With almost no budget, the binding constraint is not reach, it is focus. Nearly every startup that wastes its first year does so by doing eight things adequately instead of one thing properly.
Key takeaways
- Find one channel that works before adding a second. Diversifying too early is how small teams stall.
- Talk to twenty potential customers before spending anything. It is free and it changes what you build.
- Do the things that do not scale first. They are the only ones you can afford and they teach you the most.
- Set a single number that defines the next quarter, and let it kill everything that does not serve it.
On this page
The work that comes before marketing
If you cannot state who this is for and what changes for them, no amount of marketing will help. Vague positioning is not a communication problem you can solve with better copy; it is a strategy problem that copy will faithfully transmit.
Twenty conversations with people who have the problem will tell you more than any amount of desk research. Ask what they do today, what it costs them, what they have already tried and why they stopped. Do not pitch. You are looking for the words they use, which are almost never the words you would have chosen.
The fastest marketing win available to most startups is describing the product the way its customers already describe the problem.
Serhii Yelbaiev, Well Web Marketing
One channel, properly
Pick one acquisition channel and work it until it either produces customers repeatably or clearly will not. Running five channels on a tenth of the necessary effort produces five inconclusive results and no learning.
| Channel | Works when | Real cost |
|---|---|---|
| Direct outreach | You know exactly who the buyer is | Your time, daily, for months |
| Search content | People already search for the problem | Months before anything compounds |
| Communities | Your buyers gather somewhere specific | Genuine participation, not posting |
| Partnerships | Someone already has your audience | Relationship building, slow |
| Paid search | Demand exists and you know unit economics | Money you probably do not have yet |
Notice that four of the five cost time rather than money. That is the trade a startup is making, and it is why founder attention is the scarcest resource in the table.
Do the things that do not scale
Manual, unscalable effort is not a compromise at this stage. It is the correct strategy. It gets you customers you could not otherwise afford, and it teaches you things automation would have hidden.
- Email people individually, having actually read about them.
- Onboard your first customers personally, by video call, watching where they get stuck.
- Write the first help articles yourself, based on the questions those calls produce.
- Show up in the places your buyers already gather and be useful without pitching.
- Ask every happy customer, directly, who else has this problem.
The temptation is to build systems first because systems feel like progress. But a system that automates a process you have not yet validated just produces the wrong outcome faster and at greater expense.
The smallest useful website
You need one page that converts, not twelve that describe. A single clear page beats a full site of vague ones, and it can be built in a day.
It needs a headline naming the problem in the customer’s language, a plain explanation of what happens if they buy, evidence that it works, even a single named customer, and one action to take. Everything else can wait until people are actually arriving.
Get analytics and a conversion event working on day one. The cost of adding measurement later is not the setup; it is the three months of behaviour you cannot see and will never recover.
Keeping focus when everything looks urgent
Choose one number for the quarter and use it to refuse things. Its main value is not motivation, it is having a defensible reason to say no.
- Pick the single metric that most needs to move: paying customers, qualified conversations, weekly active users.
- Write down the two activities most likely to move it.
- Decline everything else for twelve weeks, explicitly and in writing.
- Review at the end of the quarter, not weekly.
- If the number did not move, change the activity, not the number.
When marketing becomes the constraint but a full hire is not yet justified, our monthly marketing team is built for that gap, the execution capacity of a team without the cost of assembling one.
Frequently asked questions
Before product-market fit, as little money and as much founder time as possible. Paid channels are for amplifying something that already works, not for discovering whether it does.
Neither, initially. Founders should run marketing until they understand which channel works, because that knowledge is what makes any later hire effective rather than expensive.
Outreach can produce conversations within weeks. Content and search take six to twelve months. Choose according to how long your runway allows you to wait.
Use whatever evidence you honestly have: a pilot, a beta user, your own results, or the founder’s background. Do not invent testimonials, the credibility cost when it emerges is fatal at this size.
Sources and method
Based on our own work with early-stage companies and on Well Web Marketing’s own launch. Last reviewed February 2026.