The Startup Advisory Board Agreement: A Founder's

Why the advisory board agreement is the most quietly consequential document in an early-stage company

An advisor is not an employee, not a director, not an investor. That in-between status is exactly what makes the advisory board agreement more important than founders realize. Get it right and a well-chosen advisor unlocks a customer, a technical hire, or a lead investor at exactly the moment you need it. Get it wrong and you have granted 0.5% of your company to someone who will not return your calls, has no obligation to keep your roadmap confidential, and holds a fully vested equity stake for the next seven years while contributing nothing.

This is a walkthrough of a standard early-stage advisory board agreement one section at a time, with the specific edits that separate an agreement that protects the company from one that lets an advisor coast. Nothing here is legal advice — always have final versions reviewed by counsel — but every founder should understand what each clause is actually doing before they sign it or send it to a prospective advisor.

Section 1 — Engagement of Services: define exactly what the advisor owes you

The template opens by naming both parties and reciting the reason the company wants the advisor. Then it gets to the operative language: the advisor agrees to attend at least six meetings per fiscal year, spend at least ten hours per quarter communicating with management by phone or email, and render advice on the issues discussed. It is short, it is specific, and it is the single most important paragraph in the entire agreement.

Founders skip past this section because it feels like formality. It is not. The specific numeric commitment — six meetings, ten hours per quarter — is the only defensible standard you have when an advisor stops showing up. Without it, "the advisor did not perform" is a subjective claim the advisor will contest. With it, you can point to a clause in the signed agreement and say the advisor is in breach.

Match the commitment to the equity. Six meetings per year plus ten hours per quarter is roughly 40–50 hours annually. That is the right range for a 0.1%–0.5% grant on a two-year vest. If you are granting 1% or more, push the commitment higher: monthly meetings, two intro calls per quarter, or a specific deliverable such as a named-account introduction goal. · Add a named-account or introduction expectation when relevant. If the reason you want this advisor is their network in a specific customer segment, write it in: "Advisor will make at least three warm introductions per quarter to prospects mutually agreed by the parties." Advisors who cannot commit to that specificity are the ones who will vest quietly on strategic advice they never end up delivering. · Require attendance in a specific format. Reserve the right to designate whether meetings are in-person, video, or asynchronous, and add that no-shows without at least 48 hours notice count as a missed meeting for termination purposes.

The template also gives the company the right to designate meeting time and place, subject to the advisor''s availability. That last phrase — "subject to the availability of Advisor" — is where the tension lives. An advisor with a demanding day job will invoke availability constantly. If your candidate is a public company executive or a partner at a large VC fund, expect meeting scheduling to be the friction point, and account for it either by dropping the meeting count or by shifting the commitment toward asynchronous work.

Section 2 — Compensation: the option grant is the entire economic deal

The template compensates the advisor entirely in stock options rather than cash. This is the standard early-stage structure and the right one — cash is scarce, and an advisor motivated by long-term equity upside will act more like a partner than a vendor. The specific numbers you fill into the blank are governed by market benchmarks that have held remarkably stable for a decade.

The FAST framework, adapted to reality

The Founder/Advisor Standard Template (FAST) from the Founder Institute is the reference most founders and lawyers use to size the grant. In practice it works out to roughly:

Idea stage: 0.20%–0.40% for standard engagement; up to 1.0% for a heavy-lift strategic advisor. · Startup stage (some traction, pre-seed to seed): 0.15%–0.30% standard; 0.50% strategic. · Growth stage (post product-market fit, Series A+): 0.10%–0.20% standard; 0.30%–0.50% strategic.

Founders routinely over-grant here. Two anchors keep you honest: (1) the advisor pool is finite — most companies reserve 1%–2% of the fully diluted cap for the entire advisor bench, and (2) each advisor is one of five to eight you will eventually recruit. If your first advisor takes 1%, you have almost nothing left for the next hire who might be genuinely transformative.

