Start here
How do I know if my app idea is any good?
Updated 19 August 2026 · Thresholds read live from the verdict engine
Short answer
An idea is worth building when the pain is expensive, the people who have it are reachable at a price you can afford, and someone already spends money on a worse fix. You can score that in an evening. You cannot know the answer from the score, so the evening ends with a test rather than a decision.
Five questions that filter an idea in one evening
Score each from 0 to 3, honestly, writing one line of evidence beside every score. An idea under 10 out of 15 is usually a hobby with a business attached.
1. How expensive is the pain?
Count what the problem costs in money, hours or embarrassment per month. A freelancer losing four billable hours to invoicing is a 3. Mild boredom on a commute is a 0, whatever the market size says.
2. How often does it bite?
Daily beats weekly beats yearly, because frequency is what makes a person remember they have a problem while your ad is in front of them. Tax software gets one shot a year and has to pay for the whole year in that window.
3. Are they already paying for a worse fix?
Spreadsheets, a virtual assistant, three subscriptions taped together, an agency retainer. Existing spend is the strongest evidence you will find at this stage, and it also tells you your price ceiling.
4. Can you reach them for money you have?
Name the targeting. If the answer is a Meta interest, a subreddit with 200,000 members or a keyword with commercial intent, you are fine. If the answer is hospital procurement officers, your test is twenty conversations rather than a funnel.
5. What do you have that a stranger does not?
Ten years in the trade, an audience of 5,000 who trust you, a data set nobody else can assemble, distribution inside a company. Enthusiasm is not an advantage. Neither is being first, since being first is rarely true.
The signals that flatter you and mean nothing
- Big search volume. 40,000 searches a month for a phrase with no commercial intent is an audience for a blog. Look at what the top ads on that phrase are selling before you read the number as demand.
- A large market size number. The wellness market being worth billions tells you nothing about whether 500 people will pay you $40 a year. Market sizing is a slide for investors, not evidence for you.
- No competitors. The usual reason a category is empty is that the category does not pay. Check whether anyone tried and quietly stopped, which review dates and dead landing pages will tell you.
- Friends saying they would use it. They are answering a different question, which is whether they like you. Their answer stays the same when the idea is bad.
- Upvotes and waitlist signups from your own audience. Warm traffic converts several times better than cold traffic, and it runs out.
The hour of desk research that is worth doing
Four checks, none of which need a subscription, all of which change how you write the ad later.
- Meta Ad Library. Search two or three competitors and filter to active ads. Note the date each ad started running. An ad live for 90 days is paying for itself, which means the category can be bought profitably.
- One and two-star reviews of the incumbent. Sort by most recent, read fifty, and write down the complaint that repeats. That sentence is your headline and often your wedge.
- What the category charges. List five competitors and their plans. The spread tells you the shape of the market and stops you inventing a price nobody in the category has ever paid.
- Where the conversation happens. Find the subreddit, the Discord, the Facebook group. Read a month of posts. If nobody is complaining about your problem in the places these people gather, that is data.
Worked example
Fifteen minutes that changed the pitch
A founder planning a meal-planning app read 50 recent one-star reviews of the two biggest competitors. The repeated complaint was not the recipes or the interface, it was that the shopping list ignored what was already in the cupboard.
They rewrote the promise around that single job and tested it. The cupboard-first angle converted at more than twice the rate of their original general-purpose headline, on the same product idea and the same budget. The idea did not change. The sentence did.
A score ranks ideas. Only a test decides one.
The rubric is for choosing which of your five ideas to spend $300 on. It cannot tell you whether strangers will pay, because every input to it is your own judgement. Founders who skip the test at this point usually do so because the score was flattering.
The cheapest way to convert an opinion into evidence is to put the promise and the price in front of people who have never heard of you, and count who commits. Under 1% of cold visitors reaching for a card means the market has answered. At 2% and above you have permission to spend the next three months.
The numbers
Once you stop scoring and start testing, these are the bars, read from the verdict engine that applies them.
- Funnel starts before any verdict
- 30
- Below this, don't build
- 1.0%
- At or above this, build
- 2.0%
- Visitors before a segment counts
- 50
A score of 13 out of 15 and a conversion rate of 0.4% means the score was wrong. The order matters: the rubric picks what to test, the traffic decides.
How VerifyToLaunch does this
VerifyToLaunch runs an idea check as its first step, which grades the pitch through a paid-acquisition lens (how severe the pain reads, how reachable the audience is, how plausible the price looks) and suggests sharper angles. It is a language model reading your paragraph, so treat it the way you would treat a smart friend with no skin in the game. Useful for tightening the promise, worthless as evidence.
The part that counts as evidence is what happens next: a funnel carrying your price, cold traffic, and a verdict with a confidence interval attached. We hold the verdict rather than flattering you, which is the whole reason to use it.
Common questions
- Is there a score that means an idea is good?
- No score can do that, because every input is your own judgement of your own idea. A rubric is useful for ranking several ideas against each other so you know which one to spend the first $300 testing.
- How do I check demand without spending money?
- Read one and two-star reviews of the incumbents, check the Meta Ad Library for how long competitor ads have been running, and read a month of posts in the community where these people gather. That hour tells you whether money moves in the category and what language it moves for.
- My idea scores badly but I still want to build it. Should I?
- Sometimes, yes. Build it because you want it to exist, and be honest that it is a personal project rather than a business. The mistake is spending nine months on it while believing the market will show up.
- Does a big competitor mean my idea is bad?
- Usually the opposite. A funded competitor proves people pay for the category. What matters is whether you have a wedge they will not copy quickly and a channel you can afford.
Read next
- How do I validate an app idea?The whole method: pick a commitment that costs the visitor something, buy cold traffic, count who commits.
- My idea already exists. Does that kill it?Competition proves demand. Four checks on market health, and how to find a wedge.
- How do I find out if people will actually pay?Three ways to collect a paying signal before you build, and the CAC maths behind the rate.
- Should I build an MVP or validate first?What each path costs, what each measures, and when building really is the cheaper test.
Turn the score into a number
Describe the idea and get a funnel carrying your price, ad angles with creatives and a verdict held to the thresholds above. The judgement stops being yours and starts being the market's.
One-time licence, 30-day money-back guarantee