The Zero-to-One Marketing Playbook for Startups
When you're building from scratch, you don't have a playbook. This is the one: how to define your ICP, pick your first channel, build your brand positioning, and launch your first demand-generation motion without wasting money.
Nikhil Kalanjee
3/23/20266 min read
You've built something. Maybe it works. Maybe people are starting to pay for it. But you're staring at a blank canvas when it comes to marketing, and the internet is full of advice written by people who've never actually had to make a startup work with no budget, no team, and no margin for error.
I've been in that room. I've sat across from founders who've got six weeks of runway and a product that technically works but hasn't found its people yet. I've also been inside HP when we had more budget than ideas and still managed to get marketing wrong. The lessons from both ends of that spectrum are surprisingly similar.
Here's what I've learned about taking a company from zero to one, from the marketing seat.
Forget Marketing. Start With the Question.
The biggest mistake I see founders make isn't picking the wrong channel or writing bad copy. It's starting with marketing at all.
Before you spend a single pound on ads, content, or even a logo refresh, you need to be able to answer one question with absolute clarity: Who is this for, and why should they care right now?
Not "who could this be for." Not "who would benefit from this eventually." Who is actively feeling the pain that your product solves, today, and what are they doing about it right now?
I once worked with a well-funded scale-up where we ran a simple exercise: we asked the board and leadership team to independently describe the ICP. Eight senior leaders came back with eight completely different answers. Eight. This wasn't a cash-strapped startup. They had real budget. But they were spreading it across eight different prospect profiles, and wondering why nothing was sticking. Once we aligned the entire organisation around a single ICP, every marketing decision got simpler, faster, and cheaper.
The ICP Is Not a Spreadsheet Exercise
I know "Ideal Customer Profile" sounds like something a consultant invented to charge you for a workshop. But the ICP isn't a demographic breakdown. It's a thesis about who will pay, stay, and tell others.
Here's how I work through it with founders:
Step 1: Find the desperation. Who needs this so badly they'd use it even if the UX was terrible? Those are your first customers. Not the "nice to have" market. The "can't live without" market.
Step 2: Talk to them. Not surveys. Conversations. I tell every founder I work with: you need 20 conversations before you write a single marketing brief. Not 20 responses to a Google Form. Twenty real, messy, surprising conversations where you shut up and listen.
Step 3: Look for the pattern, not the persona. Personas are fiction. Patterns are data. After 20 conversations, you'll notice that the people who light up about your product share two or three characteristics. Maybe they're all Series A SaaS companies who just lost their first head of marketing. Maybe they're DTC brands doing £1M revenue who can't figure out why their CAC keeps climbing. That's your ICP.
The Five Whys of Customer Clarity
In those conversations, I always come back to five questions. I call them the Five Whys of Customer Clarity:
1. Why is the customer facing this challenge? What's the root cause of their pain, not just the surface symptom?
2. Why is your product or solution the right one for this customer, given everything else available to them?
3. Why is your product the right one against the competition? What makes you genuinely different, not just "better"?
4. Why is now the right time to buy? And what happens if the customer delays? If there's no urgency, there's no deal.
5. Why is this great value? Ideally backed by customer feedback, whether positive or negative. Social proof isn't optional at zero-to-one. It's everything.
If you can't answer all five with confidence, you're not ready to market. You're ready to keep learning.
The First 90 Days: Do Things That Don't Scale
Paul Graham said it. Everyone quotes it. Almost nobody does it.
In the first 90 days of marketing a startup, your job is not to build a funnel. It's to learn what works by doing things manually, messily, and personally.
Days 1-30: Listen and map. Where do your ICP people actually spend their time? What do they read? Who do they trust? What communities are they in? Don't guess. Go look. Join the Slack groups. Read the subreddits. Sit in the LinkedIn comment sections.
Days 31-60: Show up and add value. Start contributing to the places where your ICP hangs out. Not pitching. Contributing. Answer questions. Share useful frameworks. Be the person who adds signal to noisy channels. This is how you build trust before you have a brand.
Days 61-90: Run small experiments. Now, and only now, test some marketing. But small. A LinkedIn post series testing different angles. A cold email to 50 people with a specific value prop. A landing page with one CTA. Measure what resonates, not what converts (yet).
One of the best CMOs I ever worked for had a saying: "invest in the miles." It stuck with me because it captures something most marketers skip. Do whatever you can to get in front of the customer. Engage with them. See and hear first-hand what the challenges are and how what you're bringing to market actually helps.
In some cases, you can't get directly to the customer. Fine. Take the next best option. Talk to sales. Talk to resellers. Talk to retailers. Talk to anybody you can. And in the absence of being able to have a conversation, look for other signals: forums, review sites, anywhere the customer has shared their unfiltered thoughts.
Obsessively try to understand who you are going after and how you're going to help them. That's the job.
