Online Marketing for Small Business: The Painful 18-Month Lesson

I spent $40K on tactics that didn't work before finding the three that did. Here's the post-mortem, in plain terms, before you make the same mistakes.

The hard truth about online marketing for small business is that most of it doesn't work for most businesses. The tactics that do work depend on what you sell, who you sell to, and how often. I spent 18 months and ~$40K finding this out. Here's the version that would have saved me the money.

What didn't work for my business

Meta Ads. We're a service business with a slow consideration cycle. Meta excels at impulse-buy products. We spent $14K and got 9 leads that closed at low value.

SEO content at scale. We published 80 articles in 6 months. Traffic grew. Conversions did not. The articles ranked for keywords that didn't match buyer intent.

Influencer partnerships in our niche. The metrics looked great on paper. None of the followers converted.

What actually worked

Google Ads for high-intent commercial keywords. Boring. Expensive per click. But the intent was right. Spent $3K, got 14 customers at 4x lifetime value over cost. The bare minimum for a service business — and the place I should have started.

Email to my existing customers. A monthly newsletter that wasn't an ad. Stories, lessons, links. The retention impact was the highest-ROI thing I did all year, and I almost cut it because the open rates were modest.

Showing up consistently in one place where my buyers were already congregating. A single industry community where I posted twice a week. Took 4 months to see results. Continues to be my best lead source.

What I'd skip if starting over

Anything with "automation" or "funnel" in the marketing copy. The truth is most automation tools are over-engineered for businesses that don't yet have a working acquisition channel.

Courses that promise to teach "the algorithm." Algorithms change quarterly. The fundamentals don't.

Hiring a freelancer to "run your ads" before you understand the basics yourself. You'll pay for their tuition and have no ability to evaluate their work.

The infrastructure that supported the work

A standing desk for the long sessions of analytics review and email writing. Noise cancelling headphones for the deep-work blocks. A mechanical keyboard — I write 2,000 words a week for the newsletter alone. Deep Work by Cal Newport is the book on protecting the focus this takes.

The honest answer

Most online marketing advice is written for ecommerce by people selling to ecommerce. If you have a service business, a high-consideration product, or a B2B sale, almost none of the standard playbook applies. Find one channel where your actual customers exist, show up there for a year, and ignore everything else.