ข้ามไปยังเนื้อหา
Parallax

Customers

We filter on the shape of the revenue cycle, not the industry label

A single location works. Not being a shop at all works. The engine predicts from the cycle, not from the sign above the door. These four have exactly one thing in common — their customers have a moment that deserves contact.

Four shapes

the same axis, four rhythms

◆ where the system acts

Replenish

buys on their own rhythm, then drifts past it

act while the gap is still recoverable

Recall

one service sets the date they should return

servicecontact in week fivedue

Expiry

a block of value that ends on a fixed date

act before it lapsesexpires

Considered

traces of interest gather before one large purchase

closest to decidingpurchase

Time runs left to right on every row, so the shapes can be compared rather than described. An illustration of the rhythm, not one account's data.

01

Replenish

The gap between purchases, per person

The system learns each customer's own cycle and flags them when they drift past it — not past the shop average.

Someone who buys coffee every three days and disappears for nine is a signal. Someone on a thirty-day rhythm is not.

industries in this shape

  • Restaurants
  • Cafés
  • Retail
  • D2C skincare and supplements
  • Pet food
02

Recall

The date the last service says they should return

The cycle is set by the last service performed, so the system knows in advance who should be contacted in which week.

A six-week treatment cycle should be contacted in week five, not week nine when they have already gone elsewhere.

industries in this shape

  • Aesthetic clinics
  • Dental
  • Auto service
  • Veterinary
  • Spas and salons
03

Expiry

The end date of a membership or course

An expiry date is the day an entire block of value disappears at once. The ranking follows value at risk, not the calendar.

A member with one session left needs contact before one whose membership expires in two months but who still attends regularly.

industries in this shape

  • Gyms
  • Golf courses and driving ranges
  • Tutoring schools
  • Insurance
  • subscription
04

Considered

Interest signals ahead of a large purchase

The system collects the traces of interest that precede a big purchase, then ranks which lead is closest to deciding.

Someone who paid for a fitting and has not bought is the hottest lead in the base, and the shortest path to a high-margin sale.

industries in this shape

  • Golf equipment
  • Automotive
  • Furniture
  • Property
  • Travel and hotels

Pilot qualification

Three criteria, none of them about size

why
We do not ask how many locations you have, what you turn over, or which POS you run — none of those predict whether the engine will work
01

At least six months of purchase history

Anything shorter is not enough for the system to learn each customer's cycle.

02

At least one way to identify a customer

Phone, email or member ID. If none exist yet, we provide point-of-sale capture so you can start accumulating.

03

At least one person who can approve a campaign

Approval is the default mode. With nobody to approve, nothing sends.

How we enter the market

Deep in one cycle at a time, never broad across industries

Forty accounts in one cycle are worth more than forty spread across four

why
Because cross-account learning happens inside a single cycle. Spread thin and no cycle accumulates enough for a signature to work, so every shop gets the same beginner-grade result
We turn down deals outside the cycle we are working, even ones that would pay — because they make the result worse for the customers we have.

A business with no way to tie an identity to a receipt is not ready for Parallax as it stands today — and we say so in the first conversation rather than selling and disappointing later.

เริ่มจากไฟล์ยอดขายที่คุณมีอยู่แล้ว

หนึ่งไฟล์ หนึ่งบรีฟ และวิธีวัดผลที่เตรียมไว้ให้ ไม่ต้องติดตั้งอะไรเพิ่ม