How it works

PRODUCT VALIDATION

From what we believe about your product to what Korean buyers actually did, and a decision you can act on.

THE PROBLEM

Most brands learn whether Korea will buy after they have paid to enter.

  1. 01

    Surveys are not sales

    “I would buy it” and paying for it are different answers.

  2. 02

    Launches chase totals

    Campaigns are built to raise money, not to explain why a product did or did not sell.

  3. 03

    Costs come first

    Certification, Korean labelling and stock are paid for before anyone knows if Koreans will buy.

Test with real Korean buyers before you commit.

Five stages. The label under each shows how certain its findings usually are, rising as the evidence moves from belief to purchase.

  1. 01

    Hypotheses

    Which message, image and bundle could work for Korean buyers, written down before anything runs.

    MEASURED: NOTHING YETHYPOTHESIS
  2. 02

    Korean landing

    Pages in Korean, each version shown to a comparable group of buyers, at the same price.

    MEASURED: VISITS, READING, SIGN-UPSINFERENCE
  3. 03

    Real campaign

    A crowdfunding campaign, opened only when the product can ship. Below the goal, nobody is charged.

    MEASURED: PLEDGESINFERENCE
  4. 04

    Behaviour data

    Pledges that were charged and kept, by version, channel and buyer group.

    MEASURED: NET PURCHASESCONFIRMED
  5. 05

    Decision report

    Enter, adjust, hold or stop, with the conditions each conclusion rests on and what is still open.

    MEASURED: WHAT REMAINS OPENUNKNOWN

WHAT WE COUNT

Not every click is demand.

We record each step a buyer takes, and weigh them differently. The decision rests on what people paid for and kept, not on how many looked.

Page visits
Who reached the page, from which channel. The denominator, never the result.
CONTEXT
Clicks and reading
Which message and image held attention. Useful for choosing the story, weak on its own.
CLUE
Sign-ups
People who asked to be told when the product launches. A real step, but free to take.
SIGNAL
Pledges
People who committed money in the campaign, at the same price as everyone else.
STRONG SIGNAL
Net purchases
Pledges that were charged and not cancelled or refunded. This is the result we decide on.
RESULT

WHAT YOU GET

A decision, and how sure we are about it.

Every conclusion states the sample, channel and period it was tested on. A confirmed result means confirmed under those conditions, not a promise about all of Korea.

Enter
The evidence supports going ahead, and says with which story.
Adjust
Something sold, something did not. Change it and test again.
Hold
Not enough signal yet. Say what would settle it.
Stop
Korea is not the next market for this product, for now.
CONFIRMED

Seen in this experiment’s own data.

INFERENCE

Reasoned from that data, not observed directly.

HYPOTHESIS

Worth believing, still to be tested.

UNKNOWN

Not answered yet, and said so.

WHO IT SUITS

Built for brands that would rather know than hope.

  1. 01

    A good fit

    A product already selling in one to three markets, that could ship to Korea, from a team that wants the honest answer, including “not yet”.

  2. 02

    Not yet a fit

    An idea without a product, a product that cannot legally be sold in Korea yet, or a brief that needs the answer to be yes.

  3. 03

    Not what we do

    Charging different buyers different prices, buying reviews, or promising a sales figure before a test has run.

What we need from you.

Four things, agreed before the test plan is written.

The product
Samples, specifications and the materials or ingredients, so certification and Korean labelling can be checked before launch.
The brand
Images, story and the claims you can stand behind. We adapt them for Korean buyers; you approve every version.
The numbers
Unit cost, shipping terms and the smallest order that makes sense, so the landed-cost engine can price the product in Korea.
A decision-maker
One person who can agree the test plan and act on the result.

QUESTIONS

Questions, answered plainly.

About a validation test

Do you change the price for different buyers?

No. Every backer sees the same product, the same price and the same benefits. We vary the message, the images and the bundle. If price itself is the open question, we ask it in a separate survey that says what it is, and label the answer as what people said rather than what they did.

Is this a forecast of Korean sales?

No. It is evidence from a defined test: who saw the page, through which channel, over which period. The report says how far that can be generalised, and marks what is still unknown.

What if the campaign does not reach its goal?

Nobody is charged, and the result still counts. The report sets out what moved buyers and what did not, and whether a changed offer is worth testing again.

Who sells the product to Korean buyers?

That is agreed before any campaign opens, and the campaign page names the seller. A campaign opens only when the product can actually be delivered in Korea, with certification and Korean labelling checked first.

How long does a test take?

It depends on the product and the channel. We agree the length, the budget for reaching buyers and the rule for stopping before anything runs, so the result cannot be read to suit anyone afterwards.

What happens to the buyers’ data?

We collect only what the test needs, with consent, and report to brands in aggregate. We do not hand over Korean buyers’ personal data.

NEXT · PRINCIPLESThe rules we hold every test to

FIRST PILOTS

We are choosing our first pilot brands. Is yours one of them?