TL;DR: When you send a venture studio your pitch, it doesn't get read the way you think. We're not scoring polish, ambition, or slide design. We run every inbound deck through an eight-part rubric; problem quality, market size, timing, founderāmarket fit, competitive advantage, validation signals, business model fit, and studio fit; and every category is really testing the same underlying thing: is this specific and true, or generic and decorative?
This article opens the rubric completely, category by category, with the signals that move a score up or down and the concrete tells we look for. The pattern underneath all eight: the highest-scoring decks aren't the most impressive ones, they're the most internally consistent. The fastest way to lose us is to tell three different versions of your own story across the deck, the written proposal, and your live product. The second fastest is to claim a moat you don't have while missing the competitors you do.
The gap between how you think you're read and how you're actually read
Most founders build a deck to impress. They optimise for a clean narrative, a big market number, a confident tone, and a design that looks funded. That instinct is understandable, and it's aimed at the wrong target.
When a deck lands in our inbox, we're not asking is this impressive? We're asking is this real, is it specific, and is there a gap here that we are the right people to fill? Those are different questions, and a deck optimised for the first one often scores badly on the second.
We publish our rubric here for a simple reason; the founders who send us decks deserve to know what happens to them on the other side of the table. It also turns out that a founder who understands the scorecard writes a fundamentally better deck, not a glossier one, a truer one.
We score across eight categories, each with a couple of sub-criteria. Here's the whole thing, and what each one is actually testing.
1. Problem Quality
Sub-criteria: clear pain point Ā· urgency and frequency
This is the heaviest category in the rubric, and it's the one founders most often get wrong by aiming too high. A problem stated at the altitude of a mission statement; "healthcare is broken," "small businesses are underserved", is true and completely inert. True is not the same as acute.
We're looking for pain that is felt, recurring, and already costing someone money.
Scores up when: the pain shows up in behaviour and budgets. People are already spending time or cash to solve it badly. The problem recurs on a known cadence. There's emotional intensity in how sufferers describe it.
Scores down when: the problem is real but abstract, no one is described as actively suffering, or the frequency is vague. A problem nobody has recently paid to escape is a weak problem, however noble.
2. Market Size
Sub-criteria: TAM / SAM / SOM clarity - growth trends
This category does not reward the biggest number, It rewards the founder who understands the structure of their market well enough to narrow it honestly, from the total addressable market down to the slice they could plausibly capture in three years.
Scores up when: the SOM is built bottom-up and small enough to be credible. The founder can explain who the first thousand customers are and why. The growth trend is specific rather than borrowed.
Scores down when: the deck leans on a giant TAM and a hand-waved "if we capture just 1% of this $X billion market." Nobody captures 1% of a huge market by wishing. The 1% line is a tell that the numbers are decorative, not load-bearing.
And there's a specific failure that costs points fast, market numbers that don't agree with each other across the deck. When slide 4 says the market is worth one figure, slide 8 says another, and the summary says a third, it may suggest that the founder pulled impressive-sounding numbers from three sources without reconciling them. Inconsistent market sizing inside a single document is not a rounding issue, it's a signal about how the whole deck was assembled.
3. Timing / Trends
Sub-criteria: market readiness - alignment with current shifts (tech, behaviour, regulation)
Every deck now claims the market is moving in its favour. This category separates a real why now from just name-dropping a trend.
A real why now points to a specific thing that changed, a model capability that recently crossed a usefulness threshold, a behaviour that became normal, a regulation that opened or closed a door, and explains why the company is possible or necessary now and wasn't two years ago.
Scores up when: the timing argument is specific and would have been false eighteen months ago.
Scores down when: the argument is "AI is hot" or "this category is growing." That's ambient, not causal. It's true for ten thousand companies and therefore an edge for none.
This is often one of the higher scores we give, because founders are good at picking timely ideas. But we're careful not to let a strong timing score inflate the rest. Timely and differentiated are not the same word.
4. FounderāMarket Fit
Sub-criteria: deep founder insight - personal story
This category is not about credentials. It's about lived, hard-to-fake insight into the specific market, the kind you can't get from a report.
Scores up when: the founder has been the user or the supplier. They know where the market breaks in ways you can't Google. Their personal story and their business thesis are the same story.
Scores down when: the founder discovered the problem secondhand, or is trying to borrow the core capability they lack; most commonly by leaning the entire technical foundation on an institution or partner rather than owning it.
