Refolk
FrameworkInvesting and deal sourcing

The Advisor-Slide Read: Working Advisor, Nominal Name, or Unconfirmed

Score every name on a deck's advisor slide on five dimensions and write a verdict of Working Advisor, Nominal Name, or Unconfirmed that a second reviewer reproduces.

21 min readLast reviewed November 4, 2026Read as Markdown

Key takeaways

  • No public filing can confirm a startup advisor: Form D Item 3 covers only executive officers, directors, and promoters, and UK company law keeps advisors off the Companies House officer register.
  • The FAST template's standard advisor grant runs 2.0 to 2.4 times the observed median grant at every stage, so a deck quoting FAST numbers is quoting the high end.
  • Only 10% of pre-seed advisors receive 1% or more of the company, so a 1% pre-seed advisor claim needs a recorded answer on what scope was agreed.
  • In Refolk's index, 760 US profiles call themselves startup advisors, and in a 25-profile sample most visible current titles were full-time founder, CEO, CTO, partner, or consultant jobs, so 'advisor' is mostly a side label.
  • Count only advisory roles started in the last 24 months, the length of the FAST vesting window, and label any load cutoff a judgement call, because no one has published a measured distribution.
  • A name earns a Working Advisor verdict only when the relationship is confirmed by the advisor's profile, a signed document, or the advisor's own words, and there is a specific, dated act from the last 90 days.

Use this when a seed deck puts five impressive names on the advisor slide and you need to know, before IC, which of them actually work with the company. It is for the associates, platform and talent partners, and angels who write the memo. It gives you five scored dimensions, one gating rule, and a verdict for each name (Working Advisor, Nominal Name, or Unconfirmed). The rules are tight enough that a second reviewer reading the same evidence should reach the same answer.

What does a name on an advisor slide actually prove?

On its own, a name on an advisor slide proves only that the founder put it there. Treat each name as a separate claim with its own evidence trail. Don't expect a public filing to confirm any of them: the law keeps advisors out of governance records on purpose.

In the US, Item 3 of Form D lists "executive officers, directors, and persons performing similar functions for the issuer, as well as persons who have functioned as a promoter of the issuer within the prior five years." "Advisor" is not one of the categories. The EDGAR Form D data file records each related person's relationship only as Executive Officer, Director, or Promoter. Title alone does not decide it, so someone called an advisor who in practice acts as a director can still appear on the form. An ordinary working advisor will not.

In the UK, section 251 of the Companies Act 2006 says a person is not a shadow director just because the directors act on advice he gives in a professional capacity. So advisors don't appear on the Companies House officer register either.

That leaves three routes that can confirm the relationship:

  1. The advisor's own public profile.
  2. Signed documents in the data room: the advisor agreement, a board resolution approving the grant, and the cap-table line.
  3. The advisor's own words, on a short call.

Checklists written for founders already say advisor names should "reflect genuine relationships where the advisor has actually spent meaningful time with the company." Nobody enforces that. An annotated collection of real advisor slides says the gap most of them leave open is stating what the advisor actually does. Nobody has published what share of named advisors list the company on their own profile. With no base rate to lean on, you grade every name from scratch.

SourceWhat it can showWhat it cannot show
Form D, Item 3Officers, directors, promotersAny advisor role
Companies House officersDirectorsAdvisors (s251)
Advisor's own profileRole and start dateWhether work continues
Data room documentsAgreement, resolution, equityActivity after signing
Short callA named recent actProof, unless corroborated

The advisor evidence ladder

  1. Deck claim
    The founder puts the name on the slide
  2. Self-listing
    The advisor lists the company on their own profile
  3. Signed documents
    Agreement, board resolution, and cap-table line in the data room
  4. Advisor's account
    The advisor confirms the role and its scope in their own words
  5. Dated work
    Artifacts and named acts from the last 90 days
Each layer down is harder to fake than the one above it, and only the bottom layer shows the relationship is live today.

