A deck is evidence
A pitch deck (pitchpresentasjon) is a claims document. Every slide asserts something that is evidenced, checkable later, or decorative, and your work is to sort the three into a written list, which is a different activity from being persuaded.
The habit worth breaking first is rewarding polish. Steve Blank put it in one sentence in 2013: “Flashy demo days are great theater, but it’s not clear there’s a correlation between giving a great PowerPoint presentation and a two minute demo and building a successful business model.” Design competence is cheap and widely available, and customer evidence is neither.
The second habit is speed. DocSend, which instruments decks for a living, reports that investors spend on average under two and a half minutes on a deck. That is a fair budget for the first screen. A deck that has passed the screen deserves thirty minutes, most of it on three slides.
Run the first screen before this read. The deck can only tell you whether a plausible business is well argued, never whether an implausible one is worth the argument.
What each slide has to prove
A deck has eleven elements to prove, adapted from Guy Kawasaki’s ten-slide model. “Underlying magic” (why us, why now, how this is going to win) is his term; market size, exit strategy and fundraising information are the three additions, and Sequoia Capital’s public outline is near-identical, so the structure is no local convention.
The test is the same whatever order the founder chose. After one pass, can you reconstruct the business on a single page? That page has to carry the nine fields of an investor brief (domain, problem, concept, market, customers, competitors, business model, traction, capital sought), plus the team and the underlying-magic question.
Use the eleven elements as your checklist on the first pass, one line each. A line you cannot fill means the deck failed at that slide, whatever else it did well, and the founder gets asked about it.
| Element | The question it must close, and the evidence that closes it |
|---|---|
| Problem | Who has it, how often, and what does it cost them today? Closed by named customers or segments with a quantified current cost. |
| Solution | What does the product actually do, and for whom? Closed by a description at the level of the user’s task, not the architecture. |
| Market size | How large is the addressable slice this company can serve? Closed by a number built up from customers and price, not down from a category total. |
| Business model | How does revenue arrive, and what does a customer cost to win and keep? Closed by prices actually charged, and an acquisition cost with a source. |
| Competition | What do buyers use instead today, including doing nothing? Closed by named alternatives with an honest account of their strengths. |
| Underlying magic | Why us, why now, and why does this win? Closed by a reason grounded in timing, access or capability, not in effort. |
| Go-to-market strategy | How are customers found and won, and at what cost? Closed by a channel that has produced a sale, with the cost of that sale. |
| Team and advisors | Why these people for this problem, and who is actually involved behind them? Closed by domain history, prior delivery, a clear split of commercial and technical work, and advisors with a role. |
| Traction and milestones | What has been proved with real money or real usage, and what is next? Closed by contracts, repeat purchases or retention, and a dated next milestone. |
| Exit strategy | Who would buy this, and why? Closed by named acquirer types with a reason each would pay. |
| Fundraising information | How much, on what terms, and what does it buy in milestones? Closed by amount, instrument, use of funds tied to a next-round milestone, and a runway figure. |
Red flags slide by slide
None of these patterns is disqualifying on its own. Each one adds an item to what you must verify before money moves, so note which ones you saw.
The team and ownership slides
- Founder biographies that are impressive and irrelevant, with senior titles from adjacent industries standing in for experience of this customer.
- A co-founder who appears on the team slide but is absent from every other document, or an equity split that nobody will describe.
- A cap table with large dormant holdings from an earlier venture, a founder already below a working stake, or shares promised verbally and never issued. Each has to be resolved before a round can be priced, and each is a negotiation with people who are not in the room.
- No mention of vesting, and no account of what happens if a founder leaves. Future dilution (utvanning) is arithmetic; a departed founder holding a third of the company is not.
- Advisers and logos used to carry credibility the team slide cannot carry itself.
The market and competition slides
- A market number that appears only as a total for the category, with no path from it to a customer count and a price.
- An empty or dismissive competition slide. Guy Kawasaki’s list of entrepreneur claims includes “No one is doing what we’re doing”, and his observation on it is the useful one: “if you have a good idea, five companies are doing the same thing. If you have a great idea, fifteen companies are doing the same thing.”
