Syntha Lab RU

Unit economics with buyout: CM1-CM3, LTV/CAC, customer survival, channel payback

Order and customer mathematics for online channels with returns: contribution per order at CM1, CM2, CM3 levels accounting for buyout, expected customer orders from a survival curve, LTV, CAC, payback and channel comparison. The decision: how much can be paid for a customer and which channels to scale.

Math and analytics Brand and retail Verified: 2026-11-13

Open interactively →

Why this document

The head of online and the finance director read this when deciding how much to pay for a customer and which channels to enter. It matters because apparel and footwear have many returns, and an order with revenue can be loss-making after logistics, return and commission. Contribution at levels CM1, CM2 and CM3 shows at which step profit disappears, and the LTV to CAC ratio by channel separates those that pay back within a year from those that only burn budget. Without this, revenue growth hides the loss.

At a glance

What is this document and why is it needed?
Order and customer mathematics for online channels with returns: contribution per order at CM1, CM2, CM3 levels accounting for buyout, expected customer orders from a survival curve, LTV, CAC, payback and channel comparison. The decision: how much can be paid for a customer and which channels to scale.
Who owns the document and how often is it rebuilt?
E-commerce finance analyst and the head of marketing; recomputed monthly by cohort and channel; the result underlies the marketing budget and scaling decisions.
What mistakes are the most common?
CM is computed on orders rather than kept ones → profit overstated by returns → compute from kept orders with losses. LTV on revenue instead of CM2 → LTV/CAC looks good for a loss-making channel → LTV on contribution only. Average CAC is used → scaling goes through expensive clicks → look at marginal CAC by channel.

For whom

Brand
A brand computes D2C and marketplace unit economics separately: D2C has no commission but has delivery, payment fees and CAC; a marketplace fixes commission and logistics and the customer belongs to the platform, so LTV from repeat orders is lower and counted by brand, not by customer.
Retail
A retailer with online computes delivery, click-and-collect and try-on unit economics and links customers through the loyalty card; offline and online orders of one customer are merged, otherwise LTV and CAC are wrong.

What it consists of

Contribution levels CM1, CM2, CM3

  • CM1 = revenue of kept orders − cost of kept units − write-offs on returns − payment fees (commissions and acquiring): product margin after returns and payment, as in the Lab dictionary. CM2 = CM1 − logistics, return handling, storage and packaging: order contribution after variable fulfillment costs. CM3 = CM2 − marketing (CAC): contribution after acquiring the customer.
  • Example per 100 orders: price RUB 4,600 including VAT (RUB 3,770 excluding VAT), buyout 68%, cost RUB 1500. Revenue RUB 256,393, CM1 of RUB 142,066, CM2 RUB 93,026 or RUB 930 per order; at a CAC of RUB 1,100 the first customer order gives CM3 of −RUB 170.

Buyout and returns

  • Buyout (share of kept orders) is the main lever: at 50% buyout the first-order CM3 is −RUB 690, at 90% RUB 466; the first order repays CAC if buyout is not below 73.9%.
  • Brand: on a marketplace a return means lost commission, logistics and part of the goods; in D2C a return also concerns repackaging and resale. Retailer: in-store try-on lowers online returns, and returning to a store saves reverse logistics.

The customer over time: survival and LTV

  • The share of customers ordering in month t after the first sets a survival curve (illustrative: 100, 22, 14, 11, 9...%). Expected orders over 12 months are 1.92; LTV (CM2 discounted at 1.5% a month) is RUB 1,739 against a CAC of RUB 1,100: LTV/CAC = 1.58.
  • LTV is computed on CM2, not on revenue or CM1; a forecast beyond the actually observed horizon (cohorts younger than 12 months) is built by extrapolating the retention curve (see the cohort document) and marked as an estimate.

Channels and payback

  • A channel is compared by LTV/CAC and payback (the month when the customer cumulative CM2 covers CAC): in the example from 0.66 (Creators) to 5.96 (CRM and email).
  • An average CAC hides differences: paid channels bring new customers while CRM and email bring back existing ones; for a scaling decision use the marginal CAC (the cost of the next customer), not the average.