The five dimensions and how to score each one

Score each name 0, 1, or 2 on five dimensions: self-confirmation, dated start, concurrent advisory load, role-to-domain fit, and evidence of work. The maximum is 10. The bands below are narrow so that two reviewers working from the same record land on the same number.

Dimension2 points1 point0 points
Self-confirmationListed as current, with start dateListed undated, or signed agreement onlyNot listed, ended, or nothing
Dated startTwo sources agreeOne sourceNo date, or after the deck
Concurrent load1 to 3 roles in 24 months4 to 6 roles7 or more roles
Role-to-domain fitCovers the top riskCovers a secondary riskCovers no named risk
Evidence of work2+ dated acts in 90 daysOne act or an older artifactPraise only

Self-confirmation

This shows that the advisor, not only the founder, is willing to put their name to the relationship in public. It misleads when the advisor added the line as a favour, or drifted away and never removed it. An undated listing caps at 1 because you can't tell which of those you're looking at.

The friction here is finding the right profile. A common name might have three possible LinkedIn pages, a GitHub account under a handle, and a personal site. Refolk lets me ask for the person by name and company in plain English and get their profiles back from LinkedIn, GitHub, and the open web in one pass.

Dated start

A start date gives the relationship a beginning you can reason from. With a date you can work out how far into a two-year vesting schedule the advisor is, and whether the role came before the raise or was put together for it. The Founder Institute recommends "working with a potential advisor for at least a month, and spending at least eight hours together" before signing. A start date a few weeks before the deck doesn't disqualify anyone, but ask what happened in that first month. A date can mislead if it was typed in loosely or the agreement was backdated. Two independent sources that agree are your defence.

Concurrent advisory load

Load tells you whether the advisor has hours to give. Count only roles that started in the last 24 months, which is the length of the FAST vesting window. Load misleads both ways. Lapsed roles padding a profile make a working advisor look like a collector. A short role list can hide a full-time job. The thresholds get their own section below.

Role-to-domain fit

Fit shows whether the advisor can reduce the risk an investor actually worries about. Before you score anyone, write the company's top risk in one line, such as regulatory clearance, an enterprise sales motion, or a hard technical problem. Fit misleads when a famous generalist sits on the slide of a company whose main risk is regulatory. Fame is not fit.

Evidence of work

This is the only dimension that shows the relationship is live now. Ian Paterson describes a working advisor as someone who "sends introductions you did not ask for... amplify your messaging, open the doors inside their mandate, and bring you talent." The traces that work leaves behind are joint posts, referred hires, joint appearances, and contributions to the company's repos or papers. It misleads when a first-month artifact is presented as current. Nobody has published how fast each kind of artifact goes stale after an advisor disengages. The 90-day window is my working choice, and it matches the call question.

How much equity should a seed advisor claim?

Grants seen in practice run at about half the FAST template. At every stage, the template's standard tier is 2.0 to 2.4 times the observed median. A deck quoting FAST numbers is quoting the high end, and a 1% claim at pre-seed puts the advisor in the top decile.

The Founder Institute's FAST advisor agreement sets a grant from two inputs. One is company maturity (idea, startup, or growth). The other is the advisor's engagement level (standard, strategic, or expert). An expert-level advisor at an early-stage startup gets 1%, vesting over two years. The same engagement at a growth-stage company gets 0.6%. Vesting has a three-month cliff. The template also notes that a technology startup often sets aside a 5% pool for a group of strategic advisors.

StageFAST standardObserved medianTemplate ÷ median
Pre-seed / idea0.5%0.21%2.4x
Seed0.25%0.12%2.1x
Series A / growth0.1%0.05%2.0x
2.4x
FAST standard advisor grant versus the observed pre-seed median
The gap holds at every stage, from 2.0x to 2.4x, so the template works as a ceiling rather than a norm.

