# The Independent Board Seat Playbook: Skill Gap to a Tie-Checked Slate

*Turn a board roster into a seat brief, a public-records candidate pool, and a ranked 5 to 8 name slate with an independence verdict per name, in three weeks.*

- Canonical URL: https://www.refolk.ai/guides/independent-board-seat-slate-playbook
- Pillar: Investing and deal sourcing
- Format: Playbook
- Published: 2026-10-31
- Last reviewed: 2026-10-31
- Reading time: 21 min
- Keywords: how to find an independent board director for a startup, independent director search series a, how to check if a board candidate is truly independent, sourcing first-time board directors, VC platform help portfolio company recruit board member

## Key takeaways

- The NVCA model voting agreement fills the independent seat with someone not an Affiliate of the Company or of any Investor, and that Affiliate test only catches control relationships such as GP, officer, director, trustee, or same-adviser funds.
- In Sandys v. Pincus, the Delaware Supreme Court found reasonable doubt about independence from co-investments, shared boards, and a co-owned airplane, so a candidate who passes the contract's Affiliate test can still be captured.
- In Refolk's index, 21 of 25 sampled US profiles holding independent director titles lead their headline with an operator role, so searching on board titles mostly finds people who already hold seats and misses first-time directors.
- Regulation S-K Item 401(e)(2) only requires directorships that exist at filing time, so a tie check has to pull older proxies to recover past shared board seats with an investor's partners.
- Equity for independent directors falls fast with stage: a median one-time grant of 1.0% below $15M raised against a 0.29% median for later-stage companies paying equity only.
- Every name on the slate needs one of three written verdicts: independent, tied but disclosable, or captured, with the evidence rows that drove it.

This playbook is for platform partners, talent partners, and angels helping a portfolio company fill the independent seat its voting agreement already provides for. It takes the current board roster to a written seat brief, a candidate pool built from public filings and operator histories, a tie map for every name, and a ranked slate of 5 to 8 people with a written independence verdict for each, inside three weeks. Most advice on how to find an independent board director for a startup stops at "write a job description and ask your network." The tie check is what decides whether common stockholders accept the seat, so it gets its own stage here.

## What the voting agreement requires of the independent seat

Under the NVCA 2020 model voting agreement, the independent seat goes to "one individual not otherwise an Affiliate of the Company or of any Investor," approved under whichever of two bracketed options the parties adopted. The approval clause decides who can sit, not the job description, so read it before you source a single name.

The form offers two approval options:

- **Dual majority.** Approval by the holders of a majority of the shares held by the Key Holders who are still providing services to the company as officers, employees or consultants, and by the holders of a majority of the shares held by the Investors. Key Holders are the people the agreement lists as such, usually the founders.
- **Mutual acceptance.** The person must be "mutually acceptable to the other members of the Board." Proto Labs used this wording in a 2011 amendment to its voting agreement that named three independent directors.

The form defines "Affiliate" to include any general partner, managing member, officer, director or trustee of a person, and any venture capital fund controlled by the same general partners or advisers. That is a control test. It catches a partner at the lead investor and that partner's sister fund. It does not catch a co-investor, a former board colleague, or someone who co-owns an asset with a partner. The tie-check section covers those.

How the votes are counted matters as much as the wording. The NVCA form contemplates per share voting, not per capita voting (one vote per holder, whatever their stake). In Salamone v. Gorman, custom wording in a voting agreement led the Delaware Supreme Court to hold that designees were nominated and removable on a per capita basis. If the company's agreement departs from the form, the same slate can pass under one reading and fail under the other.

I did not verify the NVCA removal clause, so this playbook makes no claim about the removal threshold. Read it in the company's own agreement and put it in the brief next to the approval test.

> **Rule:** Write the approval test into the brief first
>
> Before any sourcing, the brief must name the adopted approval option, whether votes count per share or per capita, and the removal clause. A slate is not cleared until it has been checked against that exact clause.

### When the first independent usually joins

No published quantitative study says when startups seat their first independent director. What exists is practitioner opinion, and it clusters around the Series A:

- CRV says Series A is when independents typically become standard, and describes 1-1-1 or 2-2-1 boards (common, investor and independent seats, in that order).
- A healthcare venture investor puts the first independent around the Series A.
- Underscore recommends adding one between seed and Series A.
- Sharon Wilson, in a First Round Review field guide, puts it two to three years before an IPO.