Brand and retailer: unit economics in practice

  • Brand: on a marketplace a brand sees the order but not the customer: LTV is computed on repeat brand purchases on the platform if the platform provides data, and CAC is replaced by on-platform promotion cost.
  • Retailer: a retailer with a loyalty card sees the customer in all channels: LTV merges stores and online, and the CAC of an online order is compared with the customer lifetime contribution in the chain.
  • Brand: the D2C cost structure: delivery, payment fees, packaging, return handling and an own call center; for a marketplace commission and platform logistics are fixed and depend on size and category.
  • Retailer: the retailer online cost structure: picking in a store or warehouse, delivery, click-and-collect, in-store return; click-and-collect and in-store return are often cheaper than courier delivery.
  • Brand: buyout for a brand depends on the size grid and description: size returns lower CM2, and improving size guidance and photos pays back faster than cutting CAC.
  • Retailer: buyout for a retailer is improved by try-on and in-store return: the share of returns brought to a store lowers reverse logistics and returns the goods to sale faster.
  • Brand: brand channels: search, social, creators, CRM, affiliates, marketplace; they are scaled by marginal CAC and CM2 payback rather than by ROAS on revenue.
  • Retailer: retailer channels: store traffic, online, the loyalty program, mailings; the contribution of online advertising is assessed including in-store purchases (ROPO), otherwise online is undervalued.
  • Brand: a brand shows finance CM1-CM3 by channel and cohort, LTV/CAC with an interval and the break-even buyout for every channel.
  • Retailer: a retailer shows online CM2 including click-and-collect and in-store returns, the contribution of an online customer to offline and the total chain customer LTV.

Where the data comes from

Data sourceFields
Orders and returnsOrders, buyout, returns by reason, delivery both ways, cost, payment fees, packaging; by channel and cohort.
Marketing and attributionSpend by channel, new customers, acquisition sources; attribution rules documented (see the incrementality document).
CRM and loyaltyCustomer identifier, order dates, online-offline merge, communication consents.

How to build it step by step

  1. Step 1. Define the unit: order, customer or cohort; for LTV the unit is the customer, for contribution the order.
  2. Brand: Step 2. Separate D2C and marketplaces by cost structure (commission, platform logistics, payment fees).
  3. Retailer: Step 3. Separate delivery, click-and-collect and in-store returns; merge online and offline customers by card.
  4. Step 4. Compute price excluding VAT, buyout, revenue of kept orders, cost of kept units, return write-offs and payment fees; get CM1.
  5. Step 5. Subtract logistics, return handling, storage and packaging (payment fees are already in CM1); get CM2 per order.
  6. Step 6. Build the customer survival curve by cohort and compute expected orders over the horizon.
  7. Step 7. Compute LTV on CM2 with discounting, CAC by channel, LTV/CAC and payback.
  8. Step 8. Check CM3 sensitivity to buyout, price, CAC and logistics; find the break-even buyout.
  9. Step 9. Rank channels by LTV/CAC and payback, estimate marginal CAC and decide which to scale and which to cut.
  10. Step 10. Fix the rules: what is in CAC, the LTV horizon, the discount rate; publish measures with an interval, not a single number.

Formulas

MeasureFormulasNote
CM1CM1 = N·b·P_net − N·b·C − N·(1 − b)·C·w − k_pay·N·b·P_netN orders, b buyout, C cost, w loss share on a return, k_pay payment fee share.
CM2CM2 = CM1 − N·(L_out + k_pack) − N·(1 − b)·(L_back + h)h return handling; payment fees are already in CM1.
CM3CM3 = CM2 − marketing; per customer CM3 = Σ CM2·orders − CACMarketing is attributed to new customers, not to all orders.
Expected ordersE[orders] = Σ_(t=0..T) S(t), S(0) = 1S(t) is the share of customers ordering in month t (survival curve).
LTVLTV = Σ_(t=0..T) S(t)·CM2_1 / (1 + δ)^tCM2_1 is the order contribution, δ the monthly discount rate; CM2 may vary over time.
CAC paybackt* = min{ t : Σ_(k≤t) S(k)·CM2_1/(1 + δ)^k ≥ CAC }The term in months; term targets come from financial policy, not from the formula.
Break-even buyoutb*: CM2_1(b*) = CACFound by iteration; below b* the first order does not repay acquisition.

