LTV to CAC Ratio Benchmark by Business Model
When Andreessen Horowitz looked at 60+ public consumer internet companies, moving the LTV:CAC ratio from roughly 2x to 3x nearly tripled company valuation. So there is no universal "good" ratio. The number 3:1 is a SaaS-margin artifact, and the defensible LTV to CAC ratio benchmark runs SaaS 3:1+, subscription mobile 4:1–5:1, DTC/ecommerce 2.5:1–4:1 (on contribution margin), and marketplaces 4:1+.
The number that surprised me
The tripling stat is the one that made me put my coffee down. Not because 3:1 is magic. It's because a16z's own 2023 write-up on why investors care about LTV:CAC frames 3x as "a rough benchmark of a consumer company's financial health" — a rough benchmark, measured inside a five-year window. Rough. Benchmark.
That word choice matters. Somewhere between 2010 and now, "3:1" stopped being a rough investor heuristic and became gravity. People quote it like a law of physics, when it's actually a valuation lever with conditions attached. The conditions are the whole story, and almost nobody who cites the number bothers to say them out loud.
I've been guilty of it myself, which I'll get to. For now, the point is this: the ratio is real, the valuation stakes are real, and the "correct" target depends entirely on what business you're in and — the part everyone skips — what margin your revenue converts into.
The canonical benchmark table
If you screenshot one thing from this piece, make it this. Every row is tied to its source so you can defend it in a room.
| Business Model | Revenue-LTV expectation | Margin-adjusted "good" range | LTV basis | Notes |
|---|---|---|---|---|
| SaaS | 3:1+ | 3:1+ genuinely holds | Revenue ≈ close to margin (75–85%) | Skok/Matrix ~2010, mature public SaaS at steady state |
| Ecommerce / DTC | Looks fine at 3:1 | 2.5:1–4:1 | 12-month cohort CM2 | Revenue overstates profit 50–70% (Eightx, 35+ brands) |
| Subscription mobile | 3:1 baseline | 4:1–5:1 | Revenue, payback <12 mo | Front-loaded churn inflates naive LTV (Adapty) |
| Marketplace | — | 4:1+ | Net take-rate contribution, not GMV | a16z looks for 4x+, tied to GMV retention |
The a16z "rough benchmark" framing sits behind that whole first column. It's a fine starting point and a terrible ending point.
Where the 3:1 rule actually came from
David Skok, then at Matrix Partners, popularized 3:1 around 2010. His data set wasn't scrappy startups burning venture money on Facebook ads. It was mature, public SaaS companies operating at steady state. High margin, predictable churn, revenue that converts almost cleanly into contribution.
That origin explains the rule's blind spot. When your gross margin is 80%, a dollar of revenue-LTV is almost a dollar of contribution-LTV, so revenue-based math and profit-based math land in roughly the same place. You can be sloppy and still be right. Move to a 55% margin business and that sloppiness quietly turns a 3:1 into something closer to 1.6:1 on the money you actually keep.
The rule isn't wrong. It's over-generalized. It was built for one model and then applied to four.
Why gross margin quietly moves the goalpost
Here's the thesis of the whole article, and it's boring in the way load-bearing things usually are: your ratio should be computed on contribution margin, not revenue.
a16z says this directly in its startup-metrics work. A common mistake is estimating LTV as the present value of revenue, or even gross margin, "instead of calculating it as net profit of the customer over the life of the relationship." Bennett Financials ran the same logic numerically and it's clean enough to memorize.
A 3:1 LTV:CAC at SaaS margins (~80%) delivers $2.40 of contribution per $1 of CAC. That same 3:1 ratio at service-business margins (~60%) delivers $1.80 of contribution per $1 of CAC. Identical ratio. Different truth. That's why Bennett argues service businesses on ~60% margins need roughly 4:1 to match what SaaS gets from 3:1.
Napkin math: same "3:1," different truth
Round numbers, because that's how I think when nobody's watching.