An independent director search at Series A sits inside this range. Going earlier or later is a judgement call, not a breach of any standard.

## What the seat costs: equity, vesting, cadence, and hours

Set the grant, vesting, meeting cadence, and expected hours before you speak to anyone. Pay shrinks quickly as a company grows, and that changes which candidates will take the seat. Underscore also puts these decisions ahead of interviews.

| Source | Stage | Grant | Vesting |
|---|---|---|---|
| Startup pay survey (second-hand) | Under $15M raised | Median 1.0% one-time | 59% use four years |
| Bolster | Under $40M EV | 0.25% per year served | 100% on change of control |
| Westaway | Early | 0.25-1% fully diluted | Monthly over 2-4 years |
| Belenzon (BenchSci) | Series B+ | 0.1-0.125% fully diluted | Monthly over 2 years |

Read the table as a trend, not a price list:

- **Early stage is equity-led.** Below $15M raised, the survey data shows companies mostly pay equity only, with a median one-time grant of 1.0%.
- **Later stage splits.** In the same data, 33% of later-stage companies pay equity only (median 0.29%) and 29% pay cash only (median $50,000 retainer).
- **Acceleration is common.** Westaway says grants usually come with single-trigger acceleration, meaning vesting speeds up on a change of control alone.

On time, Wilson estimates about three to four weeks per year per company. Underscore says board meetings every two months or every six weeks are typical. Put both in the brief so candidates can rule themselves out before the first call.

This matters for sourcing as well as for the offer. A 0.29% grant or a cash retainer attracts different people than a 1.0% early grant. When the grant sits at the low end of the table, expect more interest from people who want the seat itself, often first-time directors. That is one more reason to search operator histories, covered below.

**Independent seat brief (one page)**

```
Company:
Stage and total raised:
Board today: (name, seat type: common / investor / independent)
Approval test: (dual majority / mutually acceptable to the other members of the Board)
Vote counting: (per share / per capita), clause reference:
Removal clause, clause reference:
Skill gap, ranked from director interviews:
  1.
  2.
  3.
Disqualifiers: (competitor board seats, ties either side has ruled out)
Meeting cadence: (e.g. every six weeks)
Expected time: (e.g. three to four weeks per year)
Grant: (% fully diluted, one-time or per year served)
Vesting and acceleration:
Committee expectations: (audit / compensation / none)
Sign-off owners: (for common: ___ ; for investors: ___ )
Slate due date:
```

*Fill every line before sourcing; a blank line means the brief is not ready to sign.*

## The three-week procedure: board roster to ranked slate

The procedure has seven stages: confirm the seat mechanics, interview the directors, write the brief, build the long list, tie-check every name, approach and rank, then hand over. To fit it into three weeks, you have to run the tie check alongside the long list and check names in batches as they come in.

#### Board roster to a cleared independent director slate

1. **Confirm the seat mechanics** - Read which approval option the voting agreement adopted and whether voting counts per share or per capita. About a day; done when the approval test is written into the brief.
2. **Interview the sitting directors** - Interview every current director on board structure, culture, strengths, and gaps. Two to three days; done when you have a ranked skill gap matrix.
3. **Write the seat brief** - Put the gap, cadence, hours, grant, vesting, and acceleration on one page. Done when founders and the lead investor have both signed it.
4. **Build the long list** - Pull names from S-1 and proxy Management sections, Form 990 Part VII, and operator histories. Three to five days; done when the list includes operators with no board title.
5. **Tie-check each candidate** - Map each name against every director, investor, affiliated fund, and portfolio board, including older filings. Three to four days, overlapping the long list; done when every name has a verdict.
6. **Make soft approaches and rank** - Approach only names that passed the tie check, confirm interest and fit, and rank. Three to four days; done when 5 to 8 names are in order.
7. **Hand over the slate** - Give the board the ranked slate with each tie record, then schedule board meetings with candidates. Done when feedback is collected right after each first meeting.

### Why three weeks, when published timelines run to months

Add up the stages and you get roughly 13 working days at the short end and 18 at the long end. Most of the gap closes when the tie check overlaps the long list. The brief has no published duration, so the one day here is my own planning figure.

