Writing a hypothesis when you know almost nothing

Some accounts are easy. A public filing, a new executive, three job posts and a press release. The hypothesis nearly writes itself.

Most accounts aren't like that. A private company with a thin website, a buyer who never posts and no news in a year. You still have to write something worth answering.

You can. You just have to be honest about how little you know and make the guess work harder.

Thin facts are still facts

Start by listing what you do know, no matter how small. The industry. Rough headcount. Where they're located. How long the buyer has been in the seat. What the company sells and to whom.

That's more than it looks like. A 400-person distributor in the Midwest that's been around 40 years tells you a lot. Likely an older ERP. Likely a small IT team. Likely a founder or second-generation owner who still signs the big checks.

None of that is proof. All of it narrows the field of problems they probably have.

Borrow from the pattern, not the company

When the account is quiet, the hypothesis comes from what you've seen at companies like it. That's where your firm's experience earns its keep.

If you've worked with ten mid-size distributors, you know the three problems that come up again and again. Pick the one most likely to fit this company's size and age. Your guess is now grounded in real work, though not in this company's news.

Be careful here. "Companies like yours" is the phrase that kills cold email. Don't say it. Use the pattern to choose the guess, then write the guess about them.

Use the one fresh thing you can find

Almost every account has one recent fact if you look. A new hire on LinkedIn. A job post. A trade show booth. A new location on the website. A line in a local business paper.

That fact becomes your hook. It doesn't have to be big. It has to be real and recent, so the buyer knows you looked at them and not a list.

Say Corlan Supply, a regional industrial distributor, just posted a job for its first data analyst. That's thin. But it's something.

Noticed Corlan is hiring its first data analyst. When a distributor your size adds that role, my guess is the reporting still lives in spreadsheets pulled from an ERP nobody wants to touch. Is that accurate, or is the bigger issue something else?

One fact, one pattern, one guess, one exit.

Say less and guess smaller

When you know a lot, you can make a sharp, specific claim. When you know little, a sharp claim reads as made up.

So shrink the guess. Name one problem, not three. Name the symptom the buyer would feel, not the root cause you'd fix. "Reporting lives in spreadsheets" is easy to confirm or correct. "Your data architecture can't scale" sounds like a pitch.

Small guesses are easier to be usefully wrong about. The buyer can correct a small guess in one line.

With thin research, you're more likely to miss. That's fine, as long as you've made missing cheap for the buyer.

The exit question does that job. "Is that accurate, or is the bigger issue something else?" tells the buyer you know you might be off. It invites the correction instead of daring them to argue.

A quiet account that replies with "not really, our problem is inventory accuracy" has given you more than any database could. That reply is the research you couldn't find.

Know when to wait

Sometimes there's truly nothing. No fresh fact, no clear pattern, no reason this week over any other week. In that case, don't write the note yet.

Put the account on watch and wait for something to change. A new hire, a new site, a new leader. A hypothesis built on a real trigger next month beats one built on air today.

The point

Thin research isn't a reason to write a vague note. It's a reason to write a smaller, more honest guess, built on one real fact and the pattern you've seen at similar companies. For more on keeping the research fast, see research that takes ten minutes, not two hours and the 3x3 for a private company.