Journeys & sequences

A journey is a multi-step automation: a trigger, an ordered list of steps with delays, branching logic, and exit conditions. Zubby’s recovery campaigns are journeys with sensible defaults; you can also build your own from scratch for onboarding, replenishment, post- purchase, or win-back.

Anatomy of a journey

Built-in templates

Six templates are battle-tested and live under Journeys → Templates:

• Classic 3-Step Cart Rescue — 15 min, 1 day, 3 day send. Discount on the third step.
• Browse Bounce-Back — single touch 4 hours after browse abandonment.
• First-Order Replenishment — fires X days after first order, calibrated to your AOV-to-second-order gap.
• Winback 30/60/90 — three sends spaced 30 days apart, escalating offer.
• VIP Concierge — silent journey that pings your team when a VIP browses without buying.
• Post-Purchase Drip — review request, care guide, cross-sell. Three steps over 21 days.

Building a custom journey

Open Journeys → New. The canvas is a top-down flowchart. Drag steps from the right panel, connect them, configure each.

Best practices:

1. Start with a strong exit rule (made purchase, unsubscribed). It’s the single biggest factor in keeping a journey from feeling spammy.
2. Use branches sparingly — a 3-step linear sequence outperforms a tangled 9-step tree in 80% of tests we’ve run.
3. Test on a 10% audience before flipping fully on. Stat sig is usually visible within a week on a Growth-tier store.
4. Match channel to context. Email for considered purchases, SMS for time-sensitive (low stock, ending sale).

Conditions

Branches and exit rules support these conditions:

• Profile — lifecycle stage, VIP tier, segment membership, LTV bracket.
• Engagement — opened last email, clicked last email, replied to last email.
• Order behavior — placed order since trigger, ordered specific SKU, refunded recent order.
• Cart state — cart still contains specific SKU, cart value over $X.
• Custom event — fired from your stack via REST API.

Firing events programmatically

Trigger any journey with a custom event from your own code:

curl -X POST https://api.zubbyai.com/api/v1/merchant/{storeId}/events \
  -H "Authorization: Bearer sk_live_..." \
  -H "X-Store-ID: {storeId}" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "warranty_registered",
    "email": "shopper@example.com",
    "properties": {
      "productSku": "APEX-JACKET-M",
      "purchasedAt": "2026-04-22"
    }
  }'

Analytics

Every journey shows step-by-step performance: sent, opened, clicked, converted, revenue attributed. Compare A/B variants from the same screen. Drill into individual shopper paths to debug a flow.

Limits