Read this table with its limits in mind. The observed medians are Carta's published figures for one half-year, measured as a share of fully diluted shares. A secondary source says they rest on nearly 5,000 advisor grants. The pre-seed median has drifted down from 0.25% to 0.21%. Only 10% of pre-seed advisors received 1% or more. The FAST standard column comes from a third-party mapping, and its stage labels have not been checked against the current template text. Third-party summaries of the FAST range disagree (0.25 to 1%, 0.15 to 1%, 0.10 to 1.00%), and the percentages have changed between versions. Before you cite a number in a memo, pull the current matrix from the Founder Institute's FAST page.

Equity is not one of the five scores. Use it as an anchor check. A stated grant well above the median means one of two things: unusually deep engagement, which should show up under evidence of work, or an inflated grant. Ask what scope was agreed and write the answer down.

How many concurrent advisory roles is too many?

Nobody has measured it, so any cutoff is a judgement call, mine included. I score 1 to 3 advisory roles started in the last 24 months (this one included) as 2, 4 to 6 as 1, and 7 or more as 0. Then I drop one band if the advisor also holds a full-time operating or investing job.

The cutoffs practitioners publish come with no methodology. One warns about advisors "spread across 20+ boards". That catches out-and-out name collectors but is far too loose to test for a working advisor. Another treats an "advisory board larger than 5 people" as a red flag, but that counts the company's board, not the advisor's roles. My bands sit well below the 20-board warning because the evidence-of-work bar asks for dated acts within a 90-day window, and every extra live role competes for the same hours. If your fund prefers different cutoffs, write them down and apply them to every name. Two reviewers can only agree if they use the same rule.

The full-time-job adjustment comes from Refolk's own data. In Refolk's index of professional profiles, the self-label "startup advisor" turns out to be mostly a side title worn alongside a full-time job.

760
US profiles in Refolk's index carrying an Advisor or Startup Advisor title with the keyword "startup advisor"
In a 25-profile sample, at least 5 visible current titles contained "Advisor"; the rest were founder, CEO, CTO, partner, or consultant roles.
MarketMatching profilesAdvisor-titled in sample of 25Share (floor)
United States760at least 5at least 20%
United Kingdom91at least 1at least 4%

The US count is 8.4 times the UK count. Only the top 10 titles in each sample were visible, so the shares are floors, not estimates. In the UK sample the one visible advisor title was "Board Advisor". The practical reading is that most people who call themselves startup advisors already have a job competing for the same hours. That's why a full-time role costs a band.

Counting roles is still a manual read of each profile. Refolk's index gives aggregate counts, not a per-person tally of advisory roles. Where Refolk helps is calibration: when I want to see what a heavy load looks like in one market, I ask for it in plain English and open the profiles that come back.

Turning five scores into a verdict

Confirmation is a gate, and the total only matters once a name passes it. A name is Unconfirmed if none of the three routes confirms a current relationship. It is a Working Advisor if it is confirmed, scores 2 on evidence of work, scores at least 1 on fit, and totals 7 or more. Every other confirmed name is a Nominal Name.

From five scores to one verdict

  1. Gate
    Is a current relationship confirmed by profile, document, or the advisor's own words? If not, Unconfirmed
  2. Work test
    Does evidence of work score 2? If not, Nominal Name
  3. Fit test
    Does role-to-domain fit score at least 1? If not, Nominal Name
  4. Total
    Is the total 7 or more out of 10? If not, Nominal Name
  5. Verdict
    Passed every gate: Working Advisor
The verdict is decided by gates in a fixed order, so two reviewers reading the same record cannot reach different answers.
VerdictConditionsWeight in the memo
Working AdvisorConfirmed, work 2, fit ≥1, total ≥7Cite against the risk it covers
Nominal NameConfirmed, fails any other conditionList as network only
UnconfirmedNo route confirmsDo not cite; open founder question

A Working Advisor goes into the memo next to the specific risk they reduce, with the dated act that proves it. A Nominal Name is a real relationship that isn't doing much yet. It can fairly go in a line about the founder's network, but it carries no weight against any risk. An Unconfirmed name stays out of the risk analysis completely. List it as an open question for the founder, and ask for the agreement and an introduction.

