most business ideas die the same way.
someone has an idea. they get excited. they buy a domain. they start building. three months later they find out the market already has six versions of the same thing, all priced lower than what they planned to charge, and the people they thought would pay don't actually care about the problem they're solving.
i did this about 15 times before i figured out what was going wrong. the ideas weren't bad. the process was. i was building first and researching later. by the time i found out an idea didn't work, i'd already spent weeks or months on it.
now i spend about 1-2 hours doing structured research before i commit to anything. the whole point is to find out if an idea is worth pursuing before i write a single line of code, design a single page, or spend a single dollar.
i'm going to walk through the exact process. this is what i actually do, not a theoretical framework. you can run this today on whatever idea you're sitting on.
step 1: find out who already does this
open Claude or ChatGPT and paste this:
"i'm considering building [describe your idea in 2-3 sentences]. find 10 businesses that already do something similar. for each one, tell me: what they offer, what they charge, and what their customers complain about most in reviews on G2, Trustpilot, Reddit, or app stores."
then sit with what comes back.
the first time i ran this, i was looking at an idea that felt like a blue ocean. i couldn't find any competitors through normal googling. the prompt came back with 11 direct competitors i'd completely missed. some were on platforms i'd never checked. some were marketing themselves with different language for the same problem.
the market leader had 2.3 stars on G2. the reviews were brutal. and one complaint kept showing up across almost every competitor: onboarding takes too long and setup is confusing.
that single finding changed everything. instead of building another generic version of what already existed, the opportunity was obvious — make the thing that everyone else makes painful to use, and make it simple.
this is why you run the research before you build. you don't just find out whether competition exists. you find out where the competition is failing, and that failure becomes your angle.
if the prompt comes back with zero competitors, that's not necessarily good news. it might mean there's no demand. dig deeper — search Reddit, search Twitter, search niche forums for people describing the problem your idea solves. if nobody's talking about the problem, the market might not exist.
step 2: check if the maths can work
this is where most people lie to themselves. they assume they'll get customers, assume they'll charge a certain price, and assume the margins will work out.
don't assume. calculate.
answer these questions honestly:
— what would you charge? look at what competitors charge and price in the same range, or justify why you'd price differently.
— how many customers or sales do you need to hit $5,000/month? if the answer is 500 customers at $10 each and you have no audience, that's a problem. if it's 10 clients at $500 each and you can reach them through LinkedIn or cold outreach, that's realistic.
— what does it cost you to deliver? your time counts. if each client takes 10 hours of your time and you're charging $500, you're making $50/hour. is that sustainable? does it scale?
— how long until first revenue? if the answer is "6+ months," that's a different bet than "2 weeks." know which one you're signing up for.
the point of this step is to find out whether the idea can reach your financial target at a scale you can actually achieve with what you have right now. not theoretically. not with a big audience you don't have yet. with your current situation.
most of my 15 failed ideas died at this step when i finally started running the numbers honestly. they needed things i didn't have — a large audience, significant capital, or skills i'd need months to develop.
step 3: find the gap, not just the market
knowing competitors exist is not enough. you need to know where they're weak.
go back to Claude and paste this:
"based on the competitors you found, what is the most common complaint or unmet need across their customer reviews? what specific problem are customers consistently saying is not being solved well? if i were entering this market, what angle would give me the strongest differentiation?"
the answer to this question is your product angle. not "i'll build a better version." specifically: "i'll build the version that solves the thing everyone else gets wrong."
when i did this for my own products, the gap wasn't some clever innovation. it was usually something embarrassingly simple that the existing players had overlooked because they were focused on adding features instead of fixing the basics.
your competitors' weakest point is your strongest entry point.
step 4: argue against yourself
this is the step that hurts and the one that saves the most money.
paste this into Claude:
"i'm about to commit to building [your idea]. my budget is [amount]. my timeline is [timeframe]. argue against this decision. what are the three most likely ways this fails? what hidden costs am i not seeing? what would have to change in the market to make this irrelevant?"
read what comes back without getting defensive. the whole point is to stress-test your plan from the perspective of someone who wants you to fail. if the arguments against your idea are stronger than the arguments for it, that's worth knowing now rather than after you've spent three months building.
not every red flag is a dealbreaker. but if the devil's advocate prompt raises problems you can't answer, you need to address them before you start building.
what to do with the results
after running these four steps, you'll be in one of three positions:
the idea has clear demand, a real gap in the market, workable economics, and no dealbreaker risks. build it.
the idea has potential but the numbers don't work at your current scale, or the competition is too strong in its current form. modify the angle or park it for later.
the idea looked good on the surface but the research shows it's a dead end. kill it. that 2 hours just saved you months.
all three outcomes are valuable. even killing an idea is a win, because now you can redirect that time toward something with better odds.
i've now used this process for 4 products i shipped in the last 120 days. each one started here — not with a hunch, not with excitement, but with 2 hours of structured research that told me whether the idea was worth my time.
if you want the full process
this newsletter covered the core of the research phase. but the complete framework has 5 phases — from initial idea generation through to getting your first paying customer. i wrote the whole thing into a 16-page guide with 7 copy-paste prompts, three kill filters, and a 30-day launch plan.
it's free - and you can get it HERE
— Marcus