Startup Burn Rate: How to Calculate, Benchmark, and Cut It

Startup burn rate is the amount of cash your company spends beyond what it earns each month. Calculate it by subtracting monthly revenue from total monthly expenses — that's your net burn rate, and it's the single number that determines how long you can survive before the money runs out. If you're not tracking this metric monthly, you're guessing at your timeline, and guessing at runway is how startups die quietly.

The urgency is real right now. According to a SaaStr analysis from July 2026, companies quintupled their AI token spend in the first half of the year — and almost nobody can point to the revenue lift that justified it. New cost categories like LLM API calls, vector database hosting, and GPU compute are inflating burn rates across the startup ecosystem, often before founders even update their financial models to reflect the change.

What Is Burn Rate?

Burn rate is the speed at which a startup consumes cash over a given period — typically measured monthly. It comes in two flavors that answer different questions.

Gross burn rate is your total monthly cash outflow: salaries, rent, software, marketing, infrastructure — everything. It tells you what it costs to keep the lights on.

Net burn rate subtracts monthly revenue from that total. It tells you how fast your cash balance is actually shrinking. For pre-revenue startups, gross and net burn are identical. Once you start generating revenue, the gap between them is the most important number in your financial model.

Gross Burn Rate = Total Monthly Expenses
Net Burn Rate   = Total Monthly Expenses − Total Monthly Revenue

Here's the thing: most founders track revenue obsessively but treat expenses as a lump. That's backwards. Your monthly recurring revenue might be growing 10% month-over-month, but if your burn is growing 15%, you're actually losing ground. The net burn trend — not the absolute number — is what separates startups that reach profitability from those that run out of time.

How Do You Calculate Burn Rate Step by Step?

Calculate burn rate by summing all cash outflows for the month, then subtracting cash inflows from operations.

Step 1 — Total your monthly expenses. Pull actuals from your bank account or accounting software, not your budget. Include payroll (fully loaded — salary, benefits, taxes), rent or coworking costs, software subscriptions, cloud infrastructure, marketing spend, legal and accounting fees, and any contractor payments. This is your gross burn.

Step 2 — Total your monthly revenue. Use cash received, not accrued revenue. For SaaS businesses, this is typically your MRR collected. For e-commerce, it's net revenue after refunds and chargebacks.

Step 3 — Subtract revenue from expenses.

Net Burn Rate = Total Monthly Expenses − Total Monthly Revenue

Worked example — NovaTech (seed-stage SaaS): NovaTech spends $95,000/month (payroll $62,000, cloud infra $12,000, marketing $11,000, office and ops $10,000). They generate $18,000 in MRR.

• Gross burn rate: $95,000/month
• Net burn rate: $95,000 − $18,000 = $77,000/month
• With $1.4M in the bank: runway = $1,400,000 ÷ $77,000 = 18.2 months

That 18-month number is the one that matters in board meetings and fundraising conversations. It's also the one that changes fastest when new cost categories — like AI API spend — creep in untracked.

Step 4 — Calculate your runway. Divide your current cash balance by net burn rate:

Runway (months) = Cash Balance / Net Burn Rate

For a deeper look at what runway means for fundraising timing, see the runway metric guide.

Calculate Your Burn Rate

What's a Good Burn Rate for Your Stage?

There is no universal "good" burn rate — but there are benchmarks by stage that tell you whether your spending is proportional to your opportunity and your cash position.

The real question isn't "how much am I burning?" — it's "how many months of runway does my burn rate give me, and is my growth rate fast enough to justify it?"

Stage	Typical Net Burn	Target Runway	Growth Rate to Justify
Pre-seed	$15K–$40K/mo	18–24 months	Proving PMF, not growing yet
Seed	$30K–$80K/mo	18–24 months	10–15% MoM revenue growth
Series A	$80K–$200K/mo	18–24 months	15–20% MoM revenue growth
Series B+	$200K–$500K/mo	18–30 months	Varies by path to profitability

Two caveats worth flagging. First, these ranges assume U.S.-based startups with primarily domestic teams. Remote-first or non-US teams can operate at 40–60% of these numbers. Second, the "target runway" column is pre-fundraise — by the time you start raising, you should have at least 6 months of cushion, because rounds take longer than founders expect.

