The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.

Signals

dormancy condition (no booking past the typical re-book window), fare_drop on previously searched routes, points_expiring, seasonal-window matches

Data

trip and booking history, browsed destinations and abandoned searches in memory, loyalty points balance and expiry, channel consent and contact state

Guardrails

judge gating on every unprompted message; suppression of do-not-contact and recently-contacted travellers; frequency caps; stale-outreach cancellation on organic booking; per-cohort holdout for honest measurement

Metrics it moves

  • reactivation-rateup, because the reason to return is built from this traveller's own browsing, routes, and points rather than a mass fare blast
  • recovered-revenueup, tracked per cohort as recovered booking value against a holdout
  • opt-out-ratedown, since a timely, personally relevant trip reason gets opened instead of reported

Related use cases

Win back

Memory-personalized win-back

the all-industry parent this travel case specializes

Engage / Lifecycle

Travel lifecycle-moment outreach

the still-active counterpart, keeping travellers engaged before they lapse

Convert

Conversational travel discovery and booking

where a reactivated traveller lands to discover and book the next trip

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