Counter Evidence: The Step Most Founders Skip

Once you like an idea, the search box starts working against you. Every query you type is shaped by the answer you want, and every thread you find confirms it. An afternoon of research can feel productive while adding no information at all.

Counter evidence is the correction. It is the deliberate search for the parts of the world that disagree with your idea, collected before you have anything invested in it.

What Counter Evidence Looks Like

Counter evidence is not a vague worry. It is a specific, checkable thing that would change your plan if it turned out to be true. It usually takes one of these shapes:

• People describing the same problem as solved and not worth fixing.
• A free tool or built-in feature that already covers the workflow well enough.
• Workarounds that cost so little that nobody would switch.
• A buyer who cannot spend money on the problem, such as someone in a role with no budget.
• A reason the problem is hard to solve, stated by people who tried.

Notice that none of these are insults to your idea. They are facts about the market that you would rather learn now.

Where to Look for It

Counter evidence hides in the same places as regular evidence, in different threads. You just have to search differently.

• Search for the problem plus “solved” or “works fine”, for example “reconciling accounts works fine”.
• Search for the current tool by name and read the users who are happy with it. “Switched to” plus the tool name is a good pattern.
• Read the one-star reviews of any existing product in the space. They tell you what people tolerate and what makes them leave.
• Look inside the threads where someone asks for a tool and the answer is “just use a spreadsheet”.
• Search for the problem plus “not worth it” or “we live with it” and read why.

If you find nothing at all, be suspicious. A problem with no counter evidence usually means you have not looked for it yet.

A Counter Evidence Pass in Four Steps

Run this after you have a favorite idea, not before. Twenty minutes is enough to start.

1. Write down your claim in one sentence, then write the opposite. If your claim is “bookkeepers lose hours every month on reconciliation”, the opposite is “reconciliation is fast enough that nobody would pay to fix it”.
2. Turn the opposite into three searches in the language a satisfied person would use. Search for satisfaction, not for complaints.
3. Read the results as if you were the person arguing against your own idea. Collect the strongest lines word for word, with links.
4. Sort what you found into three piles: fatal, fixable, irrelevant. Fatal means the problem does not exist or nobody would move. Fixable means the objection points at a design choice. Irrelevant means it is about a different audience.

The third pile is where people cheat. An objection from a different audience is genuinely irrelevant, but only if you can name why that audience is different.

How Much Counter Evidence Should Sink an Idea

There is no formula, so use the shape of the finding instead of counting lines.

• Many people happy with the current option, and they are your audience. This is the strongest stop sign you will see.
• Many people happy, but a different audience from yours. Note it and continue.
• The complaint exists, but the workaround costs five minutes a month. Real pain, no market. Narrow the audience until the cost grows.
• People tried to build a solution and stopped. Read why. Sometimes it is a hard technical problem, sometimes it is a sign that nobody paid.
• The counter evidence all comes from one thread by one person. Weak. Look for it elsewhere before you act.

The decision is not “ignore the objections” or “abandon the idea”. It is whether you can name the counter evidence and still explain why your version of the solution is worth a stranger’s money.

Recording It So You Do Not Forget

Keep the counter evidence in the same place as your positive evidence, with the same level of care. That means a file per idea, and entries in four parts.

• The objection, quoted exactly, with a link.
• Which group of people it came from.
• Your answer, if you have one, and the date you wrote it.
• What the objection would have to be wrong about for your answer to hold.

The last field is uncomfortable to write and worth every minute. It tells future you exactly what to check first when the project stalls.

FAQ

Is counter evidence the same as a competitor analysis?

No. Competitor analysis is about who else sells. Counter evidence is about whether anyone needs to buy at all, including from an existing free option or a manual habit.

What if I find counter evidence after I started building?

Collect it anyway. Finding it late is worse than finding it early, but it is still much better than never finding it. Let it narrow the audience rather than cancel the project by default.

Does the tool include counter evidence?

Yes. A report has a counter evidence section alongside the problem signals, so the positives and the objections arrive together instead of one at a time.

How often should I repeat this pass?

Whenever you change the audience or the claim, and once more before you commit real money. A new audience brings new objections.

If you want the arguments against your idea collected alongside the support for it, the research tool builds the report that way: it gathers discussions from Reddit’s public archive, Hacker News, Stack Overflow, and GitHub, groups duplicates, keeps only quotes it can verify against the source text, and includes a counter evidence section. A report takes two to four minutes and costs 2.99 EUR once. Two related posts: Spotting Recurring Complaints in a Sea of Posts and Evidence-Based Idea Validation: What Counts as Proof.