Niche business — pretty word, more often leads to red than to billions. My niche is either what's already fallen out of fashion or what's floating in the air before anyone's put it in a column. I sit not at the tail of the trend and not on the bleeding edge — sweet spot. But if you think your niche is in the business idea itself, buckle up: the conglomerates either already ran the numbers and said no (99%) or left you the crumbs (1%). Niche only protects when you have something they can't buy — deep domain knowledge, community capture, local nuance. Everything else is 12-24 months before the giants show up.
Founder essay. On the word 'niche' — whether my essay is niche, what niche means in my world, why the conglomerates aren't already doing what you're doing, and when niche actually protects a business versus when it's just a pretty word for something that'll get eaten. Two scenarios, one honest test.
Friend, let's talk philosophy about the word 'niche.'
Can my essay even be called niche? I think yes — with an asterisk. Probably for people who've never heard of Odesa's particular flavor of humor or my not-always-witty jokes. Everything else is a normal human thought in unpolished delivery. Is that niche? Or is it just the voice of someone who doesn't want to write like a McKinsey Insights column?
I haven't decided yet. Let's figure it out together as we go.
Niche business is an interesting concept. In theory it could go big and suddenly you're a billionaire. In practice — more often, unfortunately — it goes red and takes you with it.
I'm one of those people who's always loved niche, well before the podcast-era boom of 'find your niche' hit in the early 2020s.
What does niche mean to me? Two categories:
Category 1 — what's fallen out of fashion. The trend everyone forgot about. Ceramics in 2015, before it came back as Pinterest aesthetic. Artisan bread in 2010, before every neighborhood had one. Therapy in 2005, when 'I'm going to see a therapist' still sounded like an admission of instability. The past that most people don't look back to — and you go there anyway, because something alive is still living there.
Category 2 — what's floating in the air, not yet named in the columns. The thing nobody's called a 'trend' yet, but you feel it in your gut coming. AI agents in 2020. Zero-alcohol cocktails in 2018. Remote-first companies in 2015. You're in it before the media starts writing 'Top 10 Trends of the Year.'
Both positions are comfortable. I'm not at the tail of the trend, where 90% of people are nervously chasing something already dead. And I'm not on the bleeding edge either, where the earliest 1% burn out because the market isn't ready. I like the sweet spot in the middle.
And that's when the chase for uniqueness kicks in. That thing nobody could ever replicate. Your snowflake business idea.
Here's where I warn you. If your uniqueness starts in the business idea itself (not the execution, not the brand, not the quality — but literally 'we do X and nobody else does X') — buckle up, this is about to get ugly.
Let's be real for a second. The conglomerates holding the global market right now — they would have found your niche and been printing money off it long before you moved a finger. They have data. Analytics. R&D departments. Harvard MBA interns writing weekly reports titled 'underserved market niches.' They see everything. They run the numbers on everything.
So if you 'found a niche nobody else sees' — there are only two possible scenarios.
Scenario A (99% probability): they saw the niche, ran the numbers, and decided it's not worth their movement. Maybe the TAM is too small. Maybe unit economics don't work in their cost structure. Maybe ROI doesn't clear their corporate hurdle rate. Maybe they tried it in 2011 and it flopped and it's on the internal DO-NOT-DO list ever since. Whatever the reason — if they passed, they had one. And often that reason is: the idea is worse than you think.
Scenario B (1% probability): they saw it and left it for the little guys, not wanting to hoover up every crumb. This happens — Amazon leaves Etsy alone because Amazon doesn't want to be in handmade. Google leaves DuckDuckGo alone because privacy-search is a narrow slice. But this is exception, not rule, and it holds only until the niche gets big enough to interest them.
Don't abandon niche. Just understand — it defends conditionally, not always. It only defends when you have something big companies cannot buy.
The honest test of your niche — three questions:
Question 1 — Can they buy it? If your defense is code that a team of 30 engineers could reproduce in six months, that's not defense. If your defense is a reputation you spent 10 years building with a specific community, that's defense. Not everything can be bought with money. What can't be bought is your moat.
Question 2 — Do they want to do it? Some things large corporations don't want to do for cultural reasons. Work 60-hour weeks in tiny teams. Answer customer emails personally. Attend local meetups in person. Run multilingual EN + ES + PT-BR support with the same care as English. Anything requiring 'normal people work abnormally long and abnormally close' — big companies refuse. That's your moat.
Question 3 — Is the niche small enough? Paradoxically, your defense is that the niche is too small to interest Amazon. If total addressable market is $10M/year, you can build a $2M ARR business and live very well. Amazon won't get off the couch for a $10M market. As soon as your niche grows to $500M, they will. And then you have 12-18 months to either sell or build a second moat.
So does my essay have a niche moat?
Run the three questions.
Can the big guys buy the voice? McKinsey Insights cannot publish 'buckle up, this is about to get ugly' — it doesn't clear their editorial review. Bain cannot publish 'I'm in therapy with my psychologist Viki' — because they have a compliance department. HBR cannot publish about the classmate who bullied you in first grade — not the format. So the first moat exists: the specific voice of a specific Ukrainian founder with a specific dialect can't be bought, only grown, and growing it doesn't pencil for a corporate content agency.
Do they want to write personal essays? No. Large corporations don't want first-person writing. It's a brand risk — what if the founder says something dumb tomorrow, the whole corporate image wobbles. They prefer anonymous 'Three Ways to Increase Conversion' posts. Second moat.
Is the niche small enough? 'Ukrainian and Russian-speaking small-business owners who read philosophical business essays' — that's a niche of, what, one million? Two? Ten? Not a billion. Amazon won't chase it. Bain won't hire a writer for it. But for one person, it's a huge audience. Third moat.
So the essay is defended. At least while I hold the voice, while I write honestly, while my niche hasn't grown to $500M where we all get eaten.
Same test — three questions — apply to your idea.
If all three answers are 'yes' — the niche defends you, you're positioned well, sleep easy. Work slowly, deeply, let it compound.
If one of the three is 'no' — start building the second moat right now. Don't wait for Amazon to look your way. By the time they do, you should already have a cushion.
If all three are 'no' — it's not a niche. It's you standing in line, convinced you're first, until the big guys show up and take the front spot.
And here's the finale. Niche is not defense. Niche is a starting position. Defense is what you build inside the niche: quality, community relationships, local depth, reputation you spent years growing, a voice that cannot be bought. Niche is the door. What you build behind that door — that's yours.
I've been living in this philosophy for years. It's been working so far. When it stops working, I'll write an essay about it. That one will be niche too.