The Channels That Actually Work at Zero-to-One
Here's what I tell every founder who asks "should I be on TikTok?"
At zero-to-one, your channel is wherever your ICP already pays attention. There is no universal answer, and anyone selling you a channel playbook is selling you their channel playbook, not yours.
That said, I see patterns:
For B2B: LinkedIn content plus warm outreach plus community participation. The founder's personal LinkedIn almost always outperforms the company page. Every time. Don't spend money on the company page yet.
For B2C / DTC: One owned channel (email or SMS) plus one discovery channel (the social platform where your customer actually browses). For most DTC brands in 2026, that's still Instagram or TikTok, but test, don't assume. Algorithms change constantly.
For both: Content that teaches, not content that sells. The startup that helps its audience solve a problem, even a small one, before asking for money will always outperform the one that leads with a product pitch.
But before you pick any channel, understand your business model. Understand your cost of acquisition and lifetime value. In some cases you may break even on the first sale but make money on future purchases. In others, you need margin on the first transaction. It sounds basic, but there are so many founders and marketers who are unclear about how their business is going to make money and what they're trying to achieve. If you don't know the economics, you can't evaluate the channel.
When to Hire (And When Not To)
This connects directly to the fractional model, but it's worth saying here: do not hire a full marketing team at zero-to-one.
Hire a fractional strategist or advisor who can help you build the playbook and test it. Then hire executors, a content person, a demand gen person, once you know what's working. Hiring a VP of Marketing or a CMO before you've validated your GTM motion is like hiring an architect before you've bought the land.
And here's a pro tip: if your entire marketing budget is the salary of your CMO or VP, you're making a mistake. Ensure you support your marketing leader by giving them either the people, the tools, or the budget to actually execute. Otherwise you end up with someone who gives you great advice but can't act on any of it. You'll both be disappointed.
The Playbook, Condensed
If I had to hand a founder a single page before they started marketing:
Nail the ICP first. 20 conversations minimum. No shortcuts.
Map the territory. Where does your ICP spend time, attention, and trust?
Show up before you sell. 30 days of contributing value before you pitch anything.
Run small experiments. Test three or four angles with minimal spend. Measure resonance.
Double down on what works. Kill the rest. Reallocate ruthlessly.
Don't hire ahead of your knowledge. Fractional first, full-time later.
The whole point of "Time to First Bite" is this: how quickly can you get from idea to first meaningful customer interaction? Not first impression. Not first click. First bite, the moment someone engages with real intent.
That's what marketing at zero-to-one is about. Not reach. Not awareness. Not brand. Just: did someone care enough to take a bite?
You've built something. Maybe it works. Maybe people are starting to pay for it. But you're staring at a blank canvas when it comes to marketing, and the internet is full of advice written by people who've never actually had to make a startup work with no budget, no team, and no margin for error.
I've been in that room. I've sat across from founders who've got six weeks of runway and a product that technically works but hasn't found its people yet. I've also been inside HP when we had more budget than ideas and still managed to get marketing wrong. The lessons from both ends of that spectrum are surprisingly similar.
Here's what I've learned about taking a company from zero to one, from the marketing seat.
Forget Marketing. Start With the Question.
The biggest mistake I see founders make isn't picking the wrong channel or writing bad copy. It's starting with marketing at all.
Before you spend a single pound on ads, content, or even a logo refresh, you need to be able to answer one question with absolute clarity: Who is this for, and why should they care right now?
Not "who could this be for." Not "who would benefit from this eventually." Who is actively feeling the pain that your product solves, today, and what are they doing about it right now?
I once worked with a well-funded scale-up where we ran a simple exercise: we asked the board and leadership team to independently describe the ICP. Eight senior leaders came back with eight completely different answers. Eight. This wasn't a cash-strapped startup. They had real budget. But they were spreading it across eight different prospect profiles, and wondering why nothing was sticking. Once we aligned the entire organisation around a single ICP, every marketing decision got simpler, faster, and cheaper.
The ICP Is Not a Spreadsheet Exercise
I know "Ideal Customer Profile" sounds like something a consultant invented to charge you for a workshop. But the ICP isn't a demographic breakdown. It's a thesis about who will pay, stay, and tell others.
Here's how I work through it with founders:
Step 1: Find the desperation. Who needs this so badly they'd use it even if the UX was terrible? Those are your first customers. Not the "nice to have" market. The "can't live without" market.
Step 2: Talk to them. Not surveys. Conversations. I tell every founder I work with: you need 20 conversations before you write a single marketing brief. Not 20 responses to a Google Form. Twenty real, messy, surprising conversations where you shut up and listen.
Step 3: Look for the pattern, not the persona. Personas are fiction. Patterns are data. After 20 conversations, you'll notice that the people who light up about your product share two or three characteristics. Maybe they're all Series A SaaS companies who just lost their first head of marketing. Maybe they're DTC brands doing £1M revenue who can't figure out why their CAC keeps climbing. That's your ICP.
