The specific problem
The advice aimed at people who ship a lot of small things is almost entirely about shipping faster, which is strange, because speed is the one thing this group is already good at. A competent solo developer can get a working version of most small tools in front of someone in three weekends. That has not been the constraint for years.
The constraint is selection. You have six ideas in a note, all of them plausible, and roughly three months a year of genuine building time. Choose the wrong one and the cost is not the code — code is cheap and you enjoy writing it — it is the twelve weeks of maintenance, support and slow disappointment attached to a product with four users.
The failure mode is specific and recognisable: you build the idea that is most fun to build. This is not a character flaw, it is a reasonable response to having no better information at the moment of choosing. The point of a priced page is to make the moment of choosing carry information.
The arithmetic of choosing badly
- months of real building time in a typical year
- 3
- ideas competing for it
- 6
- to publish a priced page for one of them
- ~8h
Evenings and weekends, after the day job takes its share
The number in most people's notes app
Our own time on the fifteen ideas listed here
Eight hours against twelve weeks
Eight hours per idea, six ideas, is roughly one long week. Against three months of building time it is a seven percent tax, and it buys you a ranking of the six that is based on something other than which one you find most interesting on a Tuesday.
That trade only works if the pages are genuinely cheap to make, which is where most attempts fall over. Building six landing pages by hand is not eight hours each — it is eight hours for the first and six for the fifth, because you rewrite the layout each time and then discover you need a second page for search traffic. The reason this site keeps its content in JSON is precisely that: the sixteenth idea cost the same as the second, and adding a page to an existing idea costs about forty minutes.
The honest caveat is that the ranking you get is coarse. It will separate the idea nobody wants from the idea two people want badly. It will not reliably separate ranks three and four, and if your six ideas are all mediocre it will tell you they are all mediocre, which is useful but not the answer you were hoping for.
Three ways to choose, compared honestly
All three are used by people who ship regularly. The third is what this site is.
| Approach | Build the fun one | Ask your audience | Publish a priced page for each |
|---|---|---|---|
| Time before you know anything | 12 weeks | 2 days | 6–8 weeks, mostly waiting |
| Cost per idea evaluated | A quarter | Almost nothing | About 8 hours |
| Who answers | Nobody. You find out from silence. | People who already like you | Strangers who searched for the problem |
| Signal quality | Definitive, and very late | Warm, and systematically inflated | Coarse, but from the right population |
| What it is good at | Keeping you shipping | Ranking ideas within a known niche | Finding the idea with buyers outside your circle |
| Main risk | A quarter spent on four users | Your audience is not the market | Not enough traffic to distinguish anything |
Time figures are from building this site. The eight hours per idea includes keyword research, the comparison numbers, and one substantive page beyond the landing.
How to run six of these at once without it becoming a job
The operational advice, which matters more than the method once you are past one idea.
- Publish all six within the same fortnightStaggered launches make the results incomparable. Traffic and your own energy both decay, and idea six will look worse than idea one for reasons that have nothing to do with the idea.
- Write one genuinely useful page per idea, not six landing pagesA calculator or a template with real numbers is what search sends people to. A landing page alone gets found by nobody and measures your ability to post links.
- Use the same form and the same tiers on all of themThe comparison is the whole point. Different forms produce different conversion rates and you will not be able to tell which difference is the idea.
- Set one deadline for all six and do not look until thenChecking daily converts a slow experiment into an anxiety loop, and the first week is the least representative one.
- Write the pass mark down for each, before publishingDifferent ideas can have different bars — a $99 tool needs fewer signups than a $9 one. Decide that in advance or you will decide it afterwards, in your own favour.
- Kill the bottom three publiclyEmail the people who signed up and say so. It costs an hour and it is the only reason anyone will believe the next set of pages.
The last item is the one that turns a batch of experiments into something with a reputation attached, which is the only durable asset in this.
What people in this position usually say
- I would rather spend those eight hours building the thing.
- Then build it — for one idea, the arithmetic genuinely favours just building. This only pays at three or more, where the eight hours buys you a ranking rather than a page. If you have one idea you cannot stop thinking about, validation is a procrastination device and you should ignore all of this.
- My audience is on one platform. I will just post there.
- Do that too, and do it first, because it is faster. What it cannot do is tell you about demand outside the audience, which is exactly the demand that determines whether the product outgrows a launch spike. Posts measure your reach; pages measure the problem.
- Six priced pages makes me look like I am spamming ideas.
- It does, if they are thin. The mitigation is not fewer ideas, it is that each one has to carry something real — a calculation, a comparison with today's prices, a template that works without the product. Six pages with substance read as a research programme. Six pages with a hero and a form read as exactly what you are worried about.
Fifteen ideas, published in one fortnight, all still unbuilt
This is the batch. The prices, the research notes, and the failure criteria are all public, including for the ones that are probably going to fail.