How To Test A New Offer

The expensive way to test an offer is to build it, launch it, and find out.

The cheap way is to try to sell it before it exists.

Asking is not testing

"Would you be interested in something like this?" gets a yes from almost everyone. People are polite, and hypothetical purchases are free.

The only signal that means anything is someone trying to give you money, or at least putting something at risk: a deposit, a booked time, a place on a waiting list with their details attached.

Interest is not demand. Every business that launched to silence had plenty of encouraging conversations first.

Sell it before you build it

Describe the offer clearly. What it is, who it is for, what it costs, what they get.

Then take it to people who already trust you: past customers, current customers, your list, your network. Not strangers. Strangers are a marketing test, and you are not testing marketing yet.

Ask them to buy it.

If several do, you have something, and now you have both the validation and the money to build it.

If they all hesitate, the conversation about why is worth more than any survey. Listen for which part they stall on: the price, the scope, whether they believe it will work, or whether the problem matters enough to pay for.

Deliver the first few by hand

The first version does not need to be efficient. It needs to be real.

Do it manually. Overservice it. Spend too long on it.

You are buying two things: proof that the outcome is deliverable, and a detailed understanding of where the work actually goes. Almost every assumption you made about effort will be wrong, and that is what sets the real price.

Systemise after you know it works, not before. Building the process for something nobody buys is the most common way to waste a quarter.

Price it properly from the start

The instinct is to discount the first version because it is unproven.

Cheap tells you almost nothing. People buy cheap things they do not really want, and then you scale something with no margin and an unclear signal.

Charge close to what you intend to charge. If you want a launch price, frame it as a foundation rate for the first few clients, with the real price stated. That way you learn whether the actual number works.

The price you can charge is part of what you are testing, not something to settle afterwards.

What a failed test looks like

Nobody buys. That is a clean result, and it cost you a week.

Before you bin it, check three things.

Were you talking to the right people. A good offer to the wrong audience fails the same way a bad one does.

Was it clear. If you had to explain it twice on every call, the offer is confusing, not unwanted. Confused people do not buy.

Was the problem urgent. Plenty of real problems are not annoying enough to pay to solve this month. That is the most common quiet failure, and it usually cannot be fixed with better copy.

Set the threshold first

Decide before you start what a yes looks like. Three sales. Five deposits. Whatever suits your size.

Write it down, because after two weeks of effort you will be tempted to accept one sale and a lot of warm feedback as proof.

A test you can talk yourself into passing is not a test.

The real advantage

Testing this way is fast enough to do repeatedly.

Most businesses run one offer for years because launching feels enormous. If a test takes two weeks and a few conversations, you can run four a year, and you only need one to land.