It’s 2025 and the startup game has fundamentally changed.

With capital more expensive and investors far more selective, growth-at-all-costs has given way to a mandate for efficient growth (Lightercapital, 2025, Capitaly, 2025). In a choppy market, capital efficiency is becoming a competitive moat – investors still want momentum, but they demand proof that you can scale without “lighting money on fire” (Capitaly, 2025).

In short, how wisely you deploy each dollar now often determines whether you reach the next milestone or run out of runway (Lightercapital, 2025).

Founders feel the pressure. Many early-stage teams are staying leaner and doing more with less, especially with new AI tools.

In fact, Carta’s data shows startup teams have gotten smaller and that the expectation today is to “do more with less”, particularly in AI startups (Peter Walker via PMF Show, 2025). This isn’t just lip service – it’s reflected in metrics and fundraising realities.

Only ~20% of seed-stage startups now make it to a Series A raise (Peter Walker via PMF Show, 2025). Those that do are typically the ones proving capital efficiency from the get-go.

Let’s dive in!

For the sake of this article, we focus on two core measures of capital efficiency: Burn Multiple and Revenue per Dollar Raised. These are the cleanest, most universally comparable ways to judge how effectively a startup converts capital into growth.

Metrics like CAC, LTV, or payback period depend on mature sales data, attribution assumptions, and stable cohorts, which most early-stage startups don’t have.

Burn Multiple and Revenue per Dollar Raised, by contrast, give you a better understanding: They rely only on cash burn and ARR, apply across sectors and stages, and can’t easily be gamed. They capture both real-time operating efficiency (Burn Multiple) and cumulative capital productivity (Revenue per Dollar Raised), giving founders and investors a direct, honest view of how far each dollar actually goes:

Burn Multiple (coined by VC David Sacks) measures how many dollars a company burns for each $1 of new ARR (Annual Recurring Revenue) it adds (Capitaly, 2025). It’s essentially net burn divided by net new ARR.

For example, burning $1.2M to add $1M ARR yields a burn multiple of 1.2× (i.e. you spent $1.20 for each $1 of ARR) (Capitaly, 2025). Lower is better: 1.0× means each dollar burned translated to a dollar of ARR; 0× or negative means you’re growing without cash burn (rare but ideal) (Capitaly, 2025). In today’s market this metric is the “ultimate efficiency scorecard” (Lightercapital, 2025).

Burn Efficiency Index (BEI) (also called “ARR per Dollar Raised”) looks at cumulative capital efficiency. It asks: How much ARR have you generated per $1 of total capital consumed (equity + debt)? This metric shows how well you convert all funding into recurring revenue. A higher BEI (more revenue per $ raised) signals you can do more with less. “How much ARR did you create per dollar burned?” is the question on everyone’s mind in 2025 (Lightercapital, 2025). The growing importance of this metric is emblematic of the mindset shift happening in venture.

For Burn Multiple, the industry has developed clear benchmarks by stage:

Pre-Seed/Seed (pre-$1M ARR): Burn multiples are often high as you search for product-market fit (typically around 2.0× to 3.0×) (Capitaly, 2025). It’s understood that early product development isn’t cheap. However, if you can keep it under ~2.0× once you start seeing consistent customer pull, you’re ahead of the game (Capitaly, 2025). Deep tech and hardtech startups may be on the higher end early due to heavy R&D, whereas product-led SaaS, and especially vertical AI companies, sometimes achieve far better efficiency sooner (Capitaly, 2025).

Series A ($1-8M ARR range): By the time you’re raising an A, expectations tighten. Typical burn multiples fall to ~1.0×–1.5× (Capitaly, 2025). In other words, for each $1 of new ARR, you’re burning about $1–1.5. Good companies in this stage often target ~1.2× or better (Capitaly, 2025).

Investors generally view anything above ~2.0× at Series A as a red flag unless your growth rate is truly exceptional (Capitaly, 2025). In fact, David Sacks suggests a <1.5× burn multiple by Series A is ideal to stay attractive (Valor, 2025). Many top-tier A-round startups keep burn around 1.0–1.3×.

Series B ($8-15M ARR) and Beyond: As scale kicks in, efficiency is expected to further improve. Series B startups often show burn in the 0.8×–1.2× range, trending toward ~1.0× as a target (Capitaly, 2025). By Series C ($15M+ ARR), the best companies drive burn multiples down to 0.5×–1.0× (Capitaly, 2025).

That means they’re adding revenue faster than they’re burning cash. An elite late-stage company might burn $0.6–0.8 to add $1 of ARR – a level of operating leverage that usually impresses growth investors. Broadly, <1.0× is “elite” efficiency at A/B rounds, and <0.7× is elite by Series C+ (Capitaly, 2025). On the other hand, going higher than 2.0× beyond Series A is a major concern (Capitaly, 2025).

