An exit multiple is the valuation multiple at which a company is acquired or goes public, most commonly revenue, EBITDA, or ARR multiples. Common variants include revenue multiple, EBITDA multiple, ARR multiple for SaaS, or user-count multiple for consumer products. It is used to compare exits across deals, inform founder valuation expectations, and serve as a primary lens through which strategic acquirers and PE firms evaluate targets. It is the shortcut metric most M&A conversations actually run on, despite the existence of more sophisticated valuation methodologies.
The major multiples by business model: SaaS / subscription: typically valued on ARR multiple (annual recurring revenue), with public-market multiples rangi...
Option exercise is the action of paying the strike price to convert vested stock options into actual shares of common stock. It triggers tax consequences that vary by option type (ISOs create an AMT adjustment but no regular income; NSOs create ordinary income on the bargain element), requiring the holder to plan for both the cash outlay and the tax liabilities that arise. It is the moment options become stock, and the timing and structure of exercise significantly affect the holder's eventual after-tax outcome.
The exercise mechanics and methods:
Registration rights are contractual rights letting preferred stockholders require the company to register their shares with the SEC for public sale. They come in three flavors: demand registration (the holder forces a filing), piggyback registration (the holder rides on a company-initiated filing), and S-3 registration (short-form post-IPO filing), with the practical effect of letting investors actually sell shares in the public markets after an IPO. It is a critical structural right because preferred stockholders cannot freely sell their shares post-IPO without their shares being registered or qualifying for an exemption like Rule 144.
The three main types of registration rights:
A buyer persona is a research-based, semi-fictional profile of an individual decision-maker inside the target customer. It captures role, goals, pain points, success criteria, information sources, objections, and typical buying behavior, used to shape messaging, content, sales scripts, and product decisions around how that specific role actually buys. It differs from ICP in scope: ICP describes the account or household; the persona describes the human inside it.
A useful persona is built from real customer research, typically a combination of 10 to 20 structured interviews with closed-won customers (the actual buyers), churned customers (the buyers whose problem you didn't solve), and lost prospects (the buyers who chose a com...
Fund life is the total contractual duration of a VC fund, typically structured as 10 years with two 1-year extension options. The "10+2" structure is divided into an investment period (typically first 5 years when new investments are made) and a harvest/exit period (years 5-10 when portfolio matures and exits). Fund life is a critical structural constraint that affects everything from when GPs can make new investments to when LPs expect distributions to how aggressively portfolio companies must pursue exits. It shapes the temporal dimension of how funds operate.
The standard structure:
Years 1-5: Investment period:
Training data is the corpus of examples (text, images, code, audio, video) used to train AI models. The quality and scale of training data are two of the three key inputs (alongside model size and compute) that determine final model capability per the empirical scaling laws. High-quality training data is increasingly the constrained resource in AI development as compute scales faster than data quality. It's the input that becomes the output: what the model can do is bounded by what it learned from.
The components of modern AI training data:
Pre-training data (foundation model training):
Quick pointer: this entry focuses on the tax-advantaged characteristics of founder-issued common stock (QSBS, 83(b), capital-gains holding-period math). For the structural setup at formation (RSPA, vesting, repurchase rights, the share split), see [Founders Stock].
Founder shares are the formation-stage common stock whose tax-advantaged characteristics convert a tiny dollar investment into a potentially massive tax-advantaged outcome. Those characteristics are Qualified Small Business Stock (QSBS) eligibility under §1202 (up to $10M-$15M or 10x basis excluded from federal capital gains per founder, with post-OBBBA stock issued after July 4, 2025 getting the $15M cap, a $75M gross-assets ceiling, and tiered holding periods),...
A stock option is the contract granting the right to purchase common stock at a fixed strike price for a defined period. It is used as the primary equity-compensation mechanic at venture-backed startups, vesting over time (typically 4 years with a 1-year cliff) before becoming exercisable. It is the standard structure for employee equity in C-corp startups, distinct from outright stock grants because the employee must pay to convert the option into actual shares.
The mechanic of a stock option:
IP assignment is the contractual transfer of intellectual property ownership from individuals (founders, employees, contractors) to the company. It is typically executed through Proprietary Information and Inventions Agreements (PIIAs) signed by employees and Confidential Information and Invention Assignment Agreements (CIIAAs) signed by contractors, ensuring that all IP created during employment or in connection with services belongs to the company rather than the individual creator. IP assignment is foundational for company ownership of its product, technology, and competitive advantages, and missing IP assignments are one of the most-common cap-table cleanup issues discovered during diligence. It is one of the most-importan...
Pitch iteration is the systematic refinement of the pitch deck and narrative based on investor feedback patterns collected across meetings. It's used to sharpen the pitch toward what consistently resonates while addressing recurring concerns, with the discipline being to iterate based on patterns (3+ investors raising the same concern) rather than individual feedback, to avoid creating Frankenstein decks that try to address every investor's specific objections while losing coherence. It is the discipline that transforms okay pitches into great ones through structured refinement.
The iteration process:
Capture investor feedback systematically: