A new business does not have a marketing strategy problem. It has a sequencing problem. Almost everything on the standard launch checklist is defensible on its own, which is exactly why founders do it all at once, badly, and run out of money before learning anything.
Here is the order the decisions actually come up, and where each one can wait.
1. Decide who you are for
This is the only decision that changes every other one, so it comes before the name, the logo, the site, and the first dollar of spend.
Pick narrowly enough that it feels uncomfortable. Not “small businesses” but “owner-operated HVAC companies in New England running two to six trucks.” The narrow version tells you which words to use, where to show up, what to charge, and who to call. The broad version tells you nothing, which is why it feels safe.
You will be wrong about some of it. That is fine, and it is cheap to correct in month three. What is expensive is having no position at all, because then nothing you learn contradicts anything.
Write it down in one sentence: who it is for, what it replaces, and why someone would switch. If you cannot finish that sentence, you do not have a marketing problem yet — you have a product question to go answer.
2. Build the minimum viable brand
Three things, and no more:
- A name you will not change. Checked against the trademark register, available as a domain you can actually type over the phone, and not a misspelling of a real word.
- One line of positioning you can say out loud without hedging. This does the work people think a tagline does.
- Somewhere to send people. One page, a working email address at your own domain, and whichever single profile your buyers actually check.
Alongside that: a wordmark, two or three colours, one typeface you will not deviate from. A designer can do this in days and it will look deliberate, which is all it needs to do in year one.
What can wait: the brand architecture workshop, the archetype deck, the tone-of-voice guidelines, the photography system, the brand book. Those exercises are genuinely useful when there is a team to keep consistent and a market position to protect. A founder with eleven customers is both of those things already.
3. Find the first customers by hand
The first ten customers come from conversations. Not campaigns, not content, not ads.
Make a list of named companies or people who match the sentence you wrote in step one. Not a persona — a list with real names on it. Then work it: your own network first, then the partners and suppliers who already serve those buyers, then the communities they sit in, then direct outreach. Expect low response rates and keep going, because the conversations are the point even when they do not convert.
Record what you hear, in their words. Which problem they led with, what they are using now, what made them hesitate, what phrase they used that you would never have written. That transcript becomes your website copy, your ad copy, and your objection handling later. Nothing you buy in year one gives you better input than this, and it costs you time rather than money.
Two things to set up while you do it, because they are nearly free and they decide whether any of it compounds: a single place where every inquiry lands, and a decision about who replies and how fast. Speed of first reply is a thing a one-person business can genuinely win on.
4. Build a real site when you know the questions
Until then, one page. What you do, who it is for, what it costs or roughly what it costs, proof if you have any, and one way to start a conversation. One page you edit every week as you learn beats a seven-page site that freezes a guess in place.
The signal that it is time to build properly is specific. You can name the three objections that come up in every call. You can name the two or three buyer types that need separating because they want different things. You know which questions people ask before they will talk to you. At that point the site has a structure to express, and a custom website is worth real money rather than being a design exercise.
Before you build anything, check that you can measure it. Analytics installed, a form submission you can see land in an inbox, and the two numbers roughly agreeing. Tracking that silently fails is worse than none, because it still gets used to make decisions.
5. Consider paid acquisition last
Paid media is an amplifier. It scales a message that already converts; it does not find one for you. Run it before you have that message and you are paying per click for information a dozen phone calls would have given you for free.
The honest entry conditions are: you know what a customer is worth to you, you have a page that converts conversations you already had, you can afford to run long enough to read a result rather than a daily fluctuation, and you have decided in advance what cost per qualified lead would make you continue.
When you do start, start where demand already exists — people searching for what you sell — rather than where demand must be created. Capture reads faster than creation, and you need a readable signal more than you need reach. One channel, funded properly, for a fixed period, with the read date written down before you launch. Our post on how to spend a small marketing budget, in order goes further into that sequence.
What first-year founders waste money on
Patterns worth naming, because each one feels like progress:
- A full identity package before there are customers. Months of revisions on something that will be revised again once you know who you are for.
- A multi-page site built to a structure nobody has validated. Usually followed by a rebuild inside eighteen months.
- Ads run to learn whether the offer works. The most expensive possible way to find out.
- Tools bought ahead of the process. A CRM, a marketing automation platform, and a scheduler, for a pipeline that fits on one screen.
- Retainers that consume the working budget. If the fee leaves nothing to spend on the actual work, the arrangement cannot succeed.
- A second channel started because the first one is not working. Two half-funded channels teach you less than one.
- Senior marketing leadership too early. A fractional CMO is for a company with revenue, a team, and budget to direct. In year one that judgement should be yours, because you are the one in the conversations.
The next thing to do
Write the sentence from step one, then take it to five people who match it and ask what they are using now. If the sentence survives those five conversations, you have something to build on. If it does not, you just saved yourself a logo.
When there is revenue and a team to point in a direction, talk to a CMO and we will look at what the numbers are telling you.