Confirmation against live work

Recent dated workNo recent dated work
Help without paper
Unconfirmed until the founder produces an agreement or the advisor confirms
Working Advisor
Cite against the specific risk the advisor covers
Unconfirmed
Leave out of the memo and ask the founder directly
Nominal Name
Mention as network only, with no weight on risk
Not confirmedConfirmed
A name needs both a confirmed relationship and recent dated work before it earns weight in a memo.

The top-left quadrant matters more than it looks. Someone can be clearly helping, with joint posts and a referred hire, and still have no agreement and no profile line. That is a real relationship the founder hasn't formalised. It stays Unconfirmed until one of the three routes closes it, and it usually takes one email to the founder.

How to grade a five-name advisor slide, step by step

Run these eight steps in order, filling one record per name. The procedure's own estimates are about 5 minutes to break the slide into claims and about 5 minutes per name for the profile check. Neither has been verified, so time your first few slides and adjust.

The advisor-slide read

  1. Turn the slide into claims
    List every advisor as one row with name, stated role, and any stated equity or time. Done when each name has its own row.
  2. Check self-confirmation
    Find the advisor's own profile and record whether it lists the company as current with a start date. Done when each row reads yes, no, or no profile, plus the date.
  3. Search records and request documents
    Search Form D Item 3 and Companies House, then request the advisor agreement, board resolution, and cap-table line. Done when each row reads documented, undocumented, or listed as a director.
  4. Count concurrent advisory load
    Count advisory roles on the profile that started in the last 24 months and note any full-time job. Done when each row has a number and a load score.
  5. Test role-to-domain fit
    Write the company's top risk in one line and compare each advisor's career to it. Done when each row reads fit or no fit with a one-line reason.
  6. Collect evidence of work
    Search for joint posts, referred hires, joint appearances, and repo or paper contributions. Done when each row lists dated artifacts or records none.
  7. Make the short call
    Ask the advisor what they did for the company in the last 90 days. Done when a specific act is named and written down, or the call is recorded as praise only.
  8. Write the verdict and get a second review
    Apply the gate and verdict rules and write one memo line per name. Done when a second reviewer, using only the record, reaches the same verdict.

When to make the call, and how many

Sources disagree on timing. Some put backchannel calls "between the partner meeting and the term sheet". Others note that "some firms start backchannel work earlier, during the first or second meeting, before the founder knows." Advisor calls compete for the same reference budget as everything else, and that budget varies a lot by investor type.

Investor typeReference calls per deal
Angels1 to 3
Institutional average10
Top-tier firms10 to 20+

The institutional average comes from a survey of 885 venture capitalists. If you are an angel with one to three calls, don't spend them on every advisor. Call the one whose fit score matters most to the company's top risk, and grade the rest from profiles and documents alone. Without a call, their evidence-of-work score rests on public artifacts.

No firm has published a standard advisor-confirmation script. This is the one I use. The question that produces a scoreable answer is the 90-day question. Everything else is context.

Advisor confirmation call (about 10 minutes)
Opening: Thanks for taking this. I am looking at [Company] ahead of an investment decision, and you are named as an advisor. I have five short questions.

1. Are you currently advising [Company]? When did that start?
2. What did you agree to help with, and is there a written agreement?
3. What is one specific thing you did for [Company] in the last 90 days?
4. Who at the company was involved in that, and what came of it?
5. Roughly how many companies are you actively advising right now?

Close: Is there anything about the company you would want an investor to know?

Record: date of call | who called | act named (verbatim) | person to corroborate with

Replace [Company] and keep question 3 word for word; write down the act exactly as the advisor states it.

To corroborate an act named on the call, ask the founder about the same act without saying where you heard it. A referred hire, an intro, or a joint appearance should be just as specific from the other side.

