Reading a pitch deck well is mostly an act of subtraction. A founder has spent weeks polishing twelve slides to make a company look inevitable. Your job is to strip the polish back and find the two or three things that are actually true, the one or two that are wishful, and the gap between them. The deck is the opening argument. You are not the audience; you're the cross-examiner.
Here's the frame that helps: a deck answers five questions, in some order. What's broken? Why these people? What have they built, and does anyone want it? What do the numbers imply? And what is the money for? Everything else - the logo wall, the gradient on the title slide - is set dressing.
What is a pitch deck actually for?
It's a marketing document. That's not a criticism; it's the point. A seed deck exists to get a founder into a second conversation, and the best ones are ruthlessly edited to do exactly that. So the deck tells you two things at once: what the company does, and how the founder thinks. A muddled deck from a founder who can't explain the problem in a sentence is itself a data point - clarity of thought shows up on the page.
Which means you read it twice. The first pass is as a normal human: does this make sense, do I get it, is the problem real? The second pass is as a sceptic, slide by slide, pen in hand.
Which slides actually matter?
A typical seed deck runs to ten or fifteen slides. They tend to follow a familiar order, and a handful do most of the work.
Problem and solution
Start here, and be hard on it. Is the problem specific and urgent, or is it a vague "the market is broken" gesture? Good problem slides make you nod before you've seen the product. If you find yourself thinking "is that really a problem?", trust that instinct - a solution chasing a problem is the most common way early companies waste two years and your cheque.
Team
At seed, this is often the slide that decides it. You're backing people more than a plan, because the plan will change. Look for founder-market fit: do these specific people have an unfair reason to win in this specific space? Domain scars beat generic pedigree. A "head of growth" hired before there's a product is a small yellow flag about priorities.
Traction
The hardest slide to fake, and the easiest to dress up. Revenue, retention, and growth rate are the honest metrics. Be wary of vanity numbers - app downloads, "users", waitlist sign-ups, LinkedIn impressions - presented without any money or repeat usage behind them. The question to keep asking is: of the people who tried this, how many came back, and how many paid?
Market size
Almost every deck shows an enormous total addressable market. Almost none earns it. A "£40bn market" computed top-down from an analyst report tells you nothing. What you want is the bottom-up version: number of realistic customers, multiplied by what they'd plausibly pay. If the founder can build the market up from the ground, they understand who they're selling to. If they can only point at a big circle, they don't.
How should you read the numbers?
Two slides usually carry the financial weight: the projections and the ask. Both reward suspicion.
Projections at seed are fiction - useful fiction, but fiction. Nobody knows what revenue looks like in year four. You read them for the founder's grip on their own model: ask what assumptions drive the hockey stick. If revenue triples every year on flat headcount and flat marketing spend, someone has dragged a cell across a spreadsheet rather than thought about it.
The ask slide is where attention sharpens. How much are they raising, at what valuation, and crucially - what does the money buy? "Eighteen months of runway to hit £1m ARR and a Series A" is a plan. "To accelerate growth" is not. A clear use of funds, broken into hires, product, and runway, is one of the most reliable signs of a founder who'll spend your capital like it's theirs.
A deck earns a meeting. It doesn't earn a cheque.
What about the cap table and the terms?
Many decks gloss over ownership, so you may have to ask. If the founders have already given away half the company before the seed round - to an accelerator, early advisers, or a previous raise on rough terms - the incentive maths gets ugly fast, and later investors will notice. A clean, founder-heavy cap table at seed is a feature, not an accident.
On terms: the deck rarely contains the term sheet, but it may hint at structure - priced equity, or an instrument like a SAFE or convertible loan note with a cap and discount. None of that is in the deck to be decided; it's in the data room to be negotiated. The deck's job is just to get you curious enough to open that conversation.
Where do SEIS and EIS fit?
UK seed decks frequently flag that a company has SEIS (the Seed Enterprise Investment Scheme) or EIS (the Enterprise Investment Scheme) advance assurance from HMRC. For a UK-taxpaying angel, these reliefs are a genuine part of the picture, so it's worth knowing what the label means.
