How to Make a Startup Pitch Deck

A pitch deck is an argument. Each slide advances one step of that argument, and the audience should be able to state your case back to you after the last slide. Most student decks fail not because the idea is weak but because the argument is missing — the slides describe a project rather than making a case.

The good news is that the structure is standard, and the discipline is learnable in an afternoon. The hard part is having something true to put in it, which is why a deck should be built after the work, not instead of it.

The argument, in order

A pitch, whether five minutes or fifteen, follows the same logical chain:

1. There is a specific problem, affecting specific people.
2. It matters — here is what it costs them today.
3. Here is our solution, and here is what it produces.
4. Here is evidence that it works and that people want it.
5. Here is the market: enough people share this problem.
6. Here is how we make money.
7. Here is why we are the ones to do it.
8. Here is what we have done so far and what we do next.
9. Here is what we are asking for.

Nine steps, roughly one slide each. Departures from this order should be deliberate, not accidental.

Slide by slide

Title. Name, one-line description of what you do in plain language, your names, contact. The one-liner should be understandable by someone outside your field. If it contains more than one buzzword, rewrite it.

Problem. One problem, stated concretely, with a named group of people. Use a specific scenario rather than a general statement. "Fest sponsorship teams of ten to fifteen students lose track of commitments across personal WhatsApp threads" beats "communication in student organisations is inefficient".

Cost of the problem. What it costs today — hours, rupees, lost opportunities. Use numbers you have actually collected. If your only evidence is fifteen interviews, say "in fifteen interviews with fest leads, eleven described losing at least one confirmed sponsor to lost follow-up." That is more persuasive than a borrowed industry statistic, and you can defend it.

Solution. What you built and what outcome it produces. Show the product, not a description of the product: one screenshot of the core flow, or thirty seconds of demo. Avoid architecture diagrams unless the audience is technical and the architecture is the point.

Evidence. The most important slide and the one most often missing. Usage, retention, paying customers, pilots, letters of intent, waiting lists people actually joined. Whatever your strongest real signal is, this is where it goes. If your only evidence is qualitative, present it precisely and label it as such.

Market. How many people share this problem, and how you reached that number. Build it upwards from countable units — number of institutions, number of teams per institution, price per team — rather than quoting a large top-down figure. A defensible small number beats an indefensible large one, and judges probe exactly here.

Business model. Who pays, how much, how often, and what it costs you to serve them. One sentence plus one simple table.

Team. Why you, specifically. Relevant experience, unusual access, domain closeness, what each person actually does. For students, honest specificity ("I ran the fest sponsorship desk for two years") is far stronger than inflated titles.

Traction and roadmap. What you have done, in dates, and the next three or four milestones with what each will prove. Milestones should be learning outcomes, not just feature releases.

The ask. What you want from this room: funding, a pilot partner, mentorship, access to a campus, a specific introduction. Be precise. A vague ask produces polite applause and nothing else.

Design rules that hold up under a projector

• One idea per slide. If a slide needs two headlines, it is two slides.
• Headline as claim. Write the slide's headline as the point it makes, not as a category label. "Eleven of fifteen teams lost a confirmed sponsor" instead of "Problem".
• Readable from the back. Large type, high contrast, few words. Assume a bad projector and a bright room.
• No paragraphs. If it is written out, you will read it aloud, and the room will read ahead of you.
• Real screenshots over mockups, once real screens exist.
• Charts with labelled axes and honest scales.
• A consistent visual language — one font pair, one accent colour, one layout grid.

Ten to twelve slides is right for most student pitches. An appendix after the final slide is useful: detailed numbers, methodology, extra screens, ready to open when a question demands them.

Evidence discipline

Judges, mentors and partners are not persuaded by adjectives. They are persuaded by claims that are specific, sourced and survivable under a follow-up question.

Three rules:

1. Never invent a number. Fabricated traction is discovered in the Q&A, and it ends the conversation permanently.
2. Label the source of every figure. Your own research, a published source, or an estimate — say which.
3. State your uncertainty first. "This is an estimate based on twenty conversations; the sample is one city" earns credibility. Being caught overstating destroys it.

If you are early and your evidence is thin, present it as early evidence with a clear plan to strengthen it. Nobody expects a first-year team to have revenue. Everyone notices a team that cannot tell the difference between what they know and what they hope.

Delivery

The deck supports you; it does not replace you. Practical preparation:

• Rehearse aloud, with a timer. Silent reading hides the parts that do not work.
• Know your first and last thirty seconds by heart. Openings and closings are where nerves cost the most.
• Rehearse the demo failing. Have a recorded video and screenshots ready.
• Assign speaking roles clearly if you are a team, and hand over explicitly.
• Leave time for questions — the questions are usually where the decision is made.

Handling questions

Answer the question that was asked, in the first sentence. Then support it. Long preambles read as evasion.

When you do not know, say so and say what you would do to find out. "We haven't tested that yet — it's the next thing we're running, and here's how" is a strong answer. Bluffing is not.

When a question is hostile, treat it as information about a weakness you may not have examined. Note it, address it briefly, and revisit it after the session.

Adapting the same deck

The argument stays constant; the emphasis shifts with the audience.

• Competition judges weigh clarity, originality and evidence, usually against a published rubric. Read the rubric and make sure each criterion is visibly addressed.
• Mentors care about your thinking process; show what you learned and how you changed your mind.
• Institutional partners care about risk, effort required from them, and whether you will still exist next semester.
• Investors care about the size of the opportunity, the model, and the team's ability to execute.

Build one solid core deck and prepare variants of the ask slide and the emphasis, rather than maintaining several decks that drift apart.

The honest test

Before you present, ask someone unfamiliar with your work to read the deck without narration and then explain your business back to you. If they can state the problem, the solution, the evidence and the ask, the deck works. If they cannot, no amount of delivery practice will rescue it — the argument needs fixing first.