Say you've got $100 revenue-LTV and $30 CAC. On paper that's a clean 3.3:1. Now stop counting money you don't keep.
| Model | Gross margin | Contribution LTV | CAC | Real ratio (on CM) |
|---|---|---|---|---|
| SaaS | 80% | $80 | $30 | 2.7:1 |
| DTC | 55% | $55 | $30 | 1.8:1 |
The headline "3.3:1" survived contact with reality only in the SaaS row, and even there it slipped. The DTC version fell to 1.8:1. You're barely clearing your acquisition cost once you subtract COGS, shipping, and payment fees. Eightx's analysis of 35+ ecommerce and CPG brands puts a number on exactly this gap: revenue-based LTV overstates customer profitability by 50–70%. That's the range between "we're crushing it" and "we're underwater," decided entirely by which number you put in the numerator.
SaaS: the model the rule was built for
3:1 is a genuine floor here, and it's a fair one, because the margin structure does most of the work for you. At 75–85% gross margin, revenue-LTV and contribution-LTV are close enough that the sloppy version rarely lies to you by much.
That's also the trap. SaaS operators can quote revenue-LTV for years, look healthy, and never get punished for it. Then they expand into services, hardware, or a low-margin usage tier and keep using the same rubric. The rule that protected them starts hiding the problem. If you're pure software above 75% margin, 3:1 is a reasonable line in the sand. Just don't assume the line travels with you when the margin profile changes.
Ecommerce / DTC: the 3:1 misfit
This is where the borrowed SaaS rule does the most damage. DTC margins live in the 50–60% band, so the revenue-LTV that looks like a comfortable 3:1 is often a 1.6:1–1.8:1 once you compute it on 12-month cohort CM2. Eightx's read is the one I'd take into a board meeting: benchmark 2.5:1–4:1 on contribution margin two, not revenue, precisely because revenue overstates profitability by 50–70%.
Vertical spread matters too, and here I'd flag the numbers as directional agency benchmarks rather than gospel. AdZeta, citing Yotpo's 2026 data, gives gross-revenue ranges that vary a lot by category: supplements around 3:1–6:1, skincare and beauty subscription 3:1–5.5:1, mid-market apparel 2:1–4:1, high-AOV durables 1.5:1–3:1. The durables number tends to spook people. But a customer who buys a $1,200 mattress every eight years is a different animal than a supplement subscriber, and the ratio should look different.
One more AdZeta warning worth internalizing: ratios above 8:1 are usually not a flex. Brands running 8:1+ are "almost exclusively acquiring through low-cost organic and referral channels with limited scalability." An 8:1 often means you can't spend to grow, not that you're a genius. If your board is cheering an 8:1, ask them what happens when you try to double it.
Subscription mobile: front-loaded churn changes the math
Mobile subscriptions carry a specific hazard. Churn is brutally front-loaded. A huge slice of trial-and-first-month users bail fast, which means any LTV you calculate early — before those cohorts have aged — flatters you. Adapty's guidance is that investors want at least 3:1 as a baseline, and for subscription apps a 4:1 to 5:1 ratio is increasingly the expectation, ideally with a CAC payback period under 12 months.
The higher bar isn't arbitrary. It's a buffer against the naive-LTV problem. When you compute LTV on a young cohort, the survivors haven't churned yet, so your average revenue-per-user is temporarily inflated. Demanding 4:1–5:1 gives you margin for error when reality catches up and the curve bends down. The payback window is the reality check. If a 5:1 ratio comes with a 20-month payback, your ratio is technically fine and your cash position is not.
Marketplaces: two-sided CAC, GMV isn't revenue
Marketplaces break the standard formula in two places at once. First, CAC is two-sided. You're paying to acquire both supply and demand, and a16z evaluates LTV versus CAC for both sides of the market. Second, and this is the one that sinks people, GMV is not revenue. Your revenue is the take rate.
The classic mistake is computing LTV on the full GMV a buyer generates, which produces a gorgeous, meaningless number. a16z is blunt that LTV should be net profit over the relationship, not gross flows, and for marketplaces the gap between GMV and contribution is enormous. If your take rate is 15%, a buyer moving $10,000 of GMV isn't a $10,000 customer. They're closer to a $1,500 customer before you subtract the cost of servicing the transaction.