Published search timelines are much longer, and they are worth knowing so the board does not mistake this plan for a full search:

| Phase | Board-search firm (FBCG) | Exec search (Richmond Group) |
|---|---|---|
| Definition | 3-6 wks | 2-3 wks |
| Sourcing | 4-10 wks | 4-6 wks |
| Interviews | 10-14 wks | 4-8 wks |
| Board-level total | not stated | 4-8 months |

Underscore says the search takes anywhere from a few weeks to a few months. The difference comes down to scope. These timelines run through the board's own interviews, while this playbook stops at handing over a tie-checked slate. A platform partner also starts with something an outside firm has to build: a working knowledge of the investor syndicate that every tie check runs against. How many candidates boards usually interview is not publicly established, so I make no claim about it.

### Where sources disagree on order

FBCG sends the prospectus to candidates before narrowing the list. RSR Partners narrows the list first and approaches only prioritised candidates. I follow RSR and put a tie check in front of any approach. No published source treats the tie check as a stage of its own; that is my addition. Approaching a candidate who turns out to be captured wastes a relationship and tells the market about a search the board has not finished defining.

## Where the candidate pool for a first independent director comes from

The pool comes from filings that name directors and operators, such as S-1 Management sections, proxy statements, and nonprofit Form 990s, and from operator career histories. Searching on board titles alone mostly returns people who already hold seats, and that is the same pool warm intros already reach.

In Refolk's index of professional profiles, the same title filters (Independent Director, Independent Board Member, Non-Executive Director) return these pools:

| Market | Matching profiles | Top headline in sample of 25 |
|---|---|---|
| United States | 3,388 | Independent Director / Independent Board Member (2 each) |
| United Kingdom | 2,077 | Non-Executive Director (4) |
| US:UK ratio | 1.63x (derived) | |

This is the only like-for-like pair I have. Use it as a sense of scale, not a market census.

**21 of 25 - US independent-director title matches whose headline leads with an operator role**

In a 25-profile sample from Refolk's index, only 4 put the board title first; the rest lead with roles such as Founder and CEO or Executive Chair.

Two things follow from the sample.

- **Title searches find incumbents.** Only about 16% of the sampled US matches lead with a board title. The people best suited to sourcing first-time board directors (operators who have scaled the function the board lacks) usually carry no board title at all, so a title search cannot find them.
- **The title pool leans toward life sciences.** Six of the top ten employers in the US sample were life-sciences companies or investors: Vaxcyte, Pliant, Bicara, Exo, Fathom, and MPM BioImpact. Biotech venture practice puts board titles on profiles more often. A software company's search therefore needs queries built on operator history.

When I ran the same titles filtered to US software and internet industries, I got back 1 profile. That reflects a mismatch in industry labels, not a real count. Always compare counts with the filter on and off.

### What each public record proves, and where it lies

| Record | What it proves | Where it lies |
|---|---|---|
| Item 401(e)(1), proxy or S-1 | Five years of employers, and whether each is an affiliate of the filer | Stops at five years |
| Item 401(e)(2) | Other public directorships | Current seats only; past seats vanish |
| Item 401(d) | Family relationships | Only what that filer discloses |
| Item 407(a)(3) | Transactions the board weighed on independence | The board's own judgement |
| Form 990 Part VII | Nonprofit seat, title, hours, pay | Omits advisory boards; former directors only above $10,000 |
| Investor portfolio pages, press releases | Private co-investments and boards | Incomplete; method unverified |

Item 407(a) also names which of a public company's directors the board considers independent. That gives you a ready list of people another board has already accepted as independent. Form 990 Part VII lists average hours per week and pay from the organisation and related organisations, so it shows whether a nonprofit seat is real work or a name on a letterhead.

Writing an operator-history query is the slow part: "former CFO of a SaaS company that went public, no current public board seat" means cross-referencing filings with career records by hand. [Refolk](/) takes that sentence as written and returns matching people across LinkedIn, GitHub, and the open web, which makes it practical to build pools that warm intros never reach.

I ran this search: `Former CFOs of US B2B SaaS companies that went public in the last five years who now hold no public board seat` - [see the full result list](https://www.refolk.ai/s/4zqmhe377q).

