# How Do I Market a Startup? The Honest Playbook

> Source: [https://botensten.com/articles/how-do-i-market-a-startup](https://botensten.com/articles/how-do-i-market-a-startup) (canonical)
> Author: Botensten — Botensten, https://botensten.com
> Published: 2026-08-22

## TL;DR

To market a startup, define one specific customer and the problem they already try to solve. Validate that demand with direct conversations and a small offer before spending on ads. Then pick a single channel — founder-led content, community, cold outreach, or search — and run it consistently for 90 days while tracking one metric. Use a clear message where the customer is the hero and you are the guide. Add channels only after the first one reliably produces customers. Focus beats budget for early startups.

Market a startup by winning one narrow audience on one channel before scaling, because [CB Insights' 2021 analysis of startup post-mortems](https://www.cbinsights.com/research/startup-failure-reasons-top/) found 35% of startups fail from "no market need." Validate demand with real conversations and a small paid offer first. Lead with founder-led content and outreach, not paid ads. Own one channel, measure it weekly, then add a second.

## What should I do before spending a dollar on marketing?

Define one customer and confirm they will pay before you spend on marketing. Talk to at least 20 potential buyers and sell a small offer to prove demand.

Guessing is the expensive part. Most early spend is wasted because the message is vague or the audience is wrong. Write down who the customer is, the problem they already pay to solve, and the exact words they use for it.

A quick pre-marketing checklist:

1. Name one narrow audience (not "small businesses" — "solo SaaS founders doing their own support").
2. Interview 20 of them and record the words they use.
3. Make one small offer and get 3-5 people to pay or pre-commit.
4. Write a one-sentence message: what you do, for whom, and the result.

If nobody pays a small amount, no ad budget will fix that. Fix the offer first.

## Which marketing channels actually work for a startup?

The channels that work early are the ones you can run yourself, consistently, for free or cheap: founder content, direct outreach, community, and search. Paid ads work later, once you know your message converts.

Pick one primary channel based on where your customer already spends attention. Running five channels badly beats nothing, but running one channel well beats all five. Here is how the main options compare for a solo or small founding team.

| Channel | Cost to start | Time to first result | Best for |
|---|---|---|---|
| Founder content (build-in-public) | Free | 4-12 weeks | Trust, inbound, hiring |
| Direct/cold outreach | Free-low | 1-2 weeks | B2B, first 10 customers |
| Community participation | Free | 2-6 weeks | Niche, high-trust markets |
| Search / SEO + AEO | Low | 3-6 months | Compound long-term demand |
| Paid ads | Medium-high | Days | Scaling a proven message |

Paul Graham's essay [Do Things That Don't Scale](https://www.paulgraham.com/ds.html) makes the case that founders should manually recruit early users by hand. That unscalable work teaches you the message you will later automate.

## How we market our own software in public

We market Botensten by shipping software with AI in the open and writing about exactly what happened — the build, the cost, the thing that broke. Build-in-public is our primary channel because it doubles as product development.

Here is the real trade-off we hit. Early on we tried to write polished "thought leadership" and it flopped — no replies, no signups. When we switched to posting the actual numbers (what a feature cost to build, how long it took, the bug that took down a service), engagement and inbound both climbed. Specific beats smart.

The mechanism is simple: a concrete story is memorable, and memorable ideas spread. Chip and Dan Heath call this "concreteness" in *Made to Stick* — vague claims slide off, specific ones stick. So we never write "we boost productivity." We write "this feature took two days and $40 in model costs."

What we changed after six months: we stopped chasing viral posts and built a boring weekly cadence instead. One build log, one lesson, one call to action. Consistency out-earned cleverness every time.

## Why does most startup marketing fail?

Most startup marketing fails because founders scale a message that was never validated, or they make themselves the hero instead of the customer. The fix is a clear message and proof of demand before spend.

Donald Miller's [Building a StoryBrand](https://storybrand.com) framework is blunt about this: the customer is the hero, and the company is the guide. When your homepage talks about your features instead of the customer's problem, people leave confused. Confused visitors do not buy.

The second failure is impatience. Marketing compounds, and founders quit a working channel at week three because it did not explode. Seth Godin argues in [This Is Marketing](https://seths.blog) that marketing is the generous, patient act of solving a specific person's problem — not a one-time launch.

Common failure patterns to avoid:

- Talking about yourself instead of the customer's problem.
- Running paid ads before organic proof of message-market fit.
- Switching channels every two weeks with no data.
- Targeting "everyone," which reaches no one.

## How do I build a repeatable marketing system?

Build a repeatable system by choosing one channel, one weekly cadence, and one metric, then reviewing it every week for a full quarter. Systems beat bursts because they compound.

Allan Dib's *1-Page Marketing Plan* is a good template: define your target, your message, your channel, and how you follow up — on a single page. Keep it that short so you actually use it.

A simple 90-day operating loop:

1. Pick one primary channel and one weekly action (for example, two build-log posts).
2. Track one metric that maps to money (signups, demos booked, replies).
3. Review weekly: keep what moved the metric, cut what did not.
4. After 90 days of a channel producing customers, add a second channel.

Do not add channel two until channel one is boring and reliable. That discipline is what turns marketing from luck into a machine.

## Related reading

- [How Do I Get My First Customers for a Startup?](/articles/get-first-customers-for-a-startup)
- [How Do I Become a Solopreneur? The Honest 90-Day Answer](/articles/how-do-i-become-a-solopreneur)
- [How Can I Use AI to Start a Business? The Honest Answer](/articles/how-to-use-ai-to-start-a-business)

## Frequently asked questions

**How do I market a startup?**

Validate one narrow audience with real conversations and a small offer, then run a single channel consistently for 90 days while tracking one money-linked metric. Add channels only after the first reliably produces customers.

**How do I market a startup with no money?**

Use free founder-led channels: direct outreach to the first customers, build-in-public content, and active participation in the communities your buyers already use. These cost time, not cash.

**How long before startup marketing works?**

Direct outreach can produce results in 1-2 weeks; founder content typically takes 4-12 weeks; SEO takes 3-6 months. Commit to a channel for at least 90 days before judging it.

**Should a startup run paid ads first?**

No. Prove your message converts organically first. Paid ads scale a working message quickly but also scale a broken one, wasting budget before you know what resonates.

**What is the biggest startup marketing mistake?**

Making the company the hero instead of the customer. Lead with the customer's problem and desired outcome; position your product as the guide that helps them win.

**How many marketing channels should a startup use?**

Start with one. Master a single channel until it reliably produces customers, then add a second. Running one channel well beats running five badly.

**What metric should a startup track for marketing?**

Track one metric that maps directly to money, such as qualified signups, demos booked, or replies from outreach. Review it weekly and cut what does not move it.