Every work stream inside the company gets a room. The meetings you already hold score it — progress against what you committed, what customers actually did, what's blocked and what it costs against runway. The investor update falls out of that book instead of being written from memory the night before.
It scores the work, never the people. What a stream committed to and what landed. Which blocker has surfaced three times with nobody clearing it. Facts a team can act on.
A board meeting is a single conversation that has to serve two portfolios at once. The founder is running a company; the investor is running a fund. Same session, same evidence — scored into two different books, at two different scopes, with a boundary down the middle that neither side has to police by hand.
One company, broken into the work streams that run it — scored on the same evidence.
One session. Digested once, and only what was approved crosses the line.
The same company as one row of a portfolio, grouped by funding stage — and the fund's own rooms alongside it.
“The meeting went great” carries no information, because it is what every founder says about every meeting. The same claim as evidence — who was in the room, what the customer actually said, what got committed — is something an investor can act on: ground reality instead of a progress report, and a reason to make the introduction that follows a real result. The handshake is worth opting into because value moves back across it.
It scores the company, never the founder. Runway stated or not stated. Board actions closed or carried. Roles the plan depends on filled or open for two quarters. Those are facts about a business, and a partner can act on every one of them. “How the founder came across” is not on the board, is not a dimension, and is not something this system is built to produce.
Two room types, not one room with two logins. An investor's row is a company; a founder's row is a work stream inside one of five domains. What the founder's rooms produce feeds the six dimensions a portfolio company is scored on — a mapping, not a mirror, which is why both sides are worth modelling.
See the investor's sideWhat a work stream is judged on at Angel is not what it is judged on at Series B+. Customer proof is nearly everything before there is a product; once there is one, whether decisions leave the room with an owner attached matters more. Same six dimensions, same engine — reweighted by where the company actually is, and asking only the questions that stage has earned.
Leads with Customer Proof at 30%.
Every stage sums to 100. Nothing is switched off — a dimension that matters less here still counts, it just costs less when it's missing. Weights that drop to zero would hide a gap instead of pricing it.
The agenda is the coverage specification, so this list is exactly what the gap scanner is allowed to raise against a Angel company.
These aren't gaps here. A flat agenda would print “Never covered · Audited financials” against a company three months old and call it a finding — and false findings are how an honest coverage figure stops being believed.
The six dimensions, weighted for the stage. How the company is actually doing — and the only reading most tools attempt.
Coverage and freshness — how much of that reading is backed by evidence, and how old the evidence is. Kept separate on purpose: a confident score on thin, stale data is the most dangerous row on the board, and averaging the two together is exactly what hides it.
The mechanism is a weighting table per axis value, not anything venture-specific. The same layer takes industry-specific weightings — a fintech and an agritech at the same stage are not de-risked by the same six numbers.
Every business judges health differently. meetapp.ai is configured to the dimensions and weights you already argue about in your own reviews — then holds every work stream to them consistently, across the whole book.
These are not invented dashboard metrics. Clarity of ownership, participation, alignment and follow-through are long-established constructs in organisational research. What has always been missing is a way to measure them without asking people to fill in a survey about themselves — self-reported, months late, and answered the way people wish things were. Reading the conversation measures the behaviour instead.
What was committed last cycle against what actually landed this one.
What customers did in response — usage, retention, renewal — not what they told you.
What is in the way, whether it is named, and whether anybody is clearing it.
Whether decisions get taken, leave with an owner attached, and hold afterwards.
What this stream costs, against what the company has left to spend.
Whether the roles this plan depends on actually exist and are filled.
Weights sum to 100. The overall score is the weighted sum of what's visible — no hidden adjustments, no black box. Change what you measure and the rooms start listening for it. This configuration reweights itself by domain, so each dimension shows the span it moves across — see the curve.
Your board isn't a portfolio unless you call it one. Here it's the domain — the thing Monday morning is actually organised around — and every work stream sits in one column of it: same rooms, same scores, triaged for whoever's accountable.
Filter the book, sort it worst-first, open a work stream, click a dimension through to the quote behind it, generate this week's digest. Fictitious data, real engine.
The founder's side of the same meetings. Every work stream inside the company gets a room, laid out by the function it serves, and scored on what the team actually said about it — so the board update is a view of the operating book rather than a document somebody writes the night before.
Every dashboard tells you what it knows. This one works on what it doesn't. The gap scanner runs continuously across the whole book, and everything it finds has an owner and a route home.
The agenda called for it; the meeting didn't reach it.
Asked for, acknowledged, not delivered.
Last signal is older than this dimension tolerates.
A commitment with no name or no date against it.
As an item to Ask, Confirm or Bring — leading the meeting.
As a request with a reply scaffold; the reply closes it.
A chase with the original ask quoted back.
A request that goes unanswered for three weeks doesn't leave the score unchanged — it lowers confidence in it. Data quality erodes the longer you go without the next update, and that erosion belongs on the screen, not in someone's head.
The board pack is written the night before, from memory, about a quarter you already lived
Delivery gets reported in effort, so a stream that has quietly stopped moving still reads busy
Chasing every function for a status that is stale by the time it is collated
What each work stream actually costs against runway is reconstructed once a quarter