*Returns finance operators with public-company experience and no current listed directorship, a first-time director pool that title searches miss.*

Other pool angles worth running, depending on the gap in the brief:

- VP Sales or CRO operators in London who scaled ARR past $50M and sit on a nonprofit board.
- Directors listed in Form 990 filings of healthcare nonprofits who were previously COOs at venture-backed digital health companies.
- Independent directors named in recent S-1 filings of fintech companies, excluding partners at venture firms.
- Former general counsels of US enterprise software companies who serve as audit or compensation committee members.

## How to check if a board candidate is truly independent

Map each candidate against every sitting director, every investor, each investor's affiliated funds, and those funds' portfolio boards, then record one of three verdicts: independent, tied but disclosable, or captured. The contract's Affiliate test is the floor, not the standard.

#### Tie check for one candidate

1. **Roster** - List every director, investor, affiliated fund, and those funds' portfolio boards
2. **Current filings** - Read the latest proxy or S-1 for five-year employers and current public seats
3. **Older filings** - Pull earlier proxies to recover directorships that have since ended
4. **Nonprofit and private** - Check Form 990 Part VII and investor portfolio pages for shared seats and co-investments
5. **Antitrust screen** - Rule out a competitor board seat above the Section 8 thresholds
6. **Verdict** - Record independent, tied but disclosable, or captured with the evidence

*A tie check runs from the board's full roster through current and older filings to a single written verdict per candidate.*

### Why the contract test is not enough

In Sandys v. Pincus, the Delaware Supreme Court found that "a network of business relationships and venture capital investments between two other directors and the controlling stockholder raised a reasonable doubt" about independence. The venture firm held 9.2% of Zynga's equity and had interlocking ties outside Zynga. The ties included co-investment and the firm investing in a company where another director sat. Co-owning an airplane also counted. The Affiliate definition would catch none of these, because none of them is a control relationship.

#### What each independence test catches

1. **Mutual acceptance** - The approval vote itself, counted per share or per capita as the clause says
2. **Delaware relationship test** - Co-investments, shared boards, and shared assets that together raise reasonable doubt
3. **NVCA Affiliate test** - GP, managing member, officer, director or trustee roles, and same-adviser funds
4. **Section 8 screen** - A competitor board seat above the thresholds, which is a strict-liability bar

*A candidate who passes the narrow contract test can still fail the wider Delaware test, so the tie map has to cover every layer.*

Section 8 of the Clayton Act bars interlocks between competitors above set dollar thresholds. It does not apply if the competitive sales of either corporation are less than two percent of that corporation's total sales. Counsel makes this call, but the tie map has to bring it up.

### The three verdicts

| Verdict | What it means | What the board does |
|---|---|---|
| Independent | No Affiliate role, no material tie in filings or portfolio records, no competitor seat | Slate it |
| Tied but disclosable | A non-control tie, such as a past shared seat or one co-investment, fully documented | Slate it with the tie disclosed to both sides in writing |
| Captured | An Affiliate under the contract, a Sandys-style web of ties, or a Section 8 interlock | Drop it and record why |

No public source sets the line between disclosable and captured. I treat a single documented, non-control tie as disclosable. I treat several ties pointing at the same investor as captured, because that pattern matches what Sandys found. Where the dossier is silent, say so. I found no source on how a candidate's LP commitment to an investor's fund, or reporting to a CEO backed by the same investor, should be treated. Log these as unresolved, default them to tied but disclosable, and get counsel's view.

**Tie map record (one per candidate)**

```
Candidate:
Sources checked: latest proxy or S-1 (date) | older proxies (dates) | Form 990s (years) | investor portfolio pages | press releases
Against director: <name> | relationship: | when: | source:
Against investor: <firm> | relationship: | when: | source:
Against affiliated fund: <fund> | relationship: | when: | source:
Against portfolio board: <company> | relationship: | when: | source:
Competitor board seats: none found / <company> | Section 8 view from counsel:
Unresolved: (LP commitments, shared assets, ties alleged but not documented)
Verdict: independent / tied but disclosable / captured
Reason, one sentence:
Checked by, date:
```

*One line per relationship found; the verdict must cite the lines that drove it.*

> The Affiliate test catches control. Common stockholders are worried about loyalty, and no definition captures that.