There's a subtlety here that cuts both ways, and it's worth stating because it changes how you should write your own deck.
Founders routinely bury their strongest fit under a generic ambition narrative, for example someone who spent three years inside a market as an operator and went through it earlier as a participant has genuinely rare, two-sided insight; but if the deck positions them as a global-vision founder from slide one, that specific fit never surfaces.
We frequently score a founder higher after a conversation than the deck did, because the deck was pitching past the founder's own best asset.
5. Competitive Advantage
Sub-criteria: unique insight, unfair advantage, IP
Two things get tested here at once here; is there a real moat, and does the founder actually know who they're competing with?
The second half fails more often than the first, and it's the single most damaging tell in a deck. A competitor slide that lists the famous incumbents, the household-name apps everyone knows, while missing the three well-funded startups doing precisely what you're pitching, tells anyone with domain knowledge that you haven't done your homework.
It's the worst possible signal to send, because it's the one thing a knowledgeable reader catches instantly.
Scores up when: the competitor map includes the real, direct threats, including the boring B2B and infrastructure players, not just the consumer brands, and the moat lives in a hard-to-copy layer: proprietary data, a network effect, accumulated domain intelligence, a distribution lock.
Scores down when: the moat rests on the technology itself (a model, a rendering pipeline, an integration anyone can buy), on a patent that's years away, or on a terminology distinction the user can't perceive. If your entire edge is that you call your thing by a more impressive name than the competitor's identical thing, that's not a moat.
Founders who claim the tech is the moat score lower than founders who can name the layer that actually compounds.
What this looks like in practice A competitor table with a green check under your product and red X's under everyone else's, where "everyone else" is four brands that don't actually do what you do, and the two startups that do exactly what you do aren't on the table at all. The table was built to make you win. It only proves you don't know the field.
6. Validation Signals
Sub-criteria: existing traction Ā· lo-fi experiments Ā· customer interest
This is where we're strictest, and it connects directly to a distinction we've written in another article, the fact that evaluation is not validation and almost every founder fails here.
They fail because they have been fed up with a business culture that extracted from Human Computer Interaction, the discipline that explains how to build products, only what was convenient and then filtered it to the bone, confusing also administration with building.
Interviews, surveys, mentor enthusiasm, accelerator praise, investor "interest", these reduce uncertainty but confirm nothing. Payment and retention confirm. (If this line of thinking is new to you, our piece on validating before you build goes deep on it.)
Scores up when: there's behavioural evidence, revenue, retention, paid pilots, real switching and the founder can decompose it. Not just "we have traction," but what the traction is made of.
Scores down when: the "validation" is a wall of compliments, or when the traction is for a different product than the one being pitched.
Two traps live in this category specifically.
The first is the naked revenue number. A five-figure or six-figure revenue line tells us almost nothing on its own.
- Is it recurring subscriptions or one-off packages?
- Growing month over month or front-loaded and decaying?
- Repeat customers or a handful of personal relationships?
- Seasonal?
The right strategy, and the right valuation, is completely different depending on the answer. A founder who leads with the headline number but can't break it down has shown us a number, not a signal.
The second is adjacent traction. A live business that serves a related market is a genuine asset, but it validates that business, not the new one.
7. Business Model Fit
Sub-criteria: monetisation potential - scalability
This category tests whether there's a clear, scalable path to revenue, and, more revealingly, whether the founder knows who actually pays.
Scores up when: there's an identified buyer, plausible margins, and a monetisation logic that matches how the market actually buys. Focus reads as confidence.
Scores down when: the deck lists every possible revenue stream at once, B2C and B2B and institutional licensing and enterprise, all at MVP stage. Listing all of them signals the founder hasn't chosen, and hasn't yet learned that each of those is a different company with a different buyer, a different sales cycle, and a different product.
The clarifying question we always end up asking:
Is the buyer the individual user, the professional using the tool, the institution, or the government body?
Those are four different businesses. A founder who has picked one for the first eighteen months scores higher than one who's keeping all four open.
8. Studio Fit
Sub-criteria: aligned with studio thesis - we have an edge (talent, tech, market access)
The final category is the one about us, and it's why a genuinely good company can still be a no from a studio specifically.
A venture studio's entire model is injecting scarce labour; product management, design, engineering, go-to-market, in exchange for equity, because that labour is what creates the value. So this category asks if there is a gap here shaped like the thing we provide.
