Magic Number, Burn Multiple, and Payback, Explained
The three metrics answer three different questions. The sales efficiency magic number asks whether your go-to-market machine turns dollars into recurring revenue fast enough. The burn multiple asks how much cash the whole company torches to add a dollar of ARR. CAC payback asks how many months until a new customer has paid back what you spent to win them. Same efficiency lens, three altitudes: the GTM team, the board, and the marketing budget.
Here's the number that made me rewrite a whole plan once. I had a paid-acquisition program I was quietly proud of. Blended CAC looked fine, the dashboard was green, everyone nodded in the QBR. Then someone asked for CAC payback by channel, and one channel came back at 29 months. Twenty-nine. We were funding it out of the healthy channels and calling the average a strategy. That's the thing about efficiency metrics: the blended number is where bad channels go to hide.
So let's put the three in one place, spell out the formulas, and, the part most articles skip, say which one you should actually be looking at depending on who's asking.
The one-table version
If you remember nothing else, remember this.
| Metric | The question | Formula | "Good" in 2025 | Who cares most |
|---|---|---|---|---|
| Magic number | Is GTM spend converting to revenue? | (Net new revenue this Q × 4) ÷ prior-Q S&M | ~0.75+ efficient, 1.0+ excellent | GTM / RevOps |
| Burn multiple | How much cash per $1 of new ARR? | Net burn ÷ net new ARR | <1 amazing, <1.5 great, <2 good | The board / CFO |
| CAC payback | Months to recover a customer's cost? | CAC ÷ (monthly ARPA × gross margin) | Under 12 top-tier, ~16 median | Marketing team |
Three metrics, three owners. They overlap on purpose. When they disagree, that disagreement is the interesting part, and I'll get to it.
Magic number: the GTM efficiency read
The magic number is the one your revenue team should live in, because it's the tightest loop between "we spent money on sales and marketing" and "revenue showed up."
The formula, per Wall Street Prep: take the increase in revenue from one quarter to the next, multiply by four to annualize it, and divide by what you spent on sales and marketing the previous quarter. The one-quarter lag is deliberate. Money you spend in Q1 mostly closes in Q2, so you compare Q2's new revenue against Q1's spend.
Napkin math, round numbers so the mechanics are obvious. Say you did $10M ARR at the end of Q1 and $12M at the end of Q2. That's $2M in net new ARR (I'm using ARR here for cleanliness; strict definitions use GAAP revenue). You spent $2M on sales and marketing in Q1. Annualize the new revenue and it's already $2M on an ARR basis, so:
$2M ÷ $2M = 1.0.
A magic number of 1.0 means every dollar of S&M bought a dollar of new annual recurring revenue inside a quarter. Wall Street Prep and most investors call anything above roughly 0.75 efficient and above 1.0 excellent. Below 0.5 is the alarm bell — usually it means acquisition cost is running well ahead of what those customers are worth, or churn is quietly eating the new-revenue line before it can compound.
Here's the reading I'd actually hang on the wall:
| Magic number | What it's telling you | The move |
|---|---|---|
| Below 0.5 | GTM isn't paying for itself | Fix ICP and retention before you spend more |
| 0.5 – 0.75 | On the right track, not efficient yet | Tune channels, don't floor it |
| 0.75 – 1.0 | Efficient. Growth is affordable | Consider adding fuel |
| Above 1.0 | You may be underspending | You can probably afford to be more aggressive |
That last row surprises people. A magic number that's too high isn't a trophy — it can mean you're leaving growth on the table because you're scared of the spend. A CFO I worked with used to say a magic number stuck at 1.5 for four quarters was a sign the marketing team needed a bigger budget, not a smaller one. She was usually right, which was annoying.
The catch: the magic number is blind to gross margin and to churn timing. It'll happily call a low-margin, high-churn business "efficient" for a quarter or two because the new-revenue numerator looks great before the cohort starts leaking. Which is why nobody sane runs it alone.
Burn multiple: the number the board opens with
If the magic number is a GTM metric, the burn multiple is a whole-company one. David Sacks introduced it in a 2020 essay with a formula so blunt it's almost rude: net burn divided by net new ARR. How much total cash did you set on fire to add one dollar of recurring revenue? Not just S&M. Everything — engineering, G&A, the office plants.
