Build your target account list from one sentence

Your best customers have things in common that no filter captures. A database lets you ask for software companies in the UK with 50 to 200 people. It can’t find the companies that look like your three best customers, because that is a description, and the database wants filters.

With Otto, your team starts from the description. You write down the companies you want the way you would brief a colleague, and Otto works out what has to be true of each one.

What changes for your team

• Lists match the brief, not just the filters. A company only comes back when it fits every part of the description.
• Every company has a reason. Each one arrives with its size, location, fit score and one line on why it is on the list, so you can forward the list and anyone can see why each company is there.
• Your ideal customer profile is applied while it searches. Your size bands and disqualifiers count from the start, and companies you exclude never come back.

How Otto reads a brief

Here is one brief from the benchmark we publish:

“Find audiobook publishers investing in AI voice.”

A company only counts if both halves hold: it publishes audiobooks, and it has verifiable AI voice activity in the last 12 months.

Otto splits the sentence into the parts it can search on, like industry, region, size and keywords, and the parts it has to check company by company. Then it:

1. Searches the live web for companies that match.
2. Applies your ideal customer profile while it searches.
3. Reviews each candidate and drops the ones that don’t fit: the wrong size, the wrong stage, or a weak match to the brief.
4. Returns each company with its size, location, fit score and the reason it is on the list.

How it compares

Company-FindAll is a set of company briefs like the one above. Each provider gets the same brief, and a company passes only when it verifiably matches every constraint at the time of the run. Otto’s figures come from a run of 1,000 semantic searches, each result graded by a model against the pass criteria. The other providers’ figures are their published results.

At its High effort, Otto found more verified matches per brief than any published result. At its Low effort, for two cents a task, it passed 13.2, more than Perplexity Agent Pro’s best result at three times the price. The chart, the other benchmarks and the full method are on our benchmarks page.

After the list

Every company lands in a workbook. From there you can score it against your ideal customer profile, find the right people inside it with Find People, and put Signal Agents on it so you hear when one of them raises, hires or launches.

Try it

In Otto, go to Find Leads, then Companies, and describe the companies you want in a sentence. From Claude, ChatGPT or Gemini, ask Otto for “companies like” your best customer and say how many you want. It tells you what the search will cost before it runs.