The Pitch Deck Slides That Investors Actually Read

An investor will spend somewhere between two and four minutes on your deck, and roughly two of every five will never reach the last slide. Those are measured figures, not folklore: DocSend has been instrumenting how investors actually read pitch decks since 2015, initially with a Harvard Business School researcher. The data says attention is unevenly distributed and predictably so, which means you can design for it.

The research, and what it actually measured

The original study came out of a collaboration between DocSend and Harvard Business School professor Tom Eisenmann, covering more than 200 pitch decks from companies that raised $360m in total. TechCrunch’s write-up in June 2015 reported the headline number: investors spent an average of 3 minutes and 44 seconds on a deck. The study also found founders needed an average of 40 investor meetings and just over 12 weeks to close a seed round, that Series A rounds averaged 9.6 weeks and contact with about 26 investors, and that there was no correlation between the number of investors contacted and the amount raised.

Two structural findings from that study still hold up as guidance. The most closely studied pages were financials, team and competition. And the recommended ordering for a ten-slide seed deck came out close to Sequoia Capital’s template, with product and team moved earlier than Sequoia had them.

DocSend has published section-level attention data since. On its own analysis of seed decks, it reports the same 3 minutes 44 seconds average and, separately, average time per section:

A caveat the numbers deserve: these are per-section averages across viewing sessions in which the section appeared, and they do not sum to the 3:44 whole-deck figure. Read them as a ranking of where attention concentrates, not as a budget you can allocate. The ranking is the useful part, and it is stable across DocSend’s later reports: business model, product and traction consistently absorb roughly twice the attention of company purpose, market size and “why now”.

What separates funded decks from unfunded ones

DocSend’s report covering 2020 and the first half of 2021, announced in September 2021, is the most useful comparison because it splits successful from unsuccessful raises. Average time per deck was 3 minutes 18 seconds. Successful decks got 3:20 against 2:40 for unsuccessful ones, a 24% gap. Within that:

• Business model in successful decks received 94% more scrutiny than in unsuccessful ones, averaging 31 seconds.
• Financials in successful decks received 236% more time.
• Traction in unsuccessful decks received 78% more time than in successful ones.

That third finding is the counterintuitive one and the most instructive. Extra time on your traction slide is not a compliment. It means the investor is trying to work out whether the growth is real, which is what happens when the chart is ambiguous, the axis is unlabelled or the metric is cumulative. Attention on business model and financials correlates with funding; attention on traction correlates with doubt.

By 2022, attention had compressed. TechCrunch reported on a DocSend analysis of 320 decks showing investors spending 24% less time on decks than in 2021, just under three minutes for decks that led to meetings, and abandoning failed decks after 2 minutes and 13 seconds. Team was the only section present in 100% of both successful and unsuccessful decks.

Slide-by-slide, in attention order

1. Company purpose (26 seconds). One sentence, no jargon, no mission statement. This slide’s only job is to stop the reader from having to guess what you do while reading the next four.
2. Problem (34 seconds). State who has the problem, how they solve it today and what that costs them. A named customer with a number beats an abstraction.
3. Solution (34 seconds). What the product does, in the buyer’s language. Not architecture.
4. Product (59 seconds). The second-highest attention section and the one most often wasted on a UI collage. Show the workflow: what the user does, what the system does, what changes as a result. DocSend’s research page notes that 90% of companies in its dataset had a product at least in alpha, so a slide that reads as vapourware stands out badly.
5. Market size (29 seconds). Low attention, high downside. Build it bottom-up from the number of buyers times a realistic annual price, and show the arithmetic. A trillion-dollar TAM sourced from an analyst PDF invites the one objection you cannot answer in a deck.
6. Traction (40 seconds). Non-cumulative, labelled, monthly, with the metric definition on the slide. If revenue is small, say the number anyway; a redacted axis reads as concealment and buys you 78% more scrutiny.
7. Business model (64 seconds). The highest-attention section. Price, unit, who signs, how the account grows, gross margin. This is where the 94% scrutiny gap lives, so put your actual price on the slide.
8. Competition (34 seconds). One of the three most-studied pages in the original HBS study. Name real competitors including the incumbent workflow, and be specific about the axis on which you win.
9. Team (38 seconds). Present in every deck DocSend examined, funded or not, so presence is table stakes and content is the differentiator. Relevance to this specific problem, not logo density.
10. Financials and the ask (37 and 32 seconds). Amount, runway it buys, the two or three milestones it funds, and a three-year model with the assumptions visible. Successful decks in the 2021 report drew more than three times the attention here, which only helps if there is something to read.

Common failure modes, ranked by how much they cost

• Length past the point of abandonment. DocSend’s research page describes the typical deck as roughly 20 pages with about 50 words per slide, and reports that only 58% of decks are viewed to completion. If your ask sits on slide 22, roughly two in five readers never see it. Put nothing load-bearing after slide 15.
• Burying the business model. The section that earns the most attention is frequently a single line reading “SaaS subscription”. That is a wasted 64 seconds of the best attention you will ever get.
• Cumulative traction charts. Any up-and-to-the-right cumulative chart is read by experienced investors as an attempt to hide a flat monthly rate.
• Top-down market sizing. Cheap to write, expensive in credibility, and it occupies a slide that gets 29 seconds of attention anyway.
• Optimising the wrong end of the funnel. The 2015 study found no correlation between the number of investors contacted and the amount raised. Sending the same weak deck to 120 firms does not fix the deck.

One caveat worth flagging

DocSend’s own data shows attention is not distributed evenly across founders. Its research summary reports that at seed stage, all-female teams saw 48% more investor time on the business model section, 77% more on traction and 200% more on market size than other teams. Those are not figures a deck can design around, but they do affect how much evidence a given founder needs to have ready behind each slide.

A three-minute edit you can run today

Open your deck and time yourself reading only the headlines. If the headlines alone do not tell the whole story in under 90 seconds, rewrite them so they do, because that is the reading pattern the data describes. Then move your business model slide earlier, put a real price on it, replace any cumulative chart with a monthly one, and delete every slide after the ask. The section that earns the most attention should not be the one you wrote last.

Sources

• TechCrunch — Lessons From A Study of Perfect Pitch Decks (DocSend and Harvard Business School, 2015)
• DocSend — What VCs really want to see inside your seed deck
• PR Newswire / DocSend — Show Us the Momentum (2021 seed report)
• TechCrunch — The science of pitch decks (DocSend 2022 analysis of 320 decks)
• Dropbox / DocSend — How DocSend data helps founders build better pitch decks
• DocSend — Startup Index pitch deck metrics