- Competitors listed but not the real alternative, which at early stage is usually a spreadsheet, an incumbent’s existing module, or the customer continuing to do nothing.
- Defensibility resting on a patent or a filing with no account of what it covers or who would have to be sued.
The traction and ask slides
- Pipeline presented as traction: signed letters of intent, pilots without payment, and “in discussions with” logos counted as customers.
- A single customer producing most of the revenue, with no second reference account. Concentration is a legitimate stage of a company’s life, but it must be visible and named.
- Customer validation from a free tier, survey respondents, or the founder’s own network, none of whom will ever be invoiced.
- A valuation (verdsettelse) set by what the founder needs rather than what the stage supports, or one that requires the next round to be priced higher than the company’s own plan can justify.
- An ask that does not reach a milestone, for example a raise covering nine months when the first proof point lands in month fourteen.
- A deck that admits no weakness anywhere. A realistic assessment of obstacles is a criterion worth scoring in its own right, and its absence tells you the founder either has not looked or will not say.
The numbers pages deserve double time
The market, traction and financials slides carry checkable arithmetic. Most of your thirty minutes goes here, because these are the claims you can test without the founder.
For market sizing, build the number back up from plausible customers, a realistic price and an achievable share within the plan’s horizon. If your bottom-up figure and the slide differ by an order of magnitude, your figure is the company’s actual opportunity and the slide is decoration.
For traction, separate money that arrived from money that was discussed, and ask what happened at the first renewal decision. Repeat purchase is the strongest evidence in an early-stage deck, and the easiest to verify because the customer will confirm it.
For financials, ask for the model itself. A budget of hardcoded numbers with no working model behind it means the founder cannot answer “what happens if the sales cycle is twice as long” without rewriting the file.
Any budget with a hockey-stick profile (flat, flat, vertical) deserves one question. What changes in the month the line turns? If the founder answers “the funding”, the projection is a wish.
Keep Kawasaki’s list at hand for these pages. “Our projections are conservative”, “We have a proven management team”, and “Patents make our product defensible” are the three claims made most and supported least.
What is missing matters as much as what is shown
Check for these four absences every time. Silence in a deck is usually deliberate, and each absence points at a specific question for the meeting.
- No competition slide. Either the founder has not looked at the market, or the honest version of that slide is uncomfortable.
- No churn, retention or repeat-purchase figure in a deck that claims paying customers. Growth without retention is a leaking bucket presented from above.
- No founder history (what they built before, what happened to it, and why they left). Founder history is the single best predictor available at this stage.
- No use of funds. An ask without a plan for the money means the milestone has not been chosen, which means the next round has not been thought about.
After the read, write down which of the eleven elements you could not fill in, and take that list into the meeting instead of your impression of the deck.
From deck to meeting
The deck raises questions and settles none. Turning the read into a meeting is four steps.
- Write down the claims that matter most to the case, the evidence that would confirm each, and who would have to provide it.
- Order the list by consequence. A claim that changes whether you invest goes first; a claim that changes the price goes second; everything else is context you can gather later.
- Send the founder the two or three questions that need preparation, and the request for the financial model. Keep the rest for the conversation, where the quality of an unrehearsed answer is itself evidence.
- In the meeting, ask which slide the founder is least confident about. The answer separates founders who have tested their own case from founders who have only presented it, and it leads straight to the part of the business that will decide the outcome.
The list is the same for everyone. Where the time goes depends on where you stand.
- A first-time angel works through all eleven elements in writing, skipping none, and spends the thirty minutes on the three numbers slides.
- An experienced angel goes straight to the three numbers slides and the four absences, and writes the list in the meeting’s order from the start.
- An angel with a sector focus spends the time on competition and underlying magic, where the founder cannot get past you with vocabulary, and checks the technical and market claims personally.
- An angel in a syndicate keeps one list per deck and gives each claim to the member best placed to verify it, so the founder is asked once and the answers sit in one place.