When you’re bootstrapping a solo AI startup, everyone has an opinion.

Freemium. Waitlists. Subscriptions. Build in public. Raise capital.

These so-called “best practices” get repeated like gospel. But here’s what I’ve learned the hard way: most of them aren’t rules — they’re just suggestions that worked once for someone else.

Some advice is useful. Most of it is conditional. And a lot of it? Just noise.

I don’t ignore all of it. But I’ve learned when to break the rules — and why it can actually work better when you do.

1. I skipped freemium. Here’s why I don’t regret it.

Freemium is almost gospel in SaaS: show your value before asking for payment, let users try it, love it and convert. It’s a model that works — until it doesn’t.

For PhotoPacks.AI, my startup that creates high-quality AI headshots, freemium wasn’t just risky — it was financially impossible. Every photo upload requires training a model on expensive GPUs, costing me $3 to $4 per user. With thousands of browsers likely to take a single free headshot and never return, offering a free sample was a fast track to bankruptcy.

Instead, I relied on social proof. I charged upfront, kept prices low at launch and focused on a seamless experience. While it turned some people away, the customers who stayed were more committed — and those were exactly the ones I wanted.

Related: 3 Essential Skills I Learned By Growing My Business From the Ground Up

2. I raised prices — and my customers got better

At first, I priced it at $9.99 — cheap, probably too cheap. I wanted traction, but I noticed a pattern: most support requests came from the lowest-paying customers. More refunds, more complaints, more time wasted.

So I doubled the price — and then raised it again. I lost about 10% of my volume, but refund requests dropped sharply. My average customer was more pleasant, emails were friendlier and I made more money with less effort.

The lesson? Pricing isn’t just about revenue. It’s a filter for the kind of customers you want to serve.

3. I ignored the hype machine

I also skipped the waitlist. Collecting emails doesn’t guarantee real users. Validation can be a false high — people often click “Join Waitlist” out of curiosity, not intent. I preferred building an MVP quickly and getting genuine feedback over chasing vanity metrics.

The same goes for social media. I tried TikTok, Facebook and Instagram, but none drove meaningful traction. What worked instead was organic search, word-of-mouth, and direct conversations with customers. So I focused on those and stopped worrying about the rest.

Related: 5 Things You Need to Stop Doing as a Solopreneur

Not all rules are useless — just overrated

There’s one rule I still live by: the golden rule. Treat others how you want to be treated. That shows up in how I handle support, how I price fairly, and how I prioritize features. It’s not revolutionary. But it works.

And here’s the bigger point: best practices are just averages. They’re designed for the “typical” startup — the one with a team, capital and time. If you’re a solo founder building something weird, niche or scrappy? You’re the exception. And exceptions make their own rules.

So learn the playbook — then feel free to throw it out.