Unit economics for marketers 2026: LTV, CAC, Payback Period with real numbers
How to calculate unit economy correctly: LTV/CAC ratio, Payback Period, ARPU. Benchmarks for Russian niches 2026, common mistakes and formulas with examples.
Once every two weeks, some founder or CMO writes to me: “We have a CAC of 3000 ₽, is this normal?” Without LTV, without margin, without Payback Period - the question is meaningless. I’m figuring out how to calculate the unit economy so that the numbers actually decide something.
I have been working with unit economics on client projects since 2019. During this time, I went through a typical path: at first I considered CAC as an advertising budget / new clients. Then I realized that this is a marketing illusion, not a metric. On the current six projects, the real CAC is on average 1.8 times higher than the “budget” one - due to the team’s salaries and the cost of tools, which no one usually considers.
The main misunderstanding about unit economics is that they think it’s about formulas. It’s more correct to think that this is about decisions. A formula without a conclusion “what to do with a budget” is an academic exercise.
1. Why 90% of companies think the unit economy is wrong
The problem is not ignorance of the formulas. The problem is that they count convenient numbers, not real ones.
Typical picture: a marketer reports a CAC of 1,500 rubles. This is the advertising budget for the month divided by new clients. Sounds good. But if you add his own salary (in proportion to the time spent on this channel) + contractors + CRM + automation services, the real CAC comes out to 2,700–3,200 rubles. Already a different picture.
Three mistakes that I see in every second audit:
1. CAC = advertising budget only. The most common. The salary of the marketing team, agent commissions, the cost of tools - all this is Cost of Acquisition, but it is taken out of the equation. As a result, CAC is reduced by one and a half to two times.
2. LTV is calculated by the average check, and not by cohort. “The average check is 5,000 ₽, they buy on average 3 times - LTV 15,000 ₽.” Problem: “on average 3 times” is the average of the entire database, including those who bought once five years ago. A real cohort of customers often shows an LTV of half as much 12 months after their first purchase.
3. They look at LTV/CAC and ignore Payback Period. LTV/CAC 4× looks great. But if the Payback Period is 24 months and the cash gap is already in 6 months, the company will not live to see the moment when the numbers “converge”. Cash flow kills earlier.
2. CAC: what actually goes into the formula
Full formula:
CAC = (advertising budget + marketing team salaries + agency commissions + cost of tools) / new clients for the period
What's included in each article:
- Advertising budget - Yandex Direct, VK Ads, Telegram Ads, target, context.
- Salaries are only that portion of working time that is spent on attraction (not on retention, not on branding). If a marketer spends 60% of his time on acquisition, we take 60% of his salary.
- Agents and contractors - SEO agency, SMM team, designers for advertising creatives.
- Tools - CRM (amoCRM, Bitrix24), email newsletter services, call tracking, analytics.
Case study: e-com client, electronics niche. Advertising budget - 500K ₽/month, new clients - 400. “Budget” CAC - 1250 ₽. We add the salary of two marketers (150K ₽ × 70% = 105K ₽), agency (60K ₽), tools (25K ₽) - total expenses 690K ₽. Real CAC - 1725 ₽. 38% higher.
Important distinction: Sales CAC vs Marketing CAC. If you have a sales department, consider separately: Marketing CAC (everything up to a lead) and Sales CAC (cost of closing a lead into a client). In B2B they are often comparable.
3. LTV: three methods of calculation (and which one is fair)
Three methods, from simple to precise:
Method 1 - simplified: LTV = average check × average number of purchases per year × average customer lifetime (years). Fast but dirty. Suitable for initial assessment when data are missing.
Method 2 - via Churn Rate: LTV = ARPU / Churn Rate. If the average monthly revenue per client is 2,000 rubles, and the outflow is 5% per month, then LTV = 2,000 / 0.05 = 40,000 rubles. Works for subscription models. Problem - Churn Rate is unstable and depends on the cohort.
Method 3 - cohort (honest): We take all clients who made their first purchase in January 2024 and look at their total revenue after 3, 6, 12, 24 months. This is real LTV, not calculated. In practice, it is usually 20–40% lower than the simplified one.
I recommend the cohort method whenever there is 6+ months of data. Method 1 is only for new products when cohorts have not yet accumulated. Method 2 - for SaaS with clear monthly churn.
About LTV/CAC Ratio For more details, see the glossary.
4. LTV/CAC Ratio: interpretation and benchmarks
The formula is simple: LTV / CAC. Interpretation:
- LTV/CAC >3 – healthy unit economy. Can be scaled.
- LTV/CAC 1–3 - works, but subtly. One change in the market or an increase in CAC - and we go into the negative.
- LTV/CAC <1 - each new client burns money. The channel needs to be stopped.
Benchmarks for Russian niches 2026:
| Niche | LTV/CAC norm | Payback Period norm | Typical CAC | Typical LTV |
|---|---|---|---|---|
| SaaS (B2B) | 4–7× | 9–18 months | 15 000–80 000 ₽ | 80 000–400 000 ₽ |
| SaaS (B2C / SMB) | 3–5× | 6–12 months | 2 000–8 000 ₽ | 8 000–35 000 ₽ |
| EdTech (online courses) | 3–5× | 3–8 months | 3 000–12 000 ₽ | 12 000–45 000 ₽ |
| E-com (average bill up to 5K ₽) | 2–4× | 1–3 months | 600–2 500 ₽ | 3 000–8 000 ₽ |
| E-com (average bill 5–30K ₽) | 2–3× | 2–6 months | 2 000–8 000 ₽ | 8 000–25 000 ₽ |
| Mobile applications (freemium) | 2–4× | 3–9 months | 150–600 ₽ | 400–2 000 ₽ |
These are market medians. The actual numbers depend on margins, outflow, and pricing policy. SaaS with high prices and low churn can produce LTV/CAC of 10-15x - this is not a mistake, it is a good business model.
