पाठशाला Pathshala · संचालन Sanchālan, Operations · Lesson 03 · Scale

Burn multiple and the discipline of efficient growth

One number that says what a crore of burn is buying: net burn divided by net new recurring revenue. Why it caught every kind of waste at once, and why India learned it the hard way after 2022.

Pathshala, The Founder Library · 10 October 2026 · 9 min read

Two companies each burn ₹4 crore in a year. One adds ₹3 crore of recurring revenue with it and the other adds ₹1 crore. The second company has the same runway, the same bank balance and the same slide about market size, and it is a worse business by a factor of three. Burn multiple is the number that says so in one figure, and since 2020 it is the first thing a growth-stage investor computes.

This lesson gives the definition and the bands as the man who coined them published them, places the number next to the two older efficiency measures it is usually confused with, explains what a bad multiple is actually telling you, and sets out what changed in India after the money stopped in 2022.

The definition and the bands

David Sacks of Craft Ventures published The Burn Multiple on 23 April 2020, a month into the pandemic, when every board was asking how much growth its cash was buying. The formula is net burn divided by net new ARR, over the same period. Net burn is cash out less cash in from operations, excluding anything raised. Net new ARR is annual recurring revenue at the end of the period less ARR at the start, which means new customers plus expansion minus churn and contraction.

His worked example: a company that burns $2 million in a quarter to add $1 million of ARR has a burn multiple of two, which he called reasonable for an early-stage startup. A company that burns $5 million to add the same $1 million has a multiple of five, which he called terrible. Sacks’s bands for venture-stage companies, reproduced from the chart in his post and carried by investors since, are: under one is amazing, one to one and a half is great, one and a half to two is good, two to three is suspect, and above three is bad. He added the staging most founders miss: a seed company might sit near three and a company after its Series A near two, and a multiple of three or more at any later stage suggests that product-market fit is not what it appears to be, or that something else in the business is wrong.

Convert to rupees and the example above reads itself. ₹4 crore of burn adding ₹3 crore of ARR is a multiple of 1.33, which is great; every crore burned bought ₹75 lakh of recurring revenue, and the new revenue repays the burn in sixteen months. The same ₹4 crore adding ₹1 crore is a multiple of four, bad on any stage, and the new revenue takes four years to repay what was spent to win it. Nobody with the second set of figures should be hiring.

The gauge places the company in Sacks’s bands. The second readout is the one to carry into a board meeting: what a crore of burn bought. Move ARR at the end down by ₹50 lakh to see what one large customer churning does to the multiple, then move burn up by ₹1 crore to see what one quarter of over-hiring does. Both effects are large and both are quiet in the monthly accounts.

Why one number catches every kind of waste

The reason the burn multiple displaced older measures is that it is hard to game. Sacks’s claim was that any serious problem in a SaaS business eventually shows up in it, and he listed the routes. A gross margin problem, where the cost of serving customers is high and does not fall with scale, raises burn without touching ARR. A sales efficiency problem, where CAC is prohibitive or productivity per salesperson is falling, raises burn for each rupee of ARR added. A churn problem nets directly against the denominator: a team that signs ₹4 crore of new business and loses ₹1 crore to churn has added ₹3 crore, and the multiple knows it even when the sales dashboard does not. A growth problem shows up as money spent on marketing, give-aways, discounts and promotions to keep the top line moving, all of which is burn. And a founder who lacks the skill or the will to control spending shows up in the numerator every quarter.

Contrast this with the metrics a growth-stage company usually presents. ARR growth alone can be bought. CAC alone ignores churn and margin. Gross margin alone ignores sales cost. Runway alone ignores what the money is doing. Burn multiple is the one figure that requires all of them to be healthy at once, which is exactly why it is uncomfortable and exactly why investors like it.

Growth tells you the company is moving. Burn multiple tells you what the movement cost.

The older cousins: Bessemer’s efficiency score and the Rule of 40

Burn multiple did not appear from nowhere. Bessemer Venture Partners had been publishing an efficiency score in its State of the Cloud reports for several years before 2020, and for companies below $30 million of ARR it is the same fraction turned upside down: net new ARR divided by net burn. Bessemer’s 2019 report set the best score for a company at that stage at above 1.5, which is a burn multiple below 0.67; their earlier rule of thumb was simply that the score should exceed one, meaning a dollar of ARR for every dollar burned. Sacks inverted the fraction so that lower is better and the bands run in the direction founders think about spending. The two measures agree completely; use whichever your investor uses and translate.

For larger companies Bessemer changes the definition. Above $30 million of ARR the efficiency score becomes year-on-year ARR growth plus free-cash-flow margin, and their benchmarks by scale target a score of 70 between $25 million and $50 million of ARR, falling to about 50 above $100 million, with the average company in their public cloud index closer to 50. That second formula is a relative of the Rule of 40.

