Section 1
Do: put the message in the headline
Give every slide a headline that states the claim rather than naming the topic. Market opportunity is a topic. Compliance teams are buying this from three vendors and none of them serve firms under two hundred staff is a claim, and a reader who sees only your headlines should still receive the argument in order. That rule does more work than any design choice. It forces you to know what each slide asserts, which reveals the slides that assert nothing, and those are usually the ones you kept because removing them felt like losing ground. It also fixes sequencing, because when the headlines are claims you can read them as a paragraph and hear where the logic jumps. The evidence then sits underneath as support: the chart, the quote, the number with its definition. One claim per slide, with what backs it visible without narration. The content of that argument is treated separately in [Telling Your Story to Investors: What Works, What Doesn't](/blog/telling-your-story-to-investors-what-works-what-doesn-t).
Section 2
Do not: build the narrative arc into the deck
Founders taught to use dramatic structure sometimes withhold the point to build tension. Withholding works when the audience is captive and has agreed to be entertained. An investor reading an attachment is neither, and will make a decision before your reveal arrives. Lead with the conclusion. State what you are building, for whom, what evidence exists, and what you are asking for, inside the first three slides. What follows is the support, structured so a sceptical reader can check the part they doubt without hunting. Tension in a pitch comes from the market situation being genuinely difficult, not from the order in which you disclose facts. Where narrative shape does help is in a single customer example, and [Using the Hero's Journey Framework in Business](/blog/using-the-hero-s-journey-framework-in-business) is a fair guide to that narrower use.
Section 3
The slide-by-slide check
Take the checklist that follows and apply it to your current deck, slide by slide. Each row asks what the slide claims, what evidence appears on it, and whether it survives without you speaking. Slides failing all three are appendix material at best.
Section 4
The appendix is the credibility instrument
Keep the main deck short and put everything a diligent reader would want into an appendix: cohort tables, unit economics with the assumptions stated, the competitive comparison including where you lose, the regulatory position, the churn definition. Then reference it explicitly, because an appendix nobody knows about is a file nobody opens. The signal this sends is worth more than the content. A founder who has prepared the detail for questions they hope are not asked is a founder who has done the work. It also protects the main sequence, since the temptation to cram everything into twelve slides comes from a fear of looking unprepared, and an appendix resolves that fear without wrecking the argument. One warning: the appendix must be accurate. Anything found wrong there costs more than the same error would have in the main deck, because the appendix is where you claimed to be rigorous.
Section 5
Language discipline
Define your terms on first use, in one clause, and then use the same term consistently for the rest of the document. Decks routinely use platform, engine, automation and intelligence to mean different things on adjacent slides, and a reader who cannot pin down what a word means in your company will assume you are inflating something. This matters most where categories are genuinely confused. If your product schedules deterministic tasks, say so, and do not describe it as learning. Investors who understand the distinction will mark you down for blurring it, and those who do not will discover it in diligence, which is worse. [Robotic Process Automation (RPA) vs AI Automation: What's the Difference?](/blog/robotic-process-automation-rpa-vs-ai-automation-whats-the-difference) is a usable reference for keeping those words honest.