## How independence checks go wrong

Most failures come from treating a source as complete when it is not: a proxy that forgets past seats, a filing that cannot see private deals, a search that only finds people with board titles. Each failure below gives what it looks like and the check that catches it.

1. **Treating the latest proxy as the full directorship history.** Item 401(e)(2) only requires directorships that exist at filing time. A seat the candidate shared with an investor's partner a few years ago is missing from the latest proxy. *Check:* pull older proxies from the same filer and read each Management section.
2. **Assuming filings show private ties.** Co-investments and private-company boards do not appear in SEC filings. In Sandys, the ties were alleged by a plaintiff, not taken from filings. *Check:* investor portfolio pages and press releases, plus a direct question to the candidate. I have not verified a method that makes this complete, so record it in the tie map as a known gap.
3. **Missing operators because the search used board titles.** In Refolk's index, 21 of 25 US matches lead with an operator title. A search on "Independent Director" returns people who already hold seats and misses first-time directors. *Check:* every long list should include names found through operator-history queries.
4. **Trusting a pool size from a filtered query.** Industry filters fail without warning. The US software-filtered title query returned 1 profile. *Check:* run every count with the filter on and off before quoting a pool size to the board.
5. **Counting the approval the wrong way.** Per share and per capita counting give different results, as Salamone showed. *Check:* reread the exact clause before you call the slate cleared, and ask counsel if it departs from the NVCA form.
6. **Reading an advisory seat as a directorship.** Form 990 does not report advisory board members as directors, so a candidate's nonprofit "board" may be advisory only. *Check:* confirm the name appears in Part VII. Former directors appear only above $10,000 of reportable pay, so a missing former seat does not prove there never was one.
7. **Missing a Section 8 interlock.** The candidate sits on a competitor's board above the thresholds. This is strict liability. *Check:* list every current board seat, compare against the company's competitors, and have counsel apply the current FTC thresholds and the two percent exemption.
8. **Independent in name only.** CRV warns that a founder can lose a vote when investor directors "and a nominally 'independent' director align against them." A candidate can pass every document check and still owe their deal flow to the lead investor. *Check:* ask the candidate how they got their past seats and who introduced them, and record the answer in the tie map.

> **Watch out:** Contract-clean is not independent
>
> The Affiliate definition only reaches control relationships. Sandys found reasonable doubt from co-investments, shared boards, and a co-owned airplane, none of which the Affiliate test would catch.

## Ranking the slate and handing it to the board

Rank tie-checked candidates on two things: how well they fill the brief's top skill gap, and how clean their tie record is. Only names that are both on-brief and independent, or on-brief with a disclosable tie both sides accept, go on the 5 to 8 name slate.

#### Fit against independence

Horizontal axis runs from Weak fit to the skill gap to Strong fit to the skill gap. Vertical axis runs from Captured or heavily tied to Independent.

| Quadrant | What it means |
| --- | --- |
| Clean but off-brief | Keep on the bench for a later seat or committee role |
| Clean and on-brief | Rank at the top of the slate |
| Tied and off-brief | Drop and record why |
| Tied but on-brief | Slate only if the tie is disclosable and both sides sign it |

*The top of the slate is reserved for candidates who fill the skill gap and carry no undisclosed tie to an investor.*

Soft approaches come after the tie check, never before. Keep the first contact short: name the company stage, the gap, the cadence, and the time commitment from the brief, and ask whether the candidate is open to a conversation. Someone who cannot commit three to four weeks a year should rule themselves out at this point, not after a board meeting.

Within each quadrant, rank by:

- **Gap fit.** Has the candidate done the specific work the directors named as missing, at a comparable stage?
- **Availability.** Do current seats and the day job leave room for the cadence in the brief?
- **Acceptability to common.** Would the founders see this person as theirs as much as the investors'? The tie record answers part of this, and the founders answer the rest.

Hand the board the ranked list, each candidate's tie record and verdict, and the brief they were measured against. RSR describes the board giving immediate feedback after the first meeting with a candidate, so schedule a short debrief straight after each one.

