Platform · Reach
Don’t start by guessing who might convert. Start from who already does.
Most lead tools start from an audience hypothesis. REACH starts from the people who actually came back at full price in this shop, and extracts what they look like.
Five steps
Start from the people who already bought, then find more of them
- 01
Pick the seed
Take the cohort that genuinely returns and pays full price — not the one with the highest lifetime total.
Those are usually different people. A high total built on discounts is not a good customer.
- 02
Extract the signature
What the seed cohort shares — first SKU, price band, channel, timing.
- 03
Expand to the lookalikes
Sync to Meta, TikTok and Google — hashed before it leaves, consented rows only — and let the platform find the lookalikes.
We never accumulate data on non-customers. We tell the ad platform who to look for.
- 04
Build the ads and the landing page
Three ad drafts and a landing page built around the SKU that cohort actually responds to. Test on a capped budget, then push the winner.
- 05
Capture and hand off
Leads are scored against the signature on arrival. The closest matches get contacted first, then everything flows into KEEP.
This is where the loop closes — the leads REACH finds become the data KEEP works and the numbers PROOF measures.
Nobody is added along the way. Reach starts from people who already bought and removes everyone it cannot legitimately message. Figures illustrate the display.
The mistake everyone makes
Pick the wrong seed and every baht spent afterwards is wrong with it
what most people pick
Highest lifetime spend
The total is high because they bought heavily on discount. The lookalikes you get back are discount-waiters too. Cost per lead may look fine; cost per returning customer gets worse.
what reach picks
Customers who return at full price
Their totals may not be the highest, but they are what keeps the shop alive. Expand from them and you get people likely to return the same way — which is the number we measure.
Ads that only bring one-time buyers aren’t bad ads. They are the platform doing exactly what it was told to look for.
Where we stand on data
We don’t accumulate data on non-customers
detail
Hashed before it leaves
Identifiers are always hashed on the way out. The receiving platform can match without ever seeing raw data.
Consented rows only
The consent ledger binds to the identity, not the file. Withdraw consent and that person drops out of every audience next cycle.
The leads belong to the shop
Not to Parallax, and never reused across accounts.
REACH is available from Growth up. Media spend is billed by the ad platform directly — never through us, never hidden in the subscription.
เริ่มจากไฟล์ยอดขายที่คุณมีอยู่แล้ว
หนึ่งไฟล์ หนึ่งบรีฟ และวิธีวัดผลที่เตรียมไว้ให้ ไม่ต้องติดตั้งอะไรเพิ่ม