How advisor checks go wrong

Most advisor-slide errors come from misreading evidence that is technically true. A profile line, a cap-table entry, or a warm call can each look like confirmation of live work when it only shows that a relationship once existed. These are the failure modes to check on every name.

  1. The favour listing or stale listing. The profile lists the company because the advisor added it as a favour, or because they disengaged long ago and never removed it. Check whether a start date was entered, whether the role still shows as current, and whether any dated artifact exists after the start date.

  2. Reading absence from Form D as a red flag. Advisors are left off Form D by design, so absence is what you should expect. The real false positive runs the other way: if a name on the advisor slide appears on Form D as a director, that person is not an advisor in the ordinary sense. Record them as a director and say so in the memo.

  3. Treating cap-table presence as proof of work. Advisor equity vests monthly. Under FAST, an advisor who disengages after four months keeps roughly four months of equity, and advisors who stop advising typically do not voluntarily return unvested equity. The cap-table line outlasts the work. Check for dated activity, not ownership.

  4. Taking the equity number at face value. A 1% claim at pre-seed puts the advisor in the top decile. That means either unusual engagement or an inflated grant. Ask what scope was agreed, and expect the evidence-of-work score to match the size of the grant.

  5. Counting padded load. The headline says "startup advisor", but the role list is padded with lapsed roles. Count only roles with a start date in the last 24 months. The reverse error also happens: a short list next to a full-time job can still mean the advisor has few hours to give.

  6. Accepting praise as evidence. The advisor praises the founder warmly but can't name an act. As one practitioner puts it, "When the reference can't name a single decision... you are looking at a red flag." Score praise-only calls 0 on evidence of work, however senior the advisor.

  7. Confusing fame with fit. A famous generalist matched against a regulated-domain risk is the classic case. Check whether the advisor's career actually covers the company's top risk, as written in one line before scoring.

  8. Treating an empty filtered search as absence. In Refolk's index, adding a CXO, VP, or Owner seniority filter to the US "startup advisor" query returned 0 profiles. That is a quirk of the filter, not a finding. If a narrow search comes back empty, rerun it without the filter before you conclude a person or profile doesn't exist.

A cap-table line proves a grant was made. It does not prove anyone still picks up the phone.

Before the memo goes to IC, and keeping the read current

Before the memo goes out, check that every name has a verdict a second reviewer can reproduce from the written record alone, without your call notes in their head. Then plan to re-run the read at the next decision point, because a Working Advisor today can be a Nominal Name in two quarters.

Per-name advisor record
ADVISOR READ | Company: [Company] | Round: [round] | Reviewer: [initials] | Date: [date]
Name:
Stated role on slide:
Stated equity or time:
Company's top risk (one line):

1 Self-confirmation (0/1/2): __ | Evidence:
2 Dated start (0/1/2): __ | Sources and dates:
3 Concurrent load (0/1/2): __ | Roles started in last 24 months: __ | Full-time job: Y/N
4 Role-to-domain fit (0/1/2): __ | Reason:
5 Evidence of work (0/1/2): __ | Dated artifacts or act named on call:

Gate: confirmed by profile / document / call / none
Form D or Companies House listing: none / director / other
Total: __ / 10
Verdict: Working Advisor / Nominal Name / Unconfirmed
Memo line:
Second reviewer verdict: __ | Match: Y/N

One block per name on the slide; the second reviewer fills the last line using only what is written above it.

Before the advisor read goes into the IC memo

  • Every name on the advisor slide has its own record and its own verdict.
  • Each verdict states which route confirmed the relationship, or that none did.
  • No name was marked down for being absent from Form D or Companies House.
  • Any name listed on Form D as a director is recorded as a director, not an advisor.
  • Concurrent load counts only roles started in the last 24 months, with full-time jobs noted.
  • The company's top risk is written in one line and every fit score refers to it.
  • Every Working Advisor verdict cites at least one dated act from the last 90 days.
  • Any pre-seed equity claim of 1% or more has a recorded answer on agreed scope.
  • A second reviewer reached the same verdict on every name from the record alone.
  • The memo cites Nominal Names as network only and does not cite Unconfirmed names at all.

