The short version
- We built a product, shipped it, put it live, and it earned nothing across its entire life.
- Nothing was broken. It worked, it looked right, and no distribution existed for it.
- Launch is an event. Distribution is a system, and the second one is the entire job.
- The test before building: name the first hundred people who will see this and how each one arrives. If you cannot, you are building on hope.
We built a product. Designed it, wrote it, integrated payments and fulfilment, put it live.
It has earned nothing across its entire life. Not a slow start. Zero.
Nothing was broken. The thing worked. It looked right. Payments processed. The failure was that no distribution existed and we had not treated that as part of the build.
What actually went wrong
We answered every question except the one that mattered.
We knew what it did, who it was for in the abstract, what it cost, how it would be fulfilled. We had not answered how the first hundred people would arrive, in specific terms, with a channel attached.
| Question we answered | Question we skipped |
|---|---|
| What does it do? | Who sees it in week one, by name or by channel? |
| Who is it for? | Where do those people already gather? |
| What does it cost? | What makes anyone look at it at all? |
| How is it fulfilled? | What happens in month two if nothing moves? |
The right-hand column is the product. The left-hand column is the artifact.
Why is this so easy to walk into?
Because building is legible and distribution is not.
A build has a definite shape. There is a list, the list gets shorter, and every day produces something you can look at. It feels like progress because it is progress, just not on the constraint.
Distribution has no natural end, no satisfying artifact, and a much worse feedback loop. So it gets deferred to launch, and at launch it turns out to be a system that needed months of lead time rather than a task that takes a week.
Launch is an event. Distribution is a system. Only one of them determines whether anybody arrives.
There is also a status problem. Building is respectable. Sitting in a community for three months so that people know who you are before you launch anything feels like avoidance, right up until it is the only reason a launch lands.
The test we use now
Before anything gets built: name the first hundred people who will see this, and how each one arrives.
Not a persona. An actual channel with an actual mechanism. An existing list of this size. A community where this audience already gathers and where we are already known. A search term with volume and a page we can rank. A partner who will introduce it.
If the honest answer is that people will find it, there is no plan. There is hope, and hope has a poor conversion rate.
The related discipline
Decide the kill criteria before launch, and write them down.
After launch, sunk cost makes that judgment much harder, and a product with no users tends to survive on the strength of the work already in it. A date and a threshold agreed in advance is the only version of this decision that gets made cleanly.
This is the same failure mode as taking on work with no defined outcome, which is covered in the work you should refuse. In both cases the problem is not effort. It is that nobody agreed in advance what success would look like, or when you would stop and check whether it had happened.
Frequently asked questions
Why do products launch and get no users?
Because launch is treated as the finish line rather than the first day of distribution. A working product with no channel reaches nobody. Nothing has to be broken for this to happen, which is what makes it so easy to walk into.
What should you decide before building a product?
How the first hundred users arrive, specifically. Name the channel, the message and the reason someone would look. If the answer is that people will find it, there is no plan, only hope.
Is 'build it and they will come' ever true?
Only when distribution already exists. A company with an audience, a list or an existing customer base is not testing whether people will come, it is reusing a channel it already built. Starting from zero is a different problem.
How much effort should go into distribution versus building?
At minimum equal, and usually more on distribution once the product works. The build has a definite end. Distribution does not, and it decays if nobody tends it.
What is the earliest signal a product will not find users?
Nobody asks for it before it exists. If you cannot get people to want a thing that is not built yet, adding the build does not create the want.
Should you kill a product that has no traction?
Decide the criteria before launch and write them down, because after launch the sunk cost makes the judgment much harder. Set a date and a threshold, then honour it.
Sources
- Chua Network delivery data across 8 client accounts (internal fact bank)
- Chua Network engagement records, anonymized (internal experience bank)
