An associate or principal will read 200+ teasers and CIMs in a year. Most of those deals are bad fits — wrong sector, wrong size, wrong dynamics. The discipline is not deciding which deal to do; it is deciding which deals to stop doing, fast.
Teaser → CIM
An anonymised teaser arrives first — 1–2 pages from the sell-side bank with sector, size, and broad financials. A buyer who wants more signs an NDA and receives the full Confidential Information Memorandum: 60–120 pages on the company, market, financials, and management. The CIM is a sell-side document. Read it as a marketing brochure, not a balanced presentation.
A disciplined screening framework
Five questions worth answering in the first read: (1) Is this in our strategy? Sector, size, geography, instrument. (2) What does the business actually do? Customers, competitors, unit economics. (3) Is the trajectory real? Growth quality, customer concentration, retention. (4) What's the entry price likely to be? Reverse-engineer from the CIM language and recent comp transactions. (5) What is the value-creation thesis? If you can't write it in 3 sentences after a 90-minute read, the deal probably isn't for you.
A useful internal habit: associates write a one-page screening memo on every CIM that gets past the teaser. The memo recommends pass / proceed / further diligence with the specific reasons. Memos are reviewed weekly. The discipline forces explicit reasoning where pattern-matching would otherwise rule.
The anchor problem
Sunk-cost reasoning is the most expensive bias in deal screening. A team that has spent two weeks on an opportunity has a strong gravitational pull to keep going, even when the early signal said no. The cleanest defence is a kill threshold: explicit criteria that trigger a deal-killer review, regardless of how much work has been done. Top firms enforce this culturally.
PE selectivity — saying no to 80%+ of deals
The typical PE or VC funnel looks roughly like this: 1,000 deals reviewed at the teaser or initial-contact stage; around 50 reach a first meeting; roughly 10 receive an indication of interest; around 3 proceed to a letter of intent; one closes. The numbers vary by firm size and strategy, but the shape is universal. The implication is that most deals die at the screening stage, not at the investment committee. IC time is expensive; the associate and principal layer exists to filter down to the 10 deals that deserve it.
The discipline that separates good screening from pattern-matching is writing explicit kill criteria before the meeting. A kill criterion is a specific, measurable threshold: TAM floor (e.g. $2B+ serviceable addressable market), gross margin floor (e.g. 60%+ for software), growth floor (e.g. 20%+ revenue CAGR for the last two years), and founder-pattern criteria (e.g. domain experience in the target vertical). If a deal fails any one of these, the pass is written before the management team walks in the room. This discipline protects the team from the charm and narrative that management teams are, by professional necessity, very good at projecting.
The 80%+ pass rate is a feature of a well-run PE or VC practice, not a failure. Portfolio construction requires it. A fund that does 15 deals over a 3-year deployment period needs to be confident that those 15 were the best risk-adjusted opportunities from the 1,000+ reviewed. The most common pass language is "interesting business, not for us right now" — which should mean something specific: outside strategy, outside size band, outside margin profile, or outside sector thesis. Writing that sentence clearly, with the actual reason, keeps the firm honest and leaves the door open for a future approach when the company has matured. Use the pitch deck extractor in this report's lab to structure the screening pass more systematically.