Keeping the read current takes three habits:

  • Re-run evidence of work at each new decision point. Nobody has published how fast advisor artifacts go stale, so don't carry a verdict forward. Refresh the 90-day window before a follow-on or a pro rata decision.
  • Re-pull the equity anchors before citing them. FAST percentages have changed between versions, and the observed medians are snapshots that have already drifted. Check the current FAST matrix and the latest published medians, and update the table in your own template.
  • Write down your load cutoffs. Mine are a judgement call. If your fund adopts different bands, put them in the rubric so every associate grades against the same line and second reviews keep matching.

Questions practitioners ask

How do I verify a startup advisor is actually involved before investing?

First confirm the relationship through one of three routes, then test for recent work. The routes are the advisor's own profile listing the company with a start date, signed documents in the data room, or the advisor confirming it on a short call. Then look for dated artifacts such as referred hires, joint posts, or repo contributions, and ask the advisor what they did in the last 90 days. The strongest single signal is a specific act they can name.

Is it a red flag if a named advisor is not on the Form D?

No. You should expect advisors to be missing from Form D. Item 3 lists only executive officers, directors, and promoters, and the EDGAR data file records relationships only in those three categories. The real false positive runs the other way. If a name from the advisor slide appears on Form D as a director, that person holds a different status from advisor, and the memo should say so.

What is the standard equity range for a startup advisor?

Grants seen in practice sit well below the template. Carta's published medians were 0.21% of fully diluted shares at pre-seed, 0.12% at seed, and 0.05% at Series A. The FAST template's standard tier runs about 2.0 to 2.4 times those figures. Only 10% of pre-seed advisors received 1% or more. The template's percentages have changed between versions, so check the current FAST matrix before you cite a range.

Do I need the founder's permission to contact a named advisor?

Common venture practice says no. Some firms start backchannel work during the first or second meeting, before the founder knows. In the US, searching public records yourself is not a consumer report and needs no consent. In the UK, the ICO's reference guidance is written for employment, and nobody has verified that it applies to advisor confirmation, so check with counsel. On this call the advisor is describing their own role, not acting as a referee.

How many advisory roles is too many for one advisor?

Nobody has measured it, so any number is a judgement call. Published warnings range from advisors spread across 20 or more boards to board-size rules that measure something else entirely. I give full marks for 1 to 3 roles started in the last 24 months, partial marks for 4 to 6, and zero for 7 or more. A full-time job drops the score one band. Count only recent roles, because lapsed listings inflate the total.

Does an advisor's line on the cap table prove they are still working?

No. Equity lags behind the work. Advisor grants vest monthly, so an advisor who walks away after four months keeps roughly four months of equity, and advisors who stop advising usually don't hand back unvested shares. The cap-table line outlives the work. Use it to confirm that a grant was made, then look for dated activity from the last 90 days.

Try it on the search you came here for

Stop building boolean strings. Just describe the person.

Type one sentence. I plan the search, read GitHub, public LinkedIn and Crunchbase records, and the open web as it is right now, and hand back a ranked list with the reason next to every name.

  1. 01Describe them

    One plain sentence. Role, city, stack, stage, whatever matters to you.

  2. 02I read the web live

    GitHub, public LinkedIn and Crunchbase records, the open web. Not a database that went stale last quarter.

  3. 03You read the shortlist

    Ranked, with the reasoning under every name. Open a profile, ask a follow-up, narrow it down.

  • No boolean, no filters, no seat to buy. One box.
  • Read at search time, so a profile updated yesterday counts today.
  • Every step visible as it runs, every name with its reason.

500 free credits on sign-up. No card, no demo call. See real searches.

Read next