That's why a16z's marketplace bar is higher: it looks for 4x or higher, tied to strong GMV retention. The 4x isn't a stricter mood. It's what you need once you've correctly shrunk the numerator down to net take-rate contribution and accounted for acquiring both sides.
A quick self-own: the ratio I once defended to a CFO
Years ago I walked into a budget review with a slide that said 6.2:1. I was proud of it. I'd built the whole channel case around it.
The CFO looked at it for about four seconds and asked, "On what margin?" And I had nothing. I'd computed LTV on revenue, hadn't touched COGS or fulfillment, and my beautiful 6.2:1 was maybe a 3:1 on the money we kept. Still fine, honestly, but I couldn't prove it in the room, which meant it wasn't fine at all. I got the budget delayed a quarter for a rework I should've done up front. That meeting is why the margin section above exists. A ratio you can't decompose in front of a skeptic isn't a metric. It's a vibe.
Getting numbers you can defend
None of this survives in a stale spreadsheet. Cohort LTV and CAC are living things. Retention curves bend, payback windows shift, and a monthly export you touch once a quarter will lie to you by omission. This work belongs in a product-analytics layer that computes cohorts, retention, and funnels on current data.
A few tools handle that rubric. Amplitude and Mixpanel are the established names for cohort and retention analysis, and both are perfectly capable of the funnel work you'd need to build defensible LTV inputs. Kixo covers the same cohort, retention, and funnel ground and adds a chat-first query layer. You ask a question in plain language and get the chart or dashboard back with a visible reasoning trail, which is handy when a CFO asks "on what margin?" and you'd like to show your work. Honest caveat: Kixo is the newer entrant of the three, so weigh maturity against the interface if that matters to your stack.
Whatever you pick, the requirement is the same. The number has to be reproducible on demand, not reconstructed from memory the night before the board deck. If you want the mechanics of building those inputs cleanly, the fundamentals of setting up cohort and retention tracking are worth getting right before you argue about benchmarks.
How to pick your own threshold
Forget memorizing the table. Answer four questions and the table tells you which row you live in.
1. Revenue or contribution-margin LTV? If revenue, stop — recompute on contribution margin before you compare to anything. Revenue-LTV overstates profitability by 50–70% per Eightx's read, and every benchmark below assumes you've done this.
2. What's your gross margin band? Above 75%, the SaaS 3:1 floor holds. In the 50–60% range, you need a higher ratio to net the same contribution per acquisition dollar. Bennett's math says roughly 4:1 to match what SaaS gets from 3:1.
3. How front-loaded is your churn, and what's your payback window? If early churn is heavy (classic subscription mobile), demand 4:1–5:1 and confirm payback under 12 months, because young cohorts inflate naive LTV.
4. One-sided or two-sided CAC? If two-sided (marketplace), account for acquiring both supply and demand, compute LTV on net take-rate contribution rather than GMV, and hold yourself to a16z's 4x+.
Work those four in order and you'll land on a threshold you can defend line by line — which, as I learned the expensive way, is the only kind worth quoting. If you want the upstream discipline of clean attribution feeding these numbers, getting your acquisition tracking honest does more for the ratio than any benchmark ever will.
FAQ
What is a good LTV:CAC ratio for SaaS? 3:1 is a genuine floor for pure software at 75–85% margins — the model David Skok's ~2010 rule was built for, and the range a16z frames as a rough benchmark of financial health.
What is a good LTV:CAC ratio for ecommerce/DTC? Aim for 2.5:1–4:1 computed on 12-month cohort contribution margin, per Eightx's analysis of 35+ brands, since revenue-based LTV overstates customer profitability by 50–70%.
What is a good LTV:CAC ratio for subscription mobile apps? Adapty's guidance is 4:1–5:1 (above the 3:1 baseline) with CAC payback ideally under 12 months, because heavy early churn inflates any LTV you calculate on young cohorts.
What is a good LTV:CAC ratio for marketplaces? a16z looks for 4x or higher, computed on net take-rate contribution rather than GMV and accounting for the cost of acquiring both sides of the market.