A pitch deck is a sales document. It's designed to make the startup look as good as possible. That's not a criticism - it's the nature of the format. But it means your job as an investor is to read between the slides.

After reviewing hundreds of pitch decks, here are the things that matter most and get overlooked most often.

The market size slide is almost always wrong

Not intentionally wrong. But the standard "TAM/SAM/SOM" framework encourages founders to start with the biggest possible number and work down. The result is a slide that says "the global market is $400 billion" when the startup's addressable segment is closer to $40 million.

This matters because it shapes the entire financial model. If the market size is inflated by 10x, every downstream projection - revenue, growth rate, market share - inherits that inflation.

What to do instead: Ignore the TAM number entirely. Focus on the bottom-up calculation. How many potential customers exist? What's the realistic price point? What's the conversion rate from awareness to purchase? If the founder can walk you through this math, the market size is grounded. If they can't, it's a guess.

The team slide hides more than it reveals

Most team slides show headshots, titles, and one-line bios. "Jane - 10 years in fintech." That tells you almost nothing about whether Jane can build this specific company.

The questions that matter:

Has anyone on the team built and sold a product in this market before?
Do the founders have complementary skills, or are they all the same profile?
Who's missing from the team, and does the founder know it?
What's the founder's track record of hiring and retaining talent?

A strong team slide isn't about impressive logos. It's about evidence that this specific group of people can execute this specific plan.

Traction numbers without context are meaningless

"10,000 users" sounds impressive until you learn they signed up for a free beta, never came back, and the startup spent $50,000 on ads to acquire them.

When evaluating traction, context is everything:

Revenue vs. users - revenue is a stronger signal than signups
Organic vs. paid - organic traction suggests product-market fit; paid traction suggests a budget
Retention vs. acquisition - keeping users matters more than getting them
Growth rate vs. absolute numbers - a startup going from 10 to 100 customers in a month is more interesting than one sitting at 10,000 with flat growth

What to ask: "Walk me through your cohort retention data." If the founder doesn't have cohort data, that's a data point in itself.

The "why now" question is often hand-waved

Timing matters more than most investors give it credit for. A good idea at the wrong time is just a bad investment.

The "why now" slide should answer: what changed in the market, technology, or regulation that makes this startup possible today but not three years ago?

Weak answers: "AI is changing everything" or "the market is growing."

Strong answers reference a specific shift - a new regulation that created demand, a technology that became affordable, a behavior change accelerated by recent events. The best founders can point to a concrete inflection point.

Competitive positioning that claims there's no competition

If the competitive landscape slide shows the startup alone in a category, one of two things is true: either the market doesn't exist yet (risky), or the founder hasn't done the research (also risky).

Every startup competes with something. Even if there's no direct competitor, customers have existing solutions - spreadsheets, manual processes, a different tool they've repurposed. Understanding what you're replacing is as important as understanding who you're competing with.

The founders who impress us most can name their top three competitors, explain what each does well, and articulate specifically where they're different. Not better at everything - different in a way that matters to a specific customer.

Unit economics buried in the appendix (or missing entirely)

The pitch deck might show a hockey-stick revenue chart, but if you can't find the unit economics - CAC, LTV, payback period, gross margin - that chart is decoration.

Early-stage startups won't have perfect unit economics. That's fine. But they should have a thesis about what the economics will look like, and early data to support it.

Questions to ask:

What does it cost to acquire one customer today?
What's the average revenue per customer per month?
How long does it take to recoup acquisition costs?
What's the gross margin after infrastructure and support?

If the founder can't answer these questions with even rough numbers, the financial model is built on assumptions, not data.

What this means for your process

None of these issues are automatic deal-killers. Early-stage startups are, by definition, works in progress. The point isn't to find a perfect company - it's to understand the real risks before you invest.

The best pitch decks are honest about what they don't know yet. The worst ones try to hide it behind polished design and big numbers.

Your diligence process should look beyond what the deck shows you and examine what it doesn't.