The Deal Screening Memo Is the Cheapest Document Your Firm Does Not Write

Firms write investment committee memos for the deals they pursue and write nothing at all for the ninety percent they kill. So the pursue decisions get argued, documented, and reviewed at exit, and the kill decisions vanish the moment the analyst closes the file. That is backwards. The kill decisions are the larger sample, they are made faster and with less information, and they are the only ones nobody grades. A deal screening memo is one page, takes twelve minutes, and is the only record a firm will ever have of the deals it walked away from and why.

Key Takeaways

• Ninety percent of deal decisions are kill decisions and almost none of them are written down, so almost none of them are reviewed.

• A screening memo needs four sections: the facts, the buy box fit, the reason for the kill, and the condition under which the answer changes.

• Without a written record, a firm cannot distinguish a disciplined pass rate from an accidental one.

• Fischhoff's 1975 work on hindsight bias established that once an outcome is known, people misremember what they believed beforehand. A contemporaneous note is the only defense.

• The most valuable field is the last one: what would have to be true for this to be a yes. It converts a dead file into a standing order.

Why does a firm write memos only for the deals it pursues?

Because the memo is treated as an approval instrument rather than a record. Committees require documentation to release capital, so documentation appears wherever capital is requested. No capital moves on a kill, so no document is required, and the decision leaves no trace beyond an unread email thread.

The consequence is a sampling problem. A firm that underwrites 40 deals a year and closes 4 has written 4 memos and made 36 undocumented decisions. Its written record of judgment comes entirely from the 10 percent where it said yes, and every process review runs on that biased tenth.

The screening decisions are also the ones made under the worst conditions. A committee memo follows weeks of diligence and a full model. A screening decision is made in twenty minutes from an offering memorandum written to sell, which is precisely why the offering memorandum should be read as the marketing document it is rather than as a data source. Fast decisions from thin information are exactly the kind that need a written record.

What belongs in a one-page deal screening memo?

Four sections and nothing else: the facts as extracted, the buy box fit line by line, the reason for the kill stated as a single binding constraint, and the condition under which the answer would change. Any longer and it will not get written. Any shorter and it is not a record.

The binding constraint field is where most drafts go wrong. A memo that lists six concerns has not made a decision, it has produced an atmosphere. If a deal fails on price, say price, and name the price that would clear. If it fails on the submarket, say the submarket, and say what evidence would change that view. One reason is falsifiable. Six reasons are a mood.

The buy box column only works if the criteria are written down as tests rather than as preferences, which is the argument for scoring a deal against the buy box so that fit is defensible.

What does the memo cost and what does it return?

It costs twelve minutes per killed deal. The return is a dataset of your own decisions, which is the only dataset that describes how your firm behaves rather than how the market behaves.

Every figure below derives from the stated inputs. A team screens 400 deals a year and kills 360. At twelve minutes each, the annual cost is 72 hours, or 1.8 weeks of one analyst.

Now the return side. After one year the firm can answer questions it currently cannot: what share of kills were priced, what share were submarket, what share were structure. Suppose the 360 memos sort as 189 priced too high, 76 outside the buy box on type or size, 58 sponsor or structure, and 37 unknown because the package was incomplete.

The last row is the one that pays for the whole exercise. Thirty-seven deals a year were declined because the package did not contain enough to evaluate them. That is not a decision, it is an omission, and no firm knows it is happening until the reasons are written down and counted. At a 10 percent close rate on properly screened deals, those 37 files represent between three and four transactions a year that were never considered.

The price row carries a second reading. If half of all kills are priced, the 189 files are worth tracking to their eventual sale price.

How does a screening memo protect against hindsight?

It preserves what you believed before the outcome was known. Baruch Fischhoff's 1975 research on hindsight bias demonstrated that once people learn how something turned out, they overestimate how predictable it was and misremember their own earlier judgments as closer to the result. A contemporaneous note is the only reliable defense.

The failure mode in acquisitions is specific and familiar. A deal the firm passed on trades eighteen months later at a substantial gain. In the retelling, everyone remembers being closer to a yes than they were, or remembers the concern that turned out to matter and forgets the four that did not. Neither version supports learning, because neither is what anyone thought at the time.

With a memo, the exercise becomes arithmetic. Pull the file. The binding constraint was price at $14,200,000 against an $18,500,000 ask. It traded at $17,900,000. The question is now precise: was the pass correct on the information available, or was the basis assumption wrong? Those have different remedies. The first requires nothing. The second requires repricing the buy box. Without the memo, the conversation produces a feeling instead of a finding.

How does the screening memo relate to the IC memo?

The screening memo is the same instrument at a tenth of the cost, applied at the stage where the decision volume is. Both make a claim that reality can contradict. The IC memo carries a full thesis, load-bearing assumptions, and kill criteria. The screening memo carries one constraint and one reopen condition.

Firms that write both get a continuous record from first look to exit. Firms that write only the second have a detailed account of four decisions and no account of three hundred and sixty. The structural logic of the longer document is covered in why an investment committee memo is a falsifiable argument, and the screening memo is that argument compressed to the smallest form that still commits the author to something.

Frequently Asked Questions

How long should a deal screening memo be?

One page, and closer to half. The constraint is not thoroughness, it is whether the document gets written at all on the four hundredth deal of the year. A template with four fixed fields that takes twelve minutes will survive. A two-page narrative will not.

Who should write the screening memo?

Whoever made the kill decision, at the moment they made it. Delegating it to a later summarizer defeats the purpose, because the value is in the contemporaneous record of the reasoning, not in a tidy account written afterward.

What do you do with screening memos once you have them?

Read them in aggregate once a quarter. Sort kills by binding constraint, count the categories, and check the reopen conditions against current market data. The individual memo is a receipt. The quarterly stack is the diagnostic.

Conclusion

A firm's judgment lives in its rejections. That is where the volume is, where the information is thinnest, and where nobody is watching. Writing one page at the moment of a kill costs under 4 percent of an analyst year and produces the only honest record of how a firm decides. It makes the pass rate auditable, turns dead files into standing orders with triggers, and surfaces the deals killed for reasons that have nothing to do with the investment. The document is cheap because it is short. It is valuable because it is the only one nobody else is writing.

Related

• The OM inbox problem: why most inbound deal flow never gets screened

• Scoring an OM against the buy box: making fit defensible

• An investment committee memo is a falsifiable argument, not a summary