EquitleEducationTry for free

Lesson 4 of 6 · Start here

How Hawk learns your judgment

6 min read · Updated Nov 16, 2026

The short answer

Hawk learns from two things: the description you write of a great fit, and every company you keep or cut afterward. Each decision sharpens its picture of what you want, so the next list is closer to the one you would have built yourself.

By the end of this lesson you will be able to:

  • Explain what the fit score measures
  • Tell the search from the description, and say what each one does
  • Use keeps and cuts to improve the next list

How Hawk learns

  1. 1
    You describeWhat a great fit looks like
  2. 2
    Hawk scoresEvery company, with reasons
  3. 3
    You decideKeep the fits, cut the rest
  4. 4
    Hawk learnsIts picture of you sharpens
  5. 5
    Next listCloser to your own picks

Each round starts from what the last one taught

One round of the loop. Each list starts from what the last one taught.

Key takeaways

  • The search decides which companies are in scope. Your description decides how each one is scored.
  • A company can match the search perfectly and still score low if it misses your description.
  • Every keep and every cut is a lesson. The more decisions you make, the more precise Hawk becomes.
  • A company you have already judged is never counted twice.

Why this matters

Two investors can look at the same list of companies and pick different ones. One wants recurring service revenue. Another wants a founder near retirement. A third rules out anything with one dominant customer. That judgment is the valuable part of sourcing, and it is the part most software ignores.

Hawk is built to learn it. That is the main difference between Equitle and a company database, and it is why a later list is better than the first.

What Hawk learns from

  • Your description. The sentences you write about what makes a company a great fit, and what to avoid.
  • Your decisions. Every company you keep, and every one you cut.
  • Your corrections. When Hawk shortlists a company you would not have picked, cutting it is the most useful thing you can do.

It does not matter where you make a decision. Keeping or cutting a company in the app, by text, by email or from an AI assistant all count the same.

What the fit score means

Every company on a list gets a fit score. The score answers one question: how well does this company match what you described as a great fit?

That is different from asking whether the company matches your search. The search, and the thesis behind it, only decide which companies are considered. The description decides the score.

Matches the search?Matches your description?Result
YesYesHigh score. This is what you asked for.
YesNoLow score. The right kind of business, but not a fit for you.
Close, and you chose adjacentYesOn the list, scored lower as a related business.
NoEitherNot on the list.

Say you search for commercial HVAC services and describe a great fit as a company that earns most of its revenue from maintenance contracts. An installer that only works on new buildings is squarely a commercial HVAC company, and it will still score low, because it is not what you said you wanted.

How to teach Hawk well

  1. Write the description you would give a new analyst. Name the things you check, not general qualities.
  2. Decide on plenty of companies, early. The first list is where Hawk has the most to learn.
  3. Cut as readily as you keep. A cut tells Hawk as much as a keep does.
  4. Be consistent. If you keep a company for a reason, keep the others that share it.
  5. Update the description when your thesis changes. Do not rely on decisions alone to carry a change of direction.

What you end up with

Over time Hawk holds a working picture of your judgment: the kinds of company you pursue, the ones you pass on, and the reasons. It is particular to you, which is why two firms can run the same search and end up with different shortlists. The practical effect is less time reading lists and more of it on companies worth a call.

Try it in Equitle

Nothing has to be set up for Hawk to learn. Keep the companies you like and cut the ones you do not, and each decision is counted once and carried into the next list Hawk builds for you.

Key terms

Sources

  1. Equitle (2026). Product overview: what Equitle does, who it is for and how a free trial starts. View source
  2. Equitle (2026). Start a hunt: the questions Equitle asks before Hawk begins. View source

Common questions

What is the difference between the search and the fit description?

The search sets which companies are considered, for example commercial HVAC services in Texas. The fit description sets how each one is scored. A company has to pass the first to be on the list, and match the second to score well.

Do decisions I make by text or from an AI assistant count?

Yes. A keep or a cut counts the same wherever you make it: in the app, by text, by email or from an assistant such as Claude or ChatGPT.

What if my thesis changes?

Update your description of a great fit, or start a new hunt for the new thesis. Hawk scores against the description it is given, so a change there shows up in the next list.