Options, not restricted stock

The template grants an option to purchase common stock at a strike price set by the board (in practice, the current 409A fair market value). This is correct. Never grant restricted stock to an advisor — the advisor gets taxed on the value of the grant at grant time whether or not they ever sell the shares, which creates immediate friction and often kills the deal. Options let the advisor decide when to exercise, and if the company fails they simply let the option expire with no tax consequence.

The strike price question

Grants are made at the current fair market value determined by a 409A valuation, which is refreshed at least annually and after every priced round. If you are between valuations, the board can set the strike using its most recent 409A. Do not use a stale valuation from twelve months ago just because it is lower — the IRS and future auditors will flag it. Refresh the 409A before granting to any advisor whose grant is material to their decision to join.

Section 3 — Option Vesting: 24 months monthly, no cliff, tied to service

The template vests the option in 24 equal monthly installments for as long as the advisor continues to serve. This is the market standard and it does two important things well: 24 months is short enough that advisors will actually reach full vesting during a period when they are useful, and monthly-with-no-cliff prevents the "advisor bails at month 11" problem that plagues one-year-cliff structures.

The critical modification most standard templates lack: add a service-based vesting condition that vesting stops if the advisor stops materially performing the Services defined in Section 1. Without this language, an advisor who stops showing up at month 6 continues to vest monthly for the next 18 months as long as they have not formally resigned. With it, the company has the right to stop vesting when performance stops.

A cleaner version of Section 3 for a founder-protective agreement:

"Beginning on the Effective Date, the Common Stock subject to the Option will vest in twenty-four (24) equal monthly installments, provided that the Advisor continues to materially perform the Services in each vesting month as reasonably determined by the Company''s CEO. The Company may suspend or terminate further vesting upon thirty (30) days written notice for material non-performance, subject to a good-faith opportunity to cure." Two related mechanics to understand: (1) the option itself will have a post-termination exercise window (typically 90 days, sometimes extended to 12 months for advisors) set by the option plan, and (2) unvested options simply lapse — they do not need to be repurchased. This is why options are so much cleaner than restricted stock for advisor grants.

Section 4 — Confidentiality and IP Assignment: the section that actually protects the company

Section 4 does the real work. It has four subsections, and every one of them matters.

4.1 — Confidential Information, defined broadly, protected for two years post-termination

The definition of Confidential Information is deliberately broad — anything designated as confidential or that should reasonably be understood as confidential given the circumstances. Specific categories are listed: technology, inventions, product ideas, customers, employee compensation, business plans, strategy, marketing, forecasts, pricing, finances, and third-party information the company is obligated to keep confidential. The obligation runs for the duration of the relationship and for two years after termination.

Two edits worth making: extend the tail period to five years if your business relies on trade secrets that take longer to become obsolete (proprietary algorithms, hardware designs, biotech IP), and add "customer lists, revenue by account, cap table, and investor communications" as explicit categories — the specificity strengthens enforceability if you ever need to sue.

4.2 — License and Assignment: any invention the advisor makes as part of the engagement belongs to the company

The template treats inventions, works of authorship, developments, know-how, improvements, and trade secrets made by the advisor within the scope of the relationship as "works made for hire" that are compensated by the equity grant. If the advisor incorporates pre-existing or independently developed work into a company invention, the company gets a royalty-free, irrevocable, worldwide, perpetual, transferable, non-exclusive license to use that work.

This is the clause that stops an advisor from later claiming they own the framework or the algorithm they helped develop. Two things to understand about it: the "works made for hire" language is legally strongest when the work is done specifically for the company at the company''s direction, and the license grant on pre-existing work is what protects you when an advisor brings a proprietary tool or framework into the collaboration and then leaves.

Founder edit: add an assignment fallback — "and to the extent any Inventions are not properly characterized as works made for hire, Advisor hereby assigns to the Company all right, title, and interest in and to such Inventions." Belt and suspenders, but this is the single sentence that has decided a dozen IP lawsuits.

