Every pitch deck in 2025 claims the same thing: “our CAC is efficient,” “we have high LTV,” “our payback period is sub-12 months.”
On paper, it looks great. In practice, it rarely holds up.
Startups are still fundamentally misrepresenting (or misunderstanding) their unit economics. The buzzwords survived the market correction. The rigor didn’t.
Why This Keeps Happening
Let’s call out the three most common reasons unit economics still don’t add up:
- Misaligned Definitions
CAC might include or exclude variable marketing spend. LTV might be projected over wildly different time horizons. Gross margin might be GAAP… or it might be “adjusted” within an inch of its life. There is no standard, and most founders exploit that. - Forward-Looking Fantasy
Most early-stage companies define LTV using projected churn curves and aspirational ARPU -as if retention magically improves over time. CAC is often based on last month’s blended cost, not actual acquisition channel economics. It’s forecasting disguised as math. - Investor Incentives Are Misaligned
VCs ask for unit economics early, so founders produce them early -even if they’re built on assumptions that wouldn’t pass a Series B diligence process. There’s pressure to show maturity, but not accountability.
So we end up with slides like this:
CAC: $115 | LTV: $890 | Payback: 7.3 months
(Footnotes: excludes cost of onboarding, support, infrastructure, and anyone with a pulse who might churn in 60 days.)
This is not analysis. It’s spreadsheet cosplay.
The Real Cost of Bad Economics
At the seed and Series A level, this misinformation creates bad feedback loops:
- Product teams optimize for features that juice short-term retention but cannibalize long-term margin.
- Marketing over-invests in channels with decent CAC but terrible second-order costs (e.g. discount-driven SEM, affiliate arbitrage).
- Investors underwrite rounds assuming efficiency that doesn’t exist, leading to flat rounds and broken cap tables later.
Even worse, when these companies do scale, their faulty unit models scale with them -often all the way to IPO S-1s that unravel under scrutiny.
We’ve seen this before: 2021’s SPAC boom was full of LTV:CAC ratios that looked impressive until the cash burn outpaced user growth. It’s 2025, and the same pattern is emerging again -just wrapped in generative AI and embedded finance decks this time.
What Investors Should Ask Instead
If you’re an investor evaluating an early-stage startup’s unit economics, stop asking for CAC and LTV in isolation. Instead, focus on how they were calculated -and whether they’ve been pressure-tested against reality.
Here’s what to ask:
- Channel-Specific CAC
Don’t settle for blended CAC. Ask how CAC breaks down by acquisition channel, how those channels scale, and whether the math includes all-in spend (creative, headcount, engineering, tools). - Retention Cohorts, Not LTV
Ask to see retention curves by signup cohort. Ignore projections. Real LTV emerges from usage patterns, not forecasts. Bonus points if the team shows variable margin over time. - Real Margin
Is gross margin defined as revenue minus COGS, or does it conveniently exclude support, infrastructure, and onboarding? If those costs are meaningful (they usually are), demand to see a version that includes them. - Payback Period by Cohort
Payback periods only matter if you understand the assumptions. Ask to see them by signup month and acquisition source. If payback extends past the customer’s actual lifetime, you’re looking at fiction. - Sensitivity Scenarios
How does the model behave if CAC goes up 30% or churn increases by 2 points? If they can’t answer that, they’re not ready to scale.
The best founders already know this. They’ve stress-tested their metrics and can defend them under scrutiny. If a team can’t walk you through those inputs confidently, it’s a red flag -not because the numbers are bad, but because the thinking behind them is.
What Good Unit Economics Actually Look Like
They’re not flashy. They’re not extrapolated over 60 months. They don’t assume your next 10 customers will behave like your first 500.
They’re grounded in observed behavior, clean definitions, and aligned incentives. For example:
- CAC is calculated per channel, inclusive of all marketing and GTM cost.
- LTV is backward-looking, based on actual retention curves and real gross margin.
- Support and infra costs are allocated based on usage, not hidden in “corporate overhead.”
- Assumptions are stated explicitly -and can be adjusted by investors.
You don’t need to be perfect. But you do need to be honest.
Bottom Line
Unit economics aren’t a formality. They’re the foundation of your business model.
Founders who build fake models for pitch decks aren’t just misleading investors -they’re misleading themselves. And investors who accept those models at face value are underwriting dysfunction.
In 2025, we’re overdue for a return to first principles:
- What does it cost to acquire a customer?
- What do they actually generate in return?
- And how do you know?
If your startup can answer those questions with clarity and confidence -not slides and assumptions -you’re already ahead of most.