Back to blog
Business & Startups

Finding First Customers When You Have No Marketing Budget

Paul Graham's 'Do Things That Don't Scale': zero-budget acquisition is hand-built, specific, slow — and the thing first-time founders skip. Six moves (30 specific DMs, communities, trade work for testimonials, ask for 2 intros, build in public, partner-trade), the ADHD case, three failures.

Samuel Culman3 December 20256 min read

The short answer: do the things that don't scale

Paul Graham's 2013 essay Do Things That Don't Scale (source) is the central document on this question for early founders. The argument is that most early-stage businesses fail not because the founders couldn't get customers but because they wouldn't do the kind of customer acquisition that's available to them. Cold DMs, personal outreach, hand-delivering value, one-on-one onboardings — these are slow, unglamorous, and unscalable. They're also the dominant way the first ten and the first hundred customers actually arrive when there's no marketing budget. The plays that work later (paid ads, content engines, SEO compounding) do not work earlier. The plays that work earlier feel beneath the dignity of "running a business". Graham's data says do them anyway.

Why the standard "build great product, market it well" advice doesn't fit zero-budget

Most marketing advice quietly assumes that the marketing engine produces customers from a top-of-funnel pool whose first contact with you was passive — they saw an ad, read a piece of content, heard about you. That pool exists when you have a budget; it doesn't exist when you don't. The replacement, at zero budget, is direct acquisition: you reach out, by name, to specific humans you think might benefit, and convert them one at a time. The early-stage maths is that fifty real conversations produce more revenue than fifty thousand impressions. The same money — zero — buys you a much higher return on direct outreach than on broad attention.

Six zero-budget moves that actually find customers

  • DM thirty people you'd ideally serve. Specific, named, with a one-sentence offer that's about them and not about you. "I built X because of Y problem — would 15 minutes of feedback be worth your time in exchange for free early access?" Specific gets answered; generic doesn't. Thirty messages this week produces a baseline of either signal or null result; either is useful.

  • Hang out where your customers already are. A subreddit, a Slack community, a specific industry conference, a hobby forum. Be useful before you sell — answer questions, share work, link to good resources. The selling happens, when it happens, as a natural extension of the participation. Communities react fastest to good faith long-term presence and slowest to drive-by promotion.

  • Trade work for testimonials. The first five customers are paying you in two things — money and credibility. The credibility is sometimes worth more than the money. Discount or free pilots in exchange for written testimonials, named case studies, or referrals create the social proof the next ten customers need to convert.

  • Ask each customer for two introductions. After delivering value, the friction of asking for an introduction is much lower than first-time outreach. Two intros per happy customer compounds quickly — five customers become ten via intros, ten become twenty, and the growth is essentially free.

  • Build in public. Document what you're building, the problems you're solving, the customers you're learning from. The audience that accumulates is qualified — they're following because they care about the same problem. Building in public substitutes for the marketing function you can't afford by turning the work itself into the discovery surface.

  • Partner-trade with people in adjacent businesses. Someone with a small audience that overlaps with yours is exchange-able value. A trade — "I'll mention you to my list if you mention me to yours" — costs both sides nothing and produces a shared pool of potential customers. The standard playbook here is non-competing services that share a customer profile.

Why this pays double for ADHD founders

Two factors. First, ADHD-energy is well-matched to the kind of intense, novel, varied work that early-stage acquisition requires — every conversation is different, every prospect is a new puzzle, the feedback loop is short. This is the part of building a business that ADHD brains tend to enjoy. Second, the standard alternative — running ad campaigns, optimising conversion funnels, building content engines — is the part that ADHD brains find brutally tedious. The zero-budget play is also the play that suits the founder profile, which is part of why early-stage scrappy growth disproportionately rewards ADHD founders even though they tend to struggle later when scaling demands operational consistency.

Where it fails (and the repair)

  • Too cold, too template. Cold outreach with templates is detected immediately and ignored. The whole leverage is specificity — name the person, name a thing they did, name a problem you think you've solved for them. If you can't write a different message for each recipient, you're sending the wrong message.

  • Confusing outreach with desperation. There's a desperate version of zero-budget acquisition that hurts the brand and the founder. The cure is to put a bar on yourself — a fixed number of outreaches per week, specific people, with a calm specific value proposition. The structure prevents the panic-spam mode that backfires.

  • Skipping the post-sale work. The customer you sold to last month is the source of three things this month: revenue, testimonials, and intros. Founders chasing only new outbound miss that the existing customers are the highest-yield outreach pool. Stay in touch; ask for the intro; trade work for case study.

FAQ

How many DMs is too many?

Thirty a week, written specifically, is sustainable for most founders. A hundred a week, copy-pasted, damages reputation faster than it converts. The constraint isn't the number — it's the specificity. If you can write thirty genuinely different, specific messages, send thirty; if you can only write five genuine ones, send five.

What if my product needs marketing budget to land?

Then the first ten customers are how you discover whether it does or doesn't. If you can't get ten through hand-built channels, no amount of paid acquisition will scale a broken-fit product. Use the zero-budget phase to test product-market fit cheaply; only spend money on broad acquisition once the fit is proven through direct contact.

Isn't this just a lot of unpaid work?

It is. Graham's central point: the unpaid work that produces the first customers is also the work that produces the most learning, the deepest understanding of the market, and the relationships that compound. It is genuinely cheaper than marketing budget would be — but it costs founder time. The trade is correct for almost every early-stage business: founder time is more available than money, and founder time spent talking to early customers is the most leveraged version of that time.

When do I stop and switch to scaled marketing?

When your hand-built acquisition is producing real demand you can't keep up with one customer at a time, and you have specific evidence of why a marketing channel will scale the same demand cheaper. "Switching to ads" before having either of those signals is the most common money-burning mistake at the first-customer stage.

What if I'm introverted and DMing strangers is excruciating?

Use the warm version first — your existing weak ties, in the Granovetter sense. Former colleagues, classmates, professional acquaintances. The warm DM is much less psychologically expensive and tends to convert better. The truly cold DM is a real lever but it's the last lever, not the first; exhaust the weak-tie pool before going to strangers.

Frequently asked questions

How many DMs is too many?
Thirty a week, written specifically, is sustainable. Hundred a week copy-pasted damages reputation faster than it converts. The constraint isn't number — it's specificity. Write thirty genuinely different specifics or send five real ones.
What if my product needs marketing budget to land?
First ten customers tell you whether it does. If you can't get ten through hand-built channels, no paid acquisition will scale a broken-fit product. Use zero-budget phase to test fit cheaply; spend on broad acquisition only after fit is proven via direct contact.
Isn't this just a lot of unpaid work?
Yes. Graham's central point: the unpaid work that produces first customers also produces the most learning, the deepest market understanding, and relationships that compound. Founder time is more available than money; founder time talking to early customers is the most leveraged version.
When do I switch to scaled marketing?
When hand-built acquisition produces real demand you can't keep up with one customer at a time, AND you have specific evidence a marketing channel will scale that demand cheaper. 'Switch to ads' before either signal is the most common money-burning mistake.
What if I'm introverted and cold DMs are excruciating?
Warm version first — existing weak ties (Granovetter). Former colleagues, classmates, professional acquaintances. The warm DM is psychologically much cheaper and converts better. The truly cold DM is the last lever; exhaust the weak-tie pool first.
Share:

Like what you're reading?

Try the platform built around the same ideas — 14 days free.

Start free trial

Read also