Another simple rule of thumb: Burn multiple under ~1.5× is excellent, 1.5–2.5× is average, and >2.5× is often a warning sign (Lightercapital, 2025). For context, the productivity app Notion reportedly operated near a 1.0× burn multiple in 2023 (meaning every $1 it burned yielded about $1 of ARR) a KPPI that helped it attract huge funding even in a tight market (Lightercapital, 2025). On the flip side, a startup burning $3-4 to get $1 of ARR (Burn Multiple 3-4×) has efficiency issues to address.

For Revenue per Dollar Raised (BEI), concrete benchmarks are still emerging, but the idea is “more is better.” The concept isn’t new per se, but it’s only recently become a formally tracked metric because the market shifted from rewarding speed to rewarding efficiency.

During the 2010s bull run, capital was cheap, so investors focused on growth rate and market capture, not on how much capital it took to get there. Metrics like CAC, LTV, and Rule of 40 dominated because everyone assumed more funding was coming. Efficiency ratios like BEI didn’t matter when follow-on rounds were easy(ish).

That changed after 2022, when interest rates rose and venture funding tightened. Suddenly, investors started asking: “How much ARR did this company actually produce for every dollar it raised?” Carta, a16z, and others began publishing capital-efficiency data, and VCs started benchmarking BEI to compare portfolio productivity. In other words, BEI became “new” not because it’s a novel formula, but because it finally answers the question the market now cares about most - how much real business value each dollar of venture capital creates. A few famous cases exemplify this well:

Today’s investors reward those who can turn cash into revenue efficiently. If you can show, say, $2M ARR on $1M burned, or $5M ARR from $3M total raised, it signals strong product-market fit and operational discipline. As Elad Gil notes, a high capital efficiency usually reflects that (a) customers desperately want your product (will pay for it), and (b) the team is frugal and doesn’t over-hire or overspend . Those fundamentals never go out of style.

As Elad Gil notes, a high capital efficiency usually reflects that (a) customers desperately want your product (will pay for it), and (b) the team is frugal and doesn’t over-hire or overspend (Gil, 2023). Those fundamentals will probably never go out of style.

Share this article with others who might benefit.

Efficient growth is about hitting key revenue milestones faster and with less funding. Both the bar for what counts as traction, and the time you have to achieve it, have changed:

Series A Benchmarks Have Risen: The days of raising a Series A on a neat story and $500k–$1M ARR are over. In 2025, VC expectations are markedly higher. Many VCs now cite ~$2M ARR as the baseline for a strong Series A (Valor, 2025), often coupled with ~3× year-over-year growth to demonstrate momentum (Valor, 2025). In practice, yes, some A rounds still happen around $1M ARR, but those are the exceptions.

For an “exceptional” Series A, you’re really looking at $2–3M+ ARR to stand out (Valor, 2025). Carta’s data confirms this trend, showing $3M ARR is essentially the new bar for Series A in 2025 (Peter Walker via PMF Show, 2025). The result: Founders must build more real revenue before Series A, often on the same seed dollars.

Longer Runways to Hit A: Because the bar is higher, it’s taking startups longer to reach these milestones. Median time from Seed to Series A has stretched beyond 2 years now (up from ~1.5–1.6 years in the late 2010s) (Peter Walker via PMF Show, 2025).

A significant subset of founders (about 25%) are taking 3.5+ years to lock in a Series A (Peter Walker via PMF Show, 2025). In other words, if you raise a seed today, you might be looking at a 2–3 year marathon before your metrics justify an A. Many teams aren’t initially planning for that long a runway, which is why extension rounds or “seed+” rounds have become common.

The sobering reality is that only ~1 in 5 seed-funded startups will successfully reach a Series A under these tougher conditions (Peter Walker via PMF Show, 2025). Efficient growth is often the differentiator for those that make it.

Time-to-ARR Milestones: Hitting revenue milestones faster is a clear advantage. A few years ago, getting to $1M ARR in ~18–24 months was considered solid execution for a B2B startup. Now, that’s merely average. According to Stripe’s recent “Indexing the AI Economy” data, the median AI startup hits $1M ARR in ~11.5 months, whereas the median traditional SaaS startup takes ~15 months (Verna, 2025).

Likewise, reaching $5M ARR used to take SaaS companies 3+ years on average; AI startups are doing it in about 2 years median (Verna, 2025). This is partly because many AI or product-led companies can monetize from day one (no lengthy POCs or sales cycles), and they often have usage-driven adoption that accelerates growth.

Case in point for First-Year ARR: Andreessen Horowitz reports that among hundreds of recent startups they tracked, the median enterprise AI company got to $2M+ ARR in its first 12 months and was able to raise a Series A just 9 months after launching monetization (a16z, 2025).