4.3 — Exclusions: the four categories courts require

Confidential Information excludes anything that (i) becomes publicly available without breach, (ii) was demonstrably known to the advisor before receiving it from the company, (iii) is received from a third party who did not disclose it wrongfully, or (iv) was independently developed by the advisor without reference to the confidential information. These four exclusions are non-negotiable — a confidentiality clause without them looks overreaching and courts will refuse to enforce it as written.

4.4 — Use of Confidential Information: only in pursuance of the relationship

The advisor may use confidential information only in service of the relationship, must not disclose it without written consent, and must take reasonable measures to prevent unauthorized use. The "commingling" language matters more than it seems — advisors who work with multiple startups in adjacent spaces are the biggest confidentiality risk, and the segregation obligation is what gives you a claim if your data ends up informing their advice to a competitor.

Section 5 — Independent Contractor: the advisor is not an employee, and neither party can pretend otherwise

This section is short and dry, and it is doing critical work. The advisor is an independent contractor, not an employee, not a partner, not a joint venturer. The advisor cannot bind the company to contracts. The company does not withhold income tax, Social Security, or benefits.

Why this matters: without an explicit independent-contractor clause, the IRS and state employment agencies can reclassify the relationship as employment, exposing the company to back payroll taxes, penalties, and in some states triple damages under wage-and-hour laws. The three factors that most often trigger reclassification are (1) the company controlling how and when the advisor works (not just the outcome), (2) the advisor working primarily or exclusively for the company, and (3) the relationship lasting years without formal renegotiation.

Founder rule: if your advisor is doing anything that looks like a full-time job — sitting in your office five days a week, running a functional area with direct reports, writing production code — they are not an advisor. Either hire them as an employee or move the relationship to a consulting agreement with a scoped statement of work. Trying to structure a de facto employee as an advisor is one of the fastest ways to get sued.

Section 6 — General Provisions: term, governing law, and the boilerplate that decides who wins in court

6.1 Term: 24 months, terminable on 30 days notice, confidentiality survives

The agreement runs 24 months, but either party can terminate on 30 days notice. This is important — an advisor who is not delivering can be removed without cause, and the founder who wants out is not locked in either. The confidentiality obligations survive termination for the two-year tail defined in Section 4.

Match the term to the vesting schedule. The template''s 24-month term aligns with 24-month vesting, which means at term end the advisor is fully vested and the natural checkpoint arrives for either renewal (with a new grant) or a clean exit. Extending the term without extending the equity is how founders end up with permanent advisors who long ago stopped adding value.

6.2 Governing Law and Venue: fill it in, do not accept the advisor''s state

The template picks New York; substitute the state where your company is incorporated or headquartered. Delaware, California, and New York are the three defensible choices for most US startups. Never leave the governing-law state blank, and never accept the advisor''s home state — that hands enforcement to whichever jurisdiction the advisor most controls.

6.3 Specific Performance: the injunction clause

The template acknowledges that a breach of the confidentiality and IP obligations would cause immediate irreparable harm that monetary damages cannot cure, and gives the company the right to seek an injunction restraining the advisor from further breach — without the company having to post a bond. This is important because the standard for getting a preliminary injunction requires the plaintiff to demonstrate irreparable harm, and having the advisor pre-acknowledge that harm in writing dramatically shortens the evidentiary fight.

6.4 Entire Agreement: the merger clause

The agreement supersedes all prior conversations, term sheets, and side letters. This means that verbal promises made during recruiting — "we will grant you the same package as our lead advisor," "the equity will accelerate on change of control" — are not enforceable unless they appear in this document. Read the agreement against every promise you made in the recruiting conversation and add anything material as an amendment or a separate side letter, signed and dated.

6.5 No Conflict: the advisor represents they are not double-committed

The advisor represents that entering into this agreement does not breach any obligation they have to another party — no NDA violation, no non-compete conflict, no fiduciary duty to a prior board they still sit on. This clause is what gives you a claim if the advisor was quietly under a non-compete with a competitor when they signed with you and the competitor later sues.