Scores up when: the gap in the company is exactly what we inject, the vertical fits our thesis, and we have a real edge, talent, technology, or market access the founder can't easily get elsewhere.
Scores down when: one of two things is true. Either the team is already complete, product, engineering, and design all covered, in which case there's nothing studio-shaped to add and we'd be paying equity to duplicate a team that exists. Or the ask is money-only, in which case the founder wants a VC or a marketing agency, not a co-builder.
There's a part of this that founders rarely see, and it's worth making explicit: the shape of the deal you're offered depends partly on us, not only just on you.
The baseline model is straightforward. We provide the workforce, product, design, engineering, go-to-market, and in exchange we take equity, commonly in the range of twenty to thirty percent when we're building alongside a founder from an early point.
But it isn't the only model, and studios differ here. We've developed others. Sometimes we take a smaller equity stake and pair it with a fee, either paid upfront, or deferred and phased, a monthly amount or a lump settled after, say, six months once the work has proven its value. Even with a fee attached, this is usually still cheaper for a founder than hiring a software house or an agency outright, and the equity we keep holds our incentives to yours.
Which of these you're offered is not a fixed menu. It depends on the studio's own state at the moment you apply, how many products we're already building, what capacity we have free, and where our cash position sits.
A studio deep inside a full portfolio structures a deal differently than one with room for a new build. Two near-identical founders arriving six months apart can be offered genuinely different terms, and the reason lives on our side of the table, not theirs. It's worth understanding when you approach any studio, that you're not only being scored against a rubric, you're arriving into an economy that has its own weather.
This is also where an ambiguous ask does real damage. When a proposal says the founder wants investment, or hands-on help, or ideally both, that's not flexible, it's unpriceable. We cannot value the equity until we know what labour we're providing. "Yes to everything" in the ask section is not accommodating; it's the one answer that stops us from being able to make an offer at all.
The signals that cut across every category
The eight scores matter, but three meta-signals sit above them and often decide the verdict regardless of the totals.
Internal consistency is the most revealing thing in the entire packet.
We don't read the deck in isolation. We read it against the written proposal and against the live product, and we watch for whether the three agree. When a deck implies the product is built and in testing, the proposal says development is just beginning, and the live website shows no trace of the product at all, those cannot all be true, and no individual category score matters as much as the fact that the story doesn't hold together. Consistency is cheap to produce and expensive to fake.
Coachability is what we're testing when we push back.
We'll often float a reframe, narrow your market, defer the moonshot feature, sequence the roadmap differently, use off-the-shelf infrastructure now instead of waiting on a proprietary build. We do this not because we're certain we're right, but because how a founder responds tells us more than the deck ever could.
A founder who engages, pressure-tests it, and either adopts or credibly rebuts it is someone we can build with. A founder welded to their original narrative is a much harder bet, no matter how good the narrative is.
What a deck buries versus what it leads with is a signal in itself.
The grand global vision goes on slide one; the single most defensible insight is often three layers down, mentioned in passing. The buried thing is frequently the real company. Part of our job in evaluation is to find it, and part of your job in writing the deck is to not make us dig.
What a score actually means
The rubric is not a gate that outputs yes or no. It's a map of where to probe.
The most common outcome, and the most useful one, is a middling total that says investigate further. That result isn't a rejection. It's the rubric telling us where the tension is: strong problem, weak validation; great timing, unclear moat; excellent founder fit, undefined ask. The number points us at the conversation worth having.
A single low category is rarely fatal on its own. A pattern of contradictions across categories usually is, not because any one of them is disqualifying, but because together they describe a story that hasn't been made to cohere yet.
What we're really evaluating
None of this is really about the deck. The deck is a proxy.
Underneath all eight categories, we're evaluating three things.
- Is there a real, specific, hard-to-fake insight at the centre of this, something the founder knows or has proven that most people don't?
- Can this founder execute, or be helped to execute?
- And are we, specifically, the right builder to close the gap?
So the takeaway for anyone about to send us a deck is the opposite of the instinct most founders start with.
Don't optimise for polish, optimise for consistency, specificity, and honesty about what's real. A deck that plainly says here is what exists, here is what doesn't yet, and here is the one insight I am certain of will score higher than a beautiful one that claims everything and holds together nowhere.
We can build around an honest gap. We can't build around a story that contradicts itself.
