How I Would Pick a Niche Starting from Scratch in 2028 (My Honest Framework)

I've been asked this question probably fifty times in the last year: "If you had to start over completely, what niche would you pick?"

I never answer with a niche, because that's not the right answer. The niche I'd pick is specific to my background, my communities, my experience. The niche you should pick is specific to yours.

What I can give you is the framework I'd actually run if I had to start from zero today. Not a vague "follow your passion" exercise. The actual decision process.

What I'm Optimizing For

Before I explain the framework, here's what I'm trying to find:

1. A niche where I have enough insider knowledge to make a credible product
2. A niche with evidence that people are paying for solutions
3. A niche specific enough that my product feels made-for-them, not made-for-everyone
4. A niche with an accessible community I can reach without a big audience

Those four criteria eliminate most of the "obvious" options (personal finance, fitness, productivity) because they're crowded, require massive authority to compete, and have generic products already winning on every search term.

Step 1: Audit Your Communities (30 minutes)

I'd make a list of every community I've been part of in the last five years:

• Subreddits I read regularly
• Facebook groups I joined
• Discord servers I'm in
• Forums, Slack groups, industry associations
• Conferences I've attended
• Newsletters I'm subscribed to

Then I'd look for patterns. Where am I genuinely engaged — commenting, helping, asking questions? Where do I feel like an insider, not a lurker?

That's the community I know. And knowing a community is the foundation of making a product it actually wants.

For me, that list includes freelancer communities, solopreneur communities, online business building spaces, and a couple of niche hobby communities. Your list will be different.

Step 2: Find the Frustrations (1 hour)

For my top two or three communities, I'd spend an hour in each one just reading for frustration.

Not "what do people want" — wants are vague and aspirational. Frustrations are specific and actionable.

In freelancer communities, the frustrations I find most often:

• "I hate how long it takes to write client proposals"
• "I never know what to say when clients ask for a discount"
• "My invoicing system is a mess"
• "I have no idea how to handle this scope creep situation"

Each of those is a product waiting to be built. Proposal templates. Difficult conversation scripts. Scope management systems. Pricing frameworks with negotiation scripts.

I'd write down fifteen of these. Not to build all of them — to find the pattern of where the most acute, most repeated pain lives.

Step 3: Apply the "Would I Buy This?" Filter (20 minutes)

For each frustration on my list, I ask: if I had this problem, would I pay $29–$49 for a solution right now?

Not "someday maybe." Not "if it was really good." Right now, with my current problem.

This filter eliminates the aspirational products (things people want in theory but won't actually buy) and keeps the pain-relief products (things people buy because they have an active problem that's costing them time or money).

"A templates for writing client proposals faster" passes this test. Freelancers hate writing proposals. They write multiple per week. A 3-hour proposal that takes 30 minutes is immediately valuable.

"A guide to manifesting clients" does not pass this test. Vague, not urgent, no specific outcome.

Step 4: Validate That People Are Paying (30 minutes)

For the top 3–5 ideas that passed the filter, I'd check:

Etsy: Search for the product type. Do results exist? Do the top products have reviews? (100+ reviews = proven demand)

Gumroad: Same search. What's being sold and for how much?

Google: Type the product name into Google. Is there paid advertising? (Paid ads = someone's making money enough to pay for clicks)

Reddit: Search for the product category in relevant subreddits. Do people recommend products or tools for this problem? What are they recommending?

I'm not looking for zero competition. I'm looking for evidence that the category exists as a paid market. Some competition is healthy — it means the market is real.

I'm looking to avoid two things: a market where a few giant players already own every search term (too competitive), and a market with no paid products at all (may not be a paying market).

Step 5: Pick and Commit (5 minutes)

By this point, I should have one or two options that score well on all the criteria:

• Community I understand
• Real, specific frustration
• Would pay for a solution immediately
• Evidence others are already paying

I pick the one where I feel most confident I could make something genuinely valuable. Not the most money theoretically. The one where I know the audience's problem well enough to solve it well.

Then I commit. Not to a niche forever — I can always make a second product in a different niche. But I commit for long enough to actually build the product and find out if the market wants it.

The Niches I'd Consider in 2028 Starting Fresh

If I were running this framework right now, these are the categories I'd be looking hardest at:

ADHD productivity — high pain, underserved by generic tools, tight community, proven market (I wrote about this specifically)

Freelancer business systems — proposals, contracts, scope management, client communication scripts — perennial pain with clear willingness to pay

Solo parent work-from-home — time management, income diversification, business structures that work around unpredictable schedules

Creator-to-product transition — the journey from "I post content" to "I sell something" — massive audience of stuck creators who know they should be monetizing but don't know how

None of these are guaranteed wins. But all of them have the ingredients: specific audience, real pain, evidence of paying behavior.

Building Once You've Picked

Once you've got the niche, the build is usually faster than you expect. Most profitable digital products in the $29–$79 range are templates, systems, frameworks, or guides — not courses, not software, not complex deliverables.

A weekend of focused work can produce a first version. Then you sell, get feedback, and refine.

For the platform, I start everything on MadeThis. It handles storefront, checkout, and delivery in one place — no need to stitch together three tools to sell one product. See my MadeThis pricing breakdown if you want to understand the economics before you decide.

The niche isn't a decision you make once and live with forever. It's a hypothesis you test. Pick one, build one product, see what the market says.

That's the only real framework.