The Rule of 40 entered circulation through Brad Feld’s post The Rule of 40% For a Healthy SaaS Company in February 2015. He had heard it from a late-stage investor at a board meeting: growth rate plus profit margin should add up to 40 per cent, with EBITDA as the profit measure. A company growing 60 per cent a year may lose 20 per cent of revenue; a company growing 20 per cent must make 20. Feld was explicit about scope: it applies to SaaS companies at scale, which he put at a minimum of $50 million in revenue.

That scope line is the point. The Rule of 40 is a yardstick for companies with hundreds of crores of revenue and the operating history to choose between growth and margin. A company at ₹20 crore of ARR that computes it will get a meaningless number, usually a large negative one, because at that stage the company should be investing nearly everything in growth. The right question at ₹20 crore is not whether growth plus margin is 40 but whether each crore of burn is buying at least half a crore of ARR. Burn multiple is the early-stage instrument; the Rule of 40 is the one you graduate to.

Computing it properly

Four details decide whether the number is honest. First, the period: compute it quarterly for the board and annually for the raise, and never on a single month, because ARR moves in lumps and one enterprise deal can make a bad quarter look brilliant. Second, net burn means operating cash flow, which excludes the round you just closed and includes the GST you paid, the TDS you deposited and the salaries that went out, because all of it is cash the business consumed. Third, ARR must be recurring: multiplying one month’s all-in bookings by twelve drags in one-time fees, implementation charges and professional services that will not recur, and none of it belongs in the denominator. Fourth, net new ARR is net of churn. Reporting gross new ARR and calling the result a burn multiple is the single most common way the figure is flattered, and a diligence analyst will catch it in an afternoon.

For businesses without recurring revenue the formula adapts: a marketplace uses net new annualised net revenue, meaning take rate, not GMV; a transactional business uses annualised gross profit. The principle is the same. Divide the cash consumed by the durable earning power it created.

What India learned after 2022

The burn multiple became the ruling metric in India not through a blog post but through a correction. Indian startups raised $42 billion in 2021 according to Inc42’s data, with 44 companies reaching unicorn valuations in a single year. In 2022 the total fell to $25 billion, a 40 per cent decline, and the unicorn count to 21. The full year 2023 closed at just over $10 billion, the lowest in seven years, and excluding rounds above $100 million the figure was about $5.5 billion, barely above the whole of 2016. Layoffs dominated the sector’s headlines for two years.

What that did to the conversation in every board room in Bengaluru, Gurugram and Mumbai is simple to describe. Between 2019 and 2021 the question was how fast, and burn was the price of speed. From the middle of 2022 the question became what the burn was buying, and companies that could not answer with a multiple under two found that the next round was a down round, a bridge from existing investors or nothing. Growth-stage term sheets written since then routinely carry burn-multiple covenants or reporting requirements, and a founder who arrives at a Series B without the number ready is signalling that they have not been reading their own accounts.

The lesson is not that burn is bad. It is that burn is a purchase, and a purchase has a price. A company with a multiple of 1.2 should probably be burning more, because its growth is cheap and the money is well spent. A company at 3.5 should stop hiring until it knows why. The discipline of efficient growth is the habit of knowing the price before writing the cheque.

A quarterly ritual, in one hour

In the first week after each quarter closes, finance produces three numbers: net operating burn for the quarter, ARR at the start, ARR at the end, with the ARR figure reconciled to the invoicing system and net of churn. The founders compute the multiple, place it in Sacks’s bands, and compare it to the previous four quarters on one line of a chart. Then they ask three questions in order. Which of the five routes, margin, sales efficiency, churn, bought growth or spending control, moved the number this quarter? What is the hiring plan for the next two quarters, and what multiple does it imply if ARR grows at the rate of the last two? And if the multiple is above two and the company is past its Series A, what single cut or single pricing change brings it under two within two quarters?

Write the answers in the board pack next to the gauge. A board that sees the multiple every quarter stops asking about runway, because runway is downstream of this number. So is the next round, and so is whether there is one.

Nothing here is financial advice. The bands are Sacks’s, the scores are Bessemer’s and the funding figures are Inc42’s; each is cited below so you can check the arithmetic against your own.

Sources

1. David Sacks, The Burn Multiple, Craft Ventures, April 2020 — The definition, the bands and the five problems a high multiple reveals.
2. Bessemer Venture Partners, State of the Cloud 2019 — Efficiency score as net new ARR over net burn for companies under $30 million of ARR; best above 1.5.
3. Bessemer Venture Partners, Scaling to $100 million — Efficiency score as growth plus FCF margin at scale; targets of 70 and 50.
4. Brad Feld, The Rule of 40% For a Healthy SaaS Company, February 2015
5. Inc42, 2022 In Review: $25 Bn Funding, 21 Unicorns, December 2022 — 2021 at $42 billion and 44 unicorns; 2022 at $25 billion and 21.
6. Inc42, 15 Charts That Defined India’s Tech And Startup Ecosystem In 2023, December 2023 — 2023 at just over $10 billion, a seven-year low.