Sacks gave it a five-band scale that's now more or less the industry vocabulary:
| Burn multiple | Sacks' label | Rough meaning |
|---|---|---|
| Under 1.0x | Amazing | You're barely burning to grow |
| 1.0 – 1.5x | Great | Near breakeven, strong efficiency |
| 1.5 – 2.0x | Good | Healthy for early-stage in growth mode |
| 2.0 – 3.0x | Suspect | Investors start asking hard questions |
| Over 3.0x | Bad | Close scrutiny required |
What's changed since 2020 isn't the formula, it's the grading curve. The 2021 cohort could raise on a burn multiple of 2 to 3 because growth was the only screen anyone cared about and money was cheap. That's over. The Growth Equity Interview Guide notes that a16z now pegs a "good" burn multiple at around 1.1x for companies between $0 and $10M ARR, and top-tier funds in 2025 expect under 1.5x from Series A onward. The bar gets lower as you get bigger, not higher — more revenue should mean more efficiency, not permission to burn more.
One honest caveat, and Sacks would agree: a beautiful burn multiple doesn't mean you can't die. You can post a 1.0x and still run out of cash if your absolute burn is large relative to the bank. The multiple is about efficiency of growth, not survival. Read it next to your runway, always.
CAC payback: the marketing team's monthly reality
CAC payback is the most tangible of the three because it's in a unit humans understand: months. Spend to acquire a customer, divide by the gross-margin-adjusted revenue that customer throws off each month, and you get the number of months until you're whole.
The gross-margin adjustment matters and gets dropped constantly. If a customer pays you $500/month but your gross margin is 80%, they're only returning $400/month of actual contribution. Skip the margin and you'll flatter every payback number you produce. I've done it. It's a great way to feel good in a deck and bad in a board meeting.
Napkin math again. CAC of $6,000. Customer pays $500/month. Gross margin 80%, so contribution is $400/month.
$6,000 ÷ $400 = 15 months.
Fifteen months to break even on that customer. Is that good? Depends entirely on who you sell to. First Page Sage's 2025 report puts the median CAC payback around 16 months, with top-quartile companies recovering CAC in roughly 6 months and the bottom quartile grinding past 24. The spread is enormous, and it tracks deal size: SMB tends to land in the 8–12 month range, mid-market 14–18, enterprise 18–24. A 15-month payback that's a disaster for a self-serve SMB tool is completely normal for an enterprise contract with a two-year commitment behind it.
The rule I use: under 12 months is the bar for a company that can self-fund its own growth, because CAC comes back fast enough to redeploy into the next cohort before you need to raise again. Under 18 is fine. North of 24 and you're financing growth with someone else's balance sheet, which is a choice, just not always a conscious one.
Which one, and when
This is the part the internet keeps splitting across three separate articles, so here's my straight answer.
Use the magic number when you're deciding whether to spend more on GTM. It's the fastest-reacting of the three and it's scoped to the exact lever a growth team controls. When someone asks "can we afford to double paid?", the magic number is the first place I look. Quarter-over-quarter, it moves.
Use the burn multiple when you're talking to the board or a prospective investor. It's the number that gets you a meeting or gets you passed on. Bessemer's commentary on the 2025–2026 vintage put burn multiple right behind net revenue retention as the metric investors screen on before they'll even take the call. If you're fundraising, know this number cold and know it before they ask.
Use CAC payback when the marketing team is allocating across channels. It's the one that exposes the 29-month channel hiding inside a healthy blended average. Payback is where I'd start any channel-mix conversation, because it's per-cohort and per-channel in a way the other two aren't.
None of the three should be blended away, and here's my one assumption box, because our house style is to show its cards:
Assumptions I'm making (disagree freely):
- You're a recurring-revenue business. These metrics get weird for usage-based or transactional models, and net new ARR isn't clean.
- You compute them per-segment, not just blended. Blended is where the 29-month channels live.
- Your gross margin number is real, not aspirational. Every one of these bends if margin is soft.
- You read each metric next to runway. Efficiency and survival are different questions.
The blended-versus-segmented point is the one that's cost me the most, so I'll die on that hill. The whole reason to run these three together is triangulation. If your magic number says GTM is efficient but your burn multiple is 2.5x, the inefficiency is somewhere other than sales and marketing — probably a bloated cost base or an R&D bet that hasn't shipped. If payback is short but the burn multiple is ugly, you're acquiring customers cheaply and then spending the savings somewhere off-camera. The metrics disagreeing is a map to where the money's actually going.
Getting these consistently means wiring revenue, spend, and cohort data into the same view rather than three spreadsheets that reconcile once a quarter. Some teams build that into a warehouse; others let a chat-first analytics tool like Kixo generate the cohort chart from a plain-language question. Either way, the goal is the same: stop letting the blended average do your thinking for you. If you want a starting layout, our unit-economics dashboard template puts CAC, payback, and margin on one page.
Run all three. Trust the disagreements. And check the channel-level payback before you're proud of anything. I learned that one the expensive way.