Counter-intuitive point: for e-com the normal LTV/CAC is lower than for SaaS - not because e-com is worse, but because the Payback Period is much shorter there. Money is returned faster, so business is less dependent on a high ratio.
5. Payback Period: when is it more important than LTV/CAC
Payback Period — the number of months until the attracted client pays off.
Formula: Payback Period = CAC / average monthly gross margin per client
Why gross margin and not revenue? Because you pay back CAC not from revenue, but from profit after cost. If the margin is 40%, and the ARPU is 3000 ₽, for the calculation we take 1200 ₽/month, and not 3000 ₽.
When Payback Period is more important than LTV/CAC:
With limited cache. LTV/CAC 5× sounds great, but if the Payback Period is 30 months, and the company has 12 months of cash, it won’t live to see payback. Investors and banks look at Payback Period for this very reason.
In markets with high uncertainty. If the niche is changing quickly, LTV after 3 years is a hypothesis. Payback Period 6 months is a fact that has already been implemented in cohorts.
When scaling. A growing business spends on attracting customers now and receives money later. The shorter the Payback Period, the smaller the “hole” in working capital during growth.
For SaaS, the norm is a Payback Period of up to 18 months. Less than 12 is good, less than 6 is excellent. More than 24 is a dangerous zone even with good LTV/CAC.
6. ARPU/ARPPU: monetization signals
ARPU (Average Revenue Per User) - average revenue per user for a period (usually a month or a year).
Formula: ARPU = revenue for the period / number of active users for the period
ARPPU (Average Revenue Per Paying User) is the same, but only for paying users. The difference between ARPU and ARPPU shows the conversion to payers and the effectiveness of monetization.
Example: mobile application, 100,000 active users (DAU/MAU), of which 8,000 are paying, revenue 2,400,000 ₽/month. ARPU = 24 ₽, ARPPU = 300 ₽. Conversion to paying people is 8%.
How to read these numbers:
- Low ARPU + high MAU = model is built on volume. You need either scale or ARPPU growth through upselling.
- High ARPPU + low conversion in paying users = problem with the monetization barrier. Try trial, freemium, lowering the first payment threshold.
- Falling ARPU with growing users = attracting a less solvent segment. Signal to reconsider targeting.
ARPU is also included in the Payback Period formula: CAC / (ARPU × margin). An increase in ARPU directly reduces the Payback Period without changing the CAC.
7. How the unit economy changes budget decisions
Four specific scenarios:
Scenario 1: two channels with different CAC, similar volume. Yandex Direct gives CAC 4500 ₽, VK Ads - 2800 ₽. With an LTV of RUB 12,000, both have a positive LTV/CAC (2.7× and 4.3×, respectively). But VK is much better. The correct decision is to redistribute the budget 70/30 in favor of VK, not 50/50.
Scenario 2: good LTV/CAC, but bad Payback Period. SaaS, CAC 60,000 ₽, LTV 300,000 ₽ (LTV/CAC = 5×, excellent). But the average monthly gross margin per client is 2500 ₽. Payback Period = 60,000 / 2500 = 24 months. If there is a cash gap after 8 months, the channel must be cut, despite the beautiful ratio.
Scenario 3: low CAC but high churn. You attract 800 ₽, but 40% of clients leave after a month. Real LTV - 3-4 purchases instead of the expected 8. LTV/CAC looked like 6x, in reality - 2.5x. The problem is not the acquisition, but the product or expectations when selling.
Scenario 4: an upsell changes the whole picture. E-com with CAC 2500 ₽ and basic LTV 5000 ₽ (LTV/CAC 2× - alarm zone). They launch an upsell for the second order: a 15% discount on the next purchase within 30 days. Conversion to repeat purchase increases from 20% to 38%. LTV becomes 7800 ₽ (LTV/CAC 3.1×) - we move into the green zone without reducing CAC.
All budget decisions are ultimately about ensuring that CAC remains below LTV/3, and the Payback Period does not go beyond the company’s cash horizon. Read more about budget redistribution in articles about performance marketing in the Russian Federation.
| Situation | LTV/CAC | Payback Period | Solution |
|---|---|---|---|
| Healthy growth | >3× | <12 months | Scaling the budget |
| Okay, but be careful | 3–5× | 12–18 months | Scalable with cache control |
| Alarm zone | 1–3× | 18–24 months | Optimizing CAC or increasing LTV |
| High ratio, bad cache | >3× | >24 months | Looking for funding or cutting back? |
| Burning money | <1× | Doesn't pay off | Stop the channel immediately |
8. What to do right now
Unit economics is not a one-time calculation. This is a model that needs to be updated every quarter: CAC (advertising auction), LTV (customer behavior), and margins (cost, tariffs) change. Companies that count once a year make decisions based on outdated data.
The minimum stack for a normal account: CRM with purchase history, cohort report (Google Sheets will do), and division of marketing expenses by item - budget, salaries, tools. It takes two days of setup and then works itself.
Related materials: unit-economy calculator CM1/CM2, LTV/CAC calculator, performance marketing RF 2026, LTV/CAC benchmarks for SaaS.
If you want to analyze your specific situation, write to Telegram or through form. Starting consultation - 0 ₽.