Even more striking, the median consumer/AI company hit $4.2M ARR in its first year and raised an A within 8 months of revenue launch (a16z, 2025). What was once “best-in-class” (e.g. $1M ARR in 12 months) is now considered the lower end of the top cohort (a16z, 2025). The gap between a merely “good” trajectory and an “exceptional” one is widening even more.

Top performers aren’t just a bit better; they’re lapping the field in that first year. Not every startup will be an AI rocketship, of course. But the broader implication is that if you’re seeking venture funding, you need a strong velocity story. Either you have rapid revenue and user growth, or at a minimum you’re shipping product at breakneck speed to signal that momentum is coming (a16z, 2025).

Sector Nuances: Capital efficiency does vary by industry. SaaS businesses (especially with product-led growth models) tend to achieve revenue milestones with less burn compared to, say, fintech or biotech startups that face regulatory overhead or require capital for things like liquidity and R&D.

Within software, AI-native companies are setting a new performance bar – data shows they scale faster and with fewer resources than prior generations (Iconic Capital, 2025). On the enterprise side, infrastructure or deep-tech startups might have higher burn early on (building hard tech or deep IP is costly) (Capitaly, 2025).

On the other hand, pure software companies leveraging viral product loops or AI automation can often reach meaningful ARR with surprisingly small teams. (For example, there are AI startups hitting $10M+ ARR with under 10 employees – essentially $1M+ revenue per employee – a scenario almost unheard of a few years ago.) Meanwhile, B2C startups historically delayed monetization, burning cash to acquire users first, but that too is changing.

In the generative AI era, consumer apps are monetizing earlier, one-third of consumer AI startups in a16z’s sample had significant up-front investment in AI model training and then saw step-function revenue spikes when those models launched (a16z, 2025).

Surprisingly, recent data even showed consumer AI startups outpacing enterprise in first-year revenue (median $4.2M vs $2M) (a16z, 2025), proving that consumers will pay for compelling AI products. In financial services and health/biotech, efficiency tends to be measured over a longer horizon becuse these startups may need more capital before revenue kicks in. But once product-market fit clicks, they can scale in a steadier, more “predictable” trajectory (Mercury, 2025).

Know Your Numbers & Benchmarks: Track your burn multiple and ARR-per-dollar monthly, not just quarterly (Lightercapital, 2025). Benchmark against your stage and sector, e.g. if you’re at $3M ARR burning $4M/year (Burn Multiple ~1.3×), you’re on track. If you’re burning $10M to add $2M ARR (5×), hit the brakes. If your efficiency metrics start to slip, treat it like a smoke alarm: Cut spend or find revenue opportunities quickly (Lightercapital, 2025).

Extend Runway & Plan for 2+ Year Cycles: In today’s climate, assume it will take 18–24 months (or more) to hit the next funding milestone (Capitaly, 2025). Reforecast your burn to survive longer: Set a low-burn plan, hire only for truly essential roles, and keep optional expenses to a minimum (Capitaly, 2025). This gives you time to reach ARR targets before needing more capital, improving your negotiating position.

Prioritize High-Efficiency Growth Levers: Not all revenue is equal, some growth comes “cheap” and some comes expensive. Double down on product-led growth and retention. If your product can acquire users virally or via self-serve, that’s essentially free revenue (Lightercapital, 2025). Aim for a viral coefficient >1 (each user brings another) if possible.

Upsells and expansions are also gold: It’s far cheaper to grow revenue from happy customers than to win new ones. Many top startups get 30%+ of new ARR from expansions in their early rounds. Strive for Net Dollar Retention > 120% so your revenue keeps compounding without proportional spend (Lightercapital, 2025). This will dramatically improve your burn multiple. Conversely, be ruthless about cutting “growth” tactics that don’t pay back, like vanity marketing spend that isn’t generating pipeline .

Safeguard Gross Margins, Even with AI: If you’re an AI-heavy startup, keep an eye on cloud and API costs which can quietly erode your margins. Target 70–80%+ gross margins even if you leverage expensive AI infrastructure (Lightercapital, 2025). Optimize your tech stack and pricing to protect margin - it’s a huge driver of efficiency. High gross margin means more cash available to reinvest in growth without needing outside capital. Smart founders are already tracking AI-adjusted margins feature-by-feature to ensure no part of the product undercuts overall efficiency (Lightercapital, 2025).

Stay Scrappy and Frugal Culturally: Finally, instill a cost-conscious mindset in your team. The most capital-efficient companies tend to behave differently – they question each hire and expense.

Avoid premature scaling (e.g. hiring a big sales team before your sales motion is proven) (Capitaly, 2025). Use automation and AI to augment your team instead of throwing bodies at a problem. Remember that every dollar saved is runway earned. Accrding to Elad Gil, many of the iconic giants (Apple, Google, Amazon, etc.) started as highly efficient, frugal operations (Gil, 2023). Emulating that DNA early on will serve you well.

Thanks to Jérôme Jaggi for his help with this post.

Stay driven,
Andre