The Five Whys of Customer Clarity
In those conversations, I always come back to five questions. I call them the Five Whys of Customer Clarity:
1. Why is the customer facing this challenge? What's the root cause of their pain, not just the surface symptom?
2. Why is your product or solution the right one for this customer, given everything else available to them?
3. Why is your product the right one against the competition? What makes you genuinely different, not just "better"?
4. Why is now the right time to buy? And what happens if the customer delays? If there's no urgency, there's no deal.
5. Why is this great value? Ideally backed by customer feedback, whether positive or negative. Social proof isn't optional at zero-to-one. It's everything.
If you can't answer all five with confidence, you're not ready to market. You're ready to keep learning.
The First 90 Days: Do Things That Don't Scale
Paul Graham said it. Everyone quotes it. Almost nobody does it.
In the first 90 days of marketing a startup, your job is not to build a funnel. It's to learn what works by doing things manually, messily, and personally.
Days 1-30: Listen and map. Where do your ICP people actually spend their time? What do they read? Who do they trust? What communities are they in? Don't guess. Go look. Join the Slack groups. Read the subreddits. Sit in the LinkedIn comment sections.
Days 31-60: Show up and add value. Start contributing to the places where your ICP hangs out. Not pitching. Contributing. Answer questions. Share useful frameworks. Be the person who adds signal to noisy channels. This is how you build trust before you have a brand.
Days 61-90: Run small experiments. Now, and only now, test some marketing. But small. A LinkedIn post series testing different angles. A cold email to 50 people with a specific value prop. A landing page with one CTA. Measure what resonates, not what converts (yet).
One of the best CMOs I ever worked for had a saying: "invest in the miles." It stuck with me because it captures something most marketers skip. Do whatever you can to get in front of the customer. Engage with them. See and hear first-hand what the challenges are and how what you're bringing to market actually helps.
In some cases, you can't get directly to the customer. Fine. Take the next best option. Talk to sales. Talk to resellers. Talk to retailers. Talk to anybody you can. And in the absence of being able to have a conversation, look for other signals: forums, review sites, anywhere the customer has shared their unfiltered thoughts.
Obsessively try to understand who you are going after and how you're going to help them. That's the job.
The Channels That Actually Work at Zero-to-One
Here's what I tell every founder who asks "should I be on TikTok?"
At zero-to-one, your channel is wherever your ICP already pays attention. There is no universal answer, and anyone selling you a channel playbook is selling you their channel playbook, not yours.
That said, I see patterns:
For B2B: LinkedIn content plus warm outreach plus community participation. The founder's personal LinkedIn almost always outperforms the company page. Every time. Don't spend money on the company page yet.
For B2C / DTC: One owned channel (email or SMS) plus one discovery channel (the social platform where your customer actually browses). For most DTC brands in 2026, that's still Instagram or TikTok, but test, don't assume. Algorithms change constantly.
For both: Content that teaches, not content that sells. The startup that helps its audience solve a problem, even a small one, before asking for money will always outperform the one that leads with a product pitch.
But before you pick any channel, understand your business model. Understand your cost of acquisition and lifetime value. In some cases you may break even on the first sale but make money on future purchases. In others, you need margin on the first transaction. It sounds basic, but there are so many founders and marketers who are unclear about how their business is going to make money and what they're trying to achieve. If you don't know the economics, you can't evaluate the channel.
When to Hire (And When Not To)
This connects directly to the fractional model, but it's worth saying here: do not hire a full marketing team at zero-to-one.
Hire a fractional strategist or advisor who can help you build the playbook and test it. Then hire executors, a content person, a demand gen person, once you know what's working. Hiring a VP of Marketing or a CMO before you've validated your GTM motion is like hiring an architect before you've bought the land.
And here's a pro tip: if your entire marketing budget is the salary of your CMO or VP, you're making a mistake. Ensure you support your marketing leader by giving them either the people, the tools, or the budget to actually execute. Otherwise you end up with someone who gives you great advice but can't act on any of it. You'll both be disappointed.
The Playbook, Condensed
If I had to hand a founder a single page before they started marketing:
Nail the ICP first. 20 conversations minimum. No shortcuts.
Map the territory. Where does your ICP spend time, attention, and trust?
Show up before you sell. 30 days of contributing value before you pitch anything.
Run small experiments. Test three or four angles with minimal spend. Measure resonance.
Double down on what works. Kill the rest. Reallocate ruthlessly.
Don't hire ahead of your knowledge. Fractional first, full-time later.
The whole point of "Time to First Bite" is this: how quickly can you get from idea to first meaningful customer interaction? Not first impression. Not first click. First bite, the moment someone engages with real intent.
That's what marketing at zero-to-one is about. Not reach. Not awareness. Not brand. Just: did someone care enough to take a bite?