A useful cross-check: your burn rate should produce a burn multiple under 2x once you're post-revenue. Burn multiple = net burn ÷ net new ARR. Anything above 2x means you're spending more than $2 for every $1 of new annual revenue — a signal that growth isn't efficient enough to sustain the spend.

Why Is AI Spending Making Burn Rate Harder to Manage?

The SaaStr report on the token ROI crisis landed a clear warning: companies across the startup ecosystem quintupled their AI token spend in H1 2026 without a measurable revenue lift. That's a burn rate problem masquerading as an innovation initiative.

AI infrastructure costs behave differently from traditional SaaS expenses in three ways that trip up financial models:

1. Usage-based scaling. Traditional SaaS costs are largely fixed or step-function (you add engineers in discrete hires). LLM API costs scale linearly — or worse, super-linearly — with usage. A feature that costs $200/month in development can cost $20,000/month once customers actually use it.

2. Hard-to-predict unit economics. The cost per API call varies by model, prompt length, and provider pricing changes. OpenAI, Anthropic, and Google have all changed pricing multiple times in 2025–2026. If your financial model uses a fixed cost-per-inference assumption, it's probably already wrong.

3. Invisible allocation. Many startups lump AI costs into "cloud infrastructure" alongside hosting, CDN, and storage. That means the burn increase from AI adoption doesn't show up as a separate line item until someone audits the bill — usually after the runway has already shortened.

On the flip side, AI can also reduce burn when deployed strategically. Vercel's COO recently shared that they replaced a 10-person SDR team with an AI-driven GTM system that costs $5,000 per year. That's going from roughly $600K+ in annual headcount costs to $5K — a massive burn reduction. The difference between these two AI outcomes comes down to whether you're modeling the ROI before you commit the spend.

The fix: add a dedicated "AI / LLM costs" line item to your financial model. Track cost per feature, cost per user, and cost per transaction separately. Set a kill threshold — if a feature's AI costs exceed X% of the revenue it generates after 90 days, cut it or optimize it.

How to Reduce Burn Rate Without Killing Growth

Audit Variable Costs Monthly — Not Quarterly

Most startups review expenses quarterly at best. By then, a $3,000/month cost creep has become a $9,000 hole in your runway. Set a monthly ritual: export your bank statement, categorize every line, and flag anything that grew more than 10% from the prior month. SaaS subscriptions are the usual suspects — the average startup pays for 3–5 tools nobody actively uses.

Renegotiate Annual Contracts Before Renewal

Vendors expect negotiation. Most SaaS companies will offer 15–25% discounts for annual commitments or multi-year deals. Infrastructure providers often have startup credit programs (AWS Activate, GCP for Startups, Azure Founders Hub) that can offset $10K–$100K in cloud costs. If you haven't applied, you're leaving free runway on the table.

Hire for Revenue Impact, Not Headcount Milestones

Payroll is the largest line item for 80%+ of startups. Before every hire, model the expected revenue impact: does this person accelerate time-to-revenue, reduce churn, or unlock a new segment? If you can't tie the role to a revenue outcome within 6 months, it's a nice-to-have, not a need-to-have.

Model Your Fundraise Timeline Backwards

Start with your target raise date, subtract 6 months for the fundraising process, and that's your "runway alarm" date. If your current burn rate puts you below 6 months of cash at that point, you need to cut now — not when the bank balance triggers panic. Build this into your SaaS financial model as a scenario: "What happens to runway if revenue growth slows to half our base case?"

Key Takeaways

• Net burn rate = total monthly expenses minus monthly revenue — track it monthly, no exceptions, and include every cost category including AI/LLM spend
• Target 18–24 months of runway at your current net burn rate after any funding round; start fundraising with at least 6 months of cushion
• AI token costs are the fastest-growing burn category in 2026 — companies quintupled spend in H1 with no measurable ROI; add a dedicated line item and set kill thresholds
• Burn multiple under 2x is the efficiency benchmark once you're post-revenue; above 2x signals that your growth doesn't justify your spend
• Monthly expense audits catch cost creep 3× faster than quarterly reviews — SaaS subscriptions and cloud costs are the most common sources of invisible burn

Your burn rate isn't a fixed number — it's a lever. The founders who survive downturns and choppy fundraising markets are the ones who model their burn against multiple revenue scenarios before the cash gets tight. Build your burn rate model in Revenue Map and run the scenarios now, while you still have time to act on them.