Under SEIS, investors can claim 50% income tax relief on up to £200,000 invested per tax year, with shares held for at least three years; gains on those shares can be exempt from capital gains tax if you held them for three years or more and received the income tax relief, and there's a separate CGT reinvestment relief on half of a gain put into SEIS shares, on up to £100,000 a year. EIS offers 30% income tax relief on up to £1m a year (or £2m if at least £1m goes to knowledge-intensive companies), again on a three-year hold, with CGT deferral and loss relief available. Both schemes can be carried back to the previous tax year. Source: HMRC guidance on the Seed Enterprise Investment Scheme and the Enterprise Investment Scheme.
Eligibility rules sit on the company's side too. SEIS is for companies trading under three years, with fewer than 25 full-time-equivalent employees and gross assets below £350,000 at the share issue, raising up to £250,000 in total. EIS company limits are far larger and were revised on 6 April 2026; if a deal turns on them, check the current figures in the gov.uk venture capital schemes guidance rather than taking the deck's word for it.
The honest caveat: advance assurance is HMRC indicating it expects a company to qualify, not a cast-iron promise. The relief is only secured once the company issues your SEIS3 or EIS3 certificate and you claim it, and you generally need to be a UK taxpayer to use it. So treat "SEIS-eligible" on a deck as a prompt to verify, not a closed question.
What are the red flags?
No single red flag should end a conversation, but a cluster should slow you down. The ones we see most:
- A problem you have to be talked into. If the founder spends three slides convincing you the problem exists, it may not.
- Top-down market sizing only. A giant TAM with no bottom-up build is a tell.
- Vanity metrics. Big "user" numbers with no revenue or retention behind them.
- A crowded cap table at seed. Founders already diluted below comfortable ownership.
- No use of funds. "To grow" is not a plan for your money.
- Friction over the data room. A founder reluctant to share detail or take follow-up questions is telling you something.
The bottom line
A deck is the trailer, not the film. The best you can conclude from reading one is whether the company is worth a real conversation - a data room, reference calls, a proper look at the numbers and the legals. Reading it well just means you walk into that conversation knowing exactly which claims you need to test.
For the avoidance of doubt: The Carry is editorial journalism, not financial advice. Nothing here is a recommendation to invest in anything. If you're weighing a deal, the sensible step is to speak to an FCA-regulated adviser first.
Frequently asked questions
How long should a startup pitch deck be?
Most seed-stage decks run to roughly 10 to 15 slides. The length matters far less than whether the core questions are answered: what the problem is, why this team, how the product works, what the numbers say, and what the round is for. A tight 11-slide deck that answers those beats a 30-slide deck that buries them.
What is the most important slide in a pitch deck?
There isn't a single one, but at seed stage the team and traction slides carry the most weight, because at that stage you are largely backing people and early evidence rather than mature financials. The problem slide matters too: if the problem is not real or not urgent, the rest of the deck is decoration.
What are the biggest red flags in a pitch deck?
Common warning signs include a vague or hand-wavy problem, a top-down market size with no bottom-up logic, vanity metrics presented without revenue or retention, a cap table already crowded before the seed round, no clear use of funds, and reluctance to share a data room or answer follow-up questions. None is automatically fatal, but each is a reason to dig further.
Do pitch decks mention SEIS or EIS eligibility?
UK seed decks often note that a company has SEIS and/or EIS advance assurance from HMRC, because the tax reliefs are a meaningful part of the proposition for UK-taxpayer angels. Advance assurance is an indication that HMRC expects the company to qualify, not a guarantee. Always confirm the position independently and check the SEIS3 or EIS3 certificate after you invest.
Is reading a pitch deck enough to make a decision?
No. A deck is a marketing document designed to win a meeting, not a complete picture. Treat it as the start of diligence: a strong deck earns a conversation, access to a data room, reference calls, and a look at the cap table and legals. This is general information, not financial advice; consider speaking to an FCA-regulated adviser before investing.