The eight mistakes founders make with advisor agreements

Over-granting to the first few advisors. The advisor pool is finite. Sizing the first grant at 1% because "this person is incredible" leaves you with nothing for the next five hires. Anchor the first grant at 0.15%–0.30% and reserve the 0.5%+ grants for advisors who are truly transformative — a hall-of-fame recruit in your exact segment who will unlock a specific commercial or capital outcome. · Vesting without a service condition. Standard templates vest monthly for 24 months regardless of whether the advisor keeps showing up. Add explicit language that vesting stops for material non-performance with a cure period. Without it, you are paying advisors in equity to disappear. · Skipping the 409A refresh before granting. Granting at a strike price based on a stale valuation invites IRS scrutiny and later cleanup costs. Get a current 409A before granting to any advisor whose stake is material. · Treating the advisor as a de facto employee. If they are running a function, sitting in the office daily, or working exclusively for you, they are an employee — hire them properly. The tax and wage-and-hour exposure from misclassification dwarfs anything you save by leaving them as an advisor. · Not writing the specific commitment into Section 1. "General strategic advice" is unenforceable. Named-account introductions, quarterly hours, and specific meeting counts are enforceable. The number of hours is less important than the fact that a number exists. · Leaving governing law blank or accepting the advisor''s state. A dispute goes wherever the agreement says it goes. Pick your state of incorporation, do not defer to the advisor. · Skipping the assignment fallback in Section 4.2. The "works made for hire" language protects most inventions, but not all. Add "and to the extent not properly characterized as works made for hire, Advisor hereby assigns…" as a second sentence. Belt and suspenders is free. · Renewing the term without renegotiating the equity. Match the term end to the vest end and treat both as a decision point. Either regrant with a new commitment or let the relationship transition to alumnus status. Auto-renewals without a fresh grant produce advisor benches full of people who have not been useful in three years.

What good advisor recruiting looks like around this document

The agreement itself is the last step in a longer process that decides whether the relationship is worth doing at all. The process worth following, in order:

Trial the advisor for 30–90 days before the grant. Have three or four working sessions on the specific problems you want their help on. Send them your board deck. Ask them to make one introduction. See if they show up, add value, and respect the confidentiality of what you share. About half of prospective advisors fail this trial silently — no follow-through, no substantive input, no intros. Do not grant equity to advisors who fail it. · Write the commitment clause first, then decide the grant. Force yourself to write Section 1 in specifics — "six meetings, ten hours per quarter, three named-account intros per quarter" — before deciding the equity number. If you cannot articulate what you want from the advisor, you are not ready to grant to them. · Reserve 1%–2% of the fully diluted cap for advisors total, and treat it as a portfolio. Assume you will end up with five to eight advisors across the company''s lifespan. Budget the pool that way, and each grant sizes itself against the alternatives you have not yet met. · Get a current 409A and price the grant against it. Refresh before granting. Document the board approval. File the option agreement properly. · Send this agreement, not a marketing template. Some advisor-recruiting platforms send simplified one-page agreements that skip half the confidentiality and IP protections. Send the full agreement, walk the advisor through Section 4 in a call, and answer their questions in writing. · Set a calendar reminder for month 20. Two months before vesting completes, decide whether to renew (new agreement, new grant, updated commitment) or let the relationship transition to alumnus status. Do not drift past the vest end.

The bottom line

An advisor grant is a call option on someone''s time, network, and judgment, paid in equity that dilutes every other stakeholder in the company. The advisory board agreement is the document that decides whether that option is worth exercising. The clauses that matter are the specific commitment in Section 1, the service condition on vesting in Section 3, the confidentiality and IP assignment in Section 4, and the governing law in Section 6.2. Fill the blanks specifically, add the service-based vesting language, refresh the 409A before granting, and set a calendar reminder for the vest-end review. Do those five things and the agreement does its job. Skip them and the advisor bench becomes exactly what most founders quietly regret — a diluted cap table with nobody on the other end of the phone.