Skip to content
Case 03 · Automated price tracking

The price dropped after they booked. The system noticed.

Prices keep moving after people book. When one drops before the change deadline, there’s money on the table, and nobody has time to recheck every booking every week. This began as a sketch of a patient watcher: check each booking against today’s price, compare like with like, and raise a hand only when the saving is likely real. Only then does a human verify. Today it watches every open booking until its deadline passes.

the idea: bookings till deadline today’s price? same thing? flag it person decides =? …every booking, until it can’t change anymore
the first sketch↓ what it became
1Oct 3

The booking goes on the watch list.

Daniel booked the hot-tub cottage for $846, with free changes until November 7. From that moment, his booking joins the watch list, along with everything else that can still change.

Designed forWatching only what can still be acted on. A booking joins the list when it’s made and leaves the moment its deadline passes.
Booking · #4803

Hot-tub Cottage

Fri 14 → Mon 17 Nov · 3 guests + dog

Free changes until Nov 7

 NOV deadline on the watch list
2Oct 10 → Oct 24

It checks again, and again, until the price moves.

Every week it looks up today’s price for the exact same stay. For three weeks nothing much changes. On October 24 the cottage shows $781, which is $65 less than Daniel paid.

Designed forPatience at scale. Every open booking is rechecked on a schedule, without anyone having to remember, and often enough that a short-lived drop isn’t missed.
$850 $780 Oct 3 Oct 10 Oct 17 Oct 24 same… same… same… paid $846 $781 −$65
3Oct 24

But is it really the same deal?

A lower price is only a saving if it buys the same thing. The cheapest option today is $749, but it can’t be changed or refunded. The $781 option matches Daniel’s booking on everything.

Designed forLike for like, only. A lower price that quietly drops a perk or a guarantee isn’t a saving, so it’s never reported as one.
compare like with like
Option A$749Same dates ✓Same home ✓Dog welcome ✓Free changes ✗
Option B$781Same dates ✓Same home ✓Dog welcome ✓Free changes ✓
 not a real saving same deal, $65 less
4Oct 24

It raises its hand.

It puts the saving on the team’s short list, with everything they need to act: what Daniel paid, today’s price, what’s included in both, and how long is left before the deadline.

Designed forA person makes the call. The system finds the saving and shows its working; someone on the team decides whether to rebook.
Price drop · #4803

Hot-tub Cottage

was $846 → now $781

Save $65

Same dates, home and perks · 14 days to deadline · Maya

RebookDecline
 on the short list
5Oct 25

A person decides.

Maya rebooks at the lower price, and Daniel gets $65 back. Now the saving counts as real, not just found. If Maya had passed on it, she’d have noted why.

Designed forHonest numbers. Savings found and savings actually captured are counted separately, so nobody mistakes a long list of alerts for money in the bank.
Decision · Oct 25 · Maya

Rebooked at $781

Daniel refunded $65

 found: $65 saved: $65 declined drops need a reason:
not like for likecustomer said notoo close to the deadline
6Nov 7

The deadline passes, and it lets go.

Once the booking can’t be changed anymore, it leaves the watch list, and any alert still open on it closes by itself. Nothing lingers to inflate the numbers.

Designed forA clean list. Nobody spends time on bookings that can’t change, and the savings still on the table are always ones that can really be captured.
NOV
Booking · #4803

Hot-tub Cottage

Watch ended Nov 7

 CLOSED open alerts close on their own

What it took to make it hold up.

Finding one lower price is easy. Watching every booking for weeks, telling a real saving from a false one, and keeping the numbers honest is the real job. Three things shaped every decision.

A

Accurate

Only like-for-like savings are reported. Every drop is checked for the same dates, the same product and the same inclusions before anyone sees it.

R

Robust

It keeps up as bookings change, get cancelled or run out of time. If it can’t price something cleanly, it says so and sets it aside. It never guesses.

L

Learns as it goes

Every declined drop carries a reason. Those reasons sharpen what counts as a real saving, so the same false alarm doesn’t come back.

watching: on (flip it any time)

The team stays in charge.

  • Nothing is rebooked without a person’s say-so.
  • Every saving shows its working: what was paid, today’s price, and what’s included in both.
  • Lists filter by agent and by size of saving, so each person sees what’s theirs to act on.

Same idea, other places.

It fits anywhere prices keep moving after purchase and the deal can still be changed before a deadline.

  • Hotel stays on flexible rates
  • Car rentals
  • Retail purchases with price protection
  • Group and event bookings