All use cases

Convert

Price-drop alert on a watched item

The jacket your shopper kept looking at just dropped below what they were willing to pay — the agent tells them the moment it does, with a one-tap path to checkout.

Get aheadSystem triggerEmailPushE-commerce & Retail

What it does

A shopper looks at an item, decides it's a little too expensive, and moves on. Days later the price falls — but the shopper has stopped checking and never finds out, so a sale that was one price-point away is lost to silence. The agent remembers the items each shopper viewed or saved without buying, watches them for price movement, and pings the shopper the moment a watched item drops to a level their behavior suggests they'll act on — with the new price and a direct path to checkout. Intent that was parked on price gets converted at the threshold instead of forgotten.

How it works

  1. 1
    Trigger. a price-change event fires on an item that this shopper viewed or saved but did not buy.
  2. 2
    Decision. the agent checks how far the price moved, what it remembers about the shopper's prior interest and likely willingness-to-pay, and consent and frequency caps; a judge gates the message before send.
  3. 3
    Action. a concise alert on the shopper's opted-in channel — email or push — leading with the new price, the saving, and a one-tap link straight to the item.
  4. 4
    Follow-through. if the shopper has already bought, or the item goes out of stock, the alert is cancelled before it sends so the message is never stale; conversions are measured against a holdout, and the client keeps owning the send channels.
Example
AgentGood news on the down jacket you were looking at — the price just dropped. It's now lower than when you last viewed it, in your size and in stock. Tap here to go straight to it.
CustomerIs that the new-season one or last year's?
AgentIt's this season's, same colour you viewed. The new price holds while stock lasts — here's the checkout link if you'd like to grab it.

Configuration

How the agent is wired for this use case.

Triggera price_changed system event from the pricing or catalog service, matched against this shopper's viewed/saved-item list.
Tools & actions
  • Pricing service · receive price-change events and the old/new price delta for the SKU
  • Watchlist store · match the changed item to shoppers who viewed or saved it without buying
  • Customer memory store · read prior interest and inferred willingness-to-pay to decide whether the drop is worth a ping
  • Product catalog · confirm the item is in stock in the shopper's size before sending
  • Messaging channel · send the alert via email or push with a deep link to the item
  • Holdout ledger · assign test or control and record any resulting order
Autonomyruns unattended; every alert is judge-gated and suppressed automatically when the shopper has already purchased or the item is out of stock. The alert states a real, current price only — no fabricated urgency or countdown the catalog cannot back.
Channelsemail · push
Escalationif the shopper replies with a question the agent cannot ground in catalog data, or raises a billing or order issue, the thread routes to chat or a human.

What you need

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

Signals

price_changed events from the pricing service, viewed/saved-item events per shopper, order_completed and stock-out events for stale-send cancellation

Data

per-shopper watchlist of viewed/saved items, price history, stock and size availability, customer memory, per-channel consent state

Guardrails

explicit consent and frequency caps; judge gating on every unprompted message; automatic cancellation on prior purchase or stock-out; the price stated must always match the live catalog, with discount framing only under merchant policy

Metrics it moves

  • conversion-rateup: parked-on-price intent is closed at the moment willingness-to-pay is met
  • recovered-revenueup and provable: orders following the alert are attributed against a holdout
  • click-through-rateup: a real price drop on an item the shopper already wanted is a far stronger open-and-click reason than a generic promotion

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