#### Before the slate goes to the board

- [ ] The brief names the adopted approval option, the per share or per capita counting method, and the removal clause.
- [ ] Every sitting director was interviewed and the skill gap is ranked in writing.
- [ ] Grant, vesting, acceleration, cadence, and expected hours are stated in the signed brief.
- [ ] The long list includes names found through operator-history queries, not only board titles.
- [ ] Every pool count was checked with industry filters on and off.
- [ ] Each candidate has a tie record covering every director, investor, affiliated fund, and portfolio board.
- [ ] Older proxies were pulled for every candidate with a public directorship history.
- [ ] Every claimed nonprofit board seat was confirmed in Form 990 Part VII.
- [ ] Counsel has reviewed each name's current board seats against the Section 8 thresholds.
- [ ] Every name on the slate has a verdict and a one-sentence reason, and no captured name remains.
- [ ] The slate holds 5 to 8 names in ranked order.

## Keeping the tie map and bench current

A tie map is only true on the day it was checked. Re-run it whenever the investor syndicate, the candidate's seats, or the antitrust thresholds change, and keep the names you did not slate as a bench for the next seat.

Re-check when any of these happen:

- **A new round closes.** New investors bring new affiliated funds and portfolio boards. Re-run the tie map for the seated independent director and every bench name.
- **The FTC updates Section 8 thresholds.** The FTC republishes the interlocking directorate thresholds every year. Read the current figures on its interlocking directorates page instead of relying on a number you noted earlier.
- **A director or candidate takes a new seat.** One new board seat can create a competitor interlock or a shared seat with an investor's partner.
- **The company changes stage.** Benchmark the grant against the equity table again. Below $15M raised and at later stage are different markets for candidates.
- **Your pool queries drift.** Industry labels shift, so re-run counts with filters on and off before each new search.

Keep the clean but off-brief names from the matrix. They are the first call for the next independent seat or a committee vacancy. Re-running the original pool queries in Refolk before that search starts surfaces operators whose careers have moved since the last slate, so the bench is not limited to whoever answered last time.

## Frequently asked questions

### How do I check if a board candidate is truly independent?

Map the candidate against every sitting director, every investor, each investor's affiliated funds, and those funds' portfolio boards, then give a written verdict: independent, tied but disclosable, or captured. Use current and older proxies, S-1 Management sections, and Form 990 Part VII for public ties, and investor portfolio pages and press releases for private ones. Passing the contract's Affiliate test is not enough, because Delaware courts have counted co-investments and shared assets as ties.

### When should a startup add its first independent director?

Most practitioners put it around the Series A, but no published quantitative study establishes a standard. CRV says independents typically become standard at Series A, Underscore recommends adding one between seed and Series A, and Sharon Wilson suggests two to three years before an IPO. Treat timing as a judgement call driven by the board's skill gap and the voting agreement, not a fixed rule.

### How much equity does an independent director get at Series A?

Published benchmarks for early-stage companies range from about 0.25% to 1% of fully diluted shares. Survey data quoted second-hand shows a median one-time grant of 1.0% for companies under $15M raised, with 59% using four-year vesting. Bolster suggests 0.25% per year served below $40M enterprise value. Later-stage grants fall to around 0.1% to 0.29%, and some companies pay a cash retainer instead.

### Can a VC platform team run the independent director search for a portfolio company?

Yes, and the platform or talent partner is well placed to run it because they already know the investor syndicate the tie check has to cover. The risk is that the slate comes only from the lead investor's network, which is exactly what common stockholders distrust. Build the pool from public filings and operator histories, and document a tie verdict per name so the founders can see the check was done.

### Do LP commitments make a board candidate non-independent?

That is not publicly established. I found no source on how a candidate's LP commitment to an investor's fund, or reporting to a CEO backed by the same investor, is treated. Record any such relationship in the tie map, default it to tied but disclosable, and ask the company's counsel for a view before the slate goes to the board.

### How long does an independent director search take?

Published timelines run from a few weeks to several months. Underscore says a few weeks to a few months, and one board-search firm lists 3 to 6 weeks of needs assessment, 4 to 10 weeks of search, and 10 to 14 weeks of interviews. Reaching a tie-checked slate in three weeks is possible when the platform partner already knows the syndicate and overlaps sourcing with the tie check.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/guides/independent-board-seat-slate-playbook*