Formulas in notation

CM2 per order
cm2 = CM1/N − L_out − k − (1 − b)·(L_back + h)
Variable fulfillment costs per order; payment fees are already in CM1.
CM3 per customer
CM3 = E[orders]·cm2 − CAC
E[orders] is the expected order count.
Payback
t* = min{ t : cm2·Σ_{k≤t} S(k)/(1+δ)^k ≥ CAC }
The first month in which cumulative contribution covers CAC.

Worked example

Contribution per 100 D2C orders: from revenue to CM2 and acquisition channels

Channel / measureCAC, RUB12-month LTV, RUBLTV/CACPayback, months
Search ads1,1001,7391.581
Social1,5001,4780.99no payback in 12 months
Creators2,1001,3910.66no payback in 12 months
CRM and email3502,0875.960
Affiliates9001,2171.353

Contribution per 100 orders at 68% buyout: revenue RUB 256,393, CM1 of RUB 142,066, CM2 RUB 93,026 (RUB 930 per order). 12-month LTV is RUB 1,739 for the base channel; channel repeat factors (a multiplier on the curve) and CAC are illustrative. Channels with LTV/CAC below 1 do not pay back within a year.

Figures

From revenue to CM2 per 100 orders
Waterfall: Revenue of kept units (excl. VAT) +256, COGS of kept units −102, Return write-offs −7.2, Payment fees (CM1) −5.1, Logistics, returns, packaging (CM2) −49, CM2 contribution 93050100150200250300Revenue of kept units (excl. VAT)Revenue of keptunits (excl. VAT)Revenue of kept units (excl. VAT): +256+256COGS of kept unitsCOGS of kept unitsCOGS of kept units: −102−102Return write-offsReturn write-offsReturn write-offs: −7.2−7.2Payment fees (CM1)Payment fees (CM1)Payment fees (CM1): −5.1−5.1Logistics, returns, packaging (CM2)Logistics, returns,packaging (CM2)Logistics, returns, packaging (CM2): −49−49CM2 contributionCM2 contributionCM2 contribution: 9393RUB thousand
  • positive
  • negative
  • total
Data as a table
StepChangeLevel
Revenue of kept units (excl. VAT)+256256
COGS of kept units−102154
Return write-offs−7.2147
Payment fees (CM1)−5.1142
Logistics, returns, packaging (CM2)−4993.1
CM2 contribution—93

From revenue to CM2 per 100 orders. Cost and logistics take the main part; returns add losses and reverse logistics costs.

Cumulative customer contribution and CAC
Line chart: Cumulative CM2 per customer, RUB: 930 → 1,739; CAC, RUB: 1,100 → 1,1008001,0001,2001,4001,6001,80001234567891011paybackCumulative CM2 per customer, RUB, 0: 930Cumulative CM2 per customer, RUB, 1: 1,132Cumulative CM2 per customer, RUB, 2: 1,258Cumulative CM2 per customer, RUB, 3: 1,356Cumulative CM2 per customer, RUB, 4: 1,435Cumulative CM2 per customer, RUB, 5: 1,500Cumulative CM2 per customer, RUB, 6: 1,555Cumulative CM2 per customer, RUB, 7: 1,601Cumulative CM2 per customer, RUB, 8: 1,642Cumulative CM2 per customer, RUB, 9: 1,679Cumulative CM2 per customer, RUB, 10: 1,711Cumulative CM2 per customer, RUB, 11: 1,739CAC, RUB, 0: 1,100CAC, RUB, 1: 1,100CAC, RUB, 2: 1,100CAC, RUB, 3: 1,100CAC, RUB, 4: 1,100CAC, RUB, 5: 1,100CAC, RUB, 6: 1,100CAC, RUB, 7: 1,100CAC, RUB, 8: 1,100CAC, RUB, 9: 1,100CAC, RUB, 10: 1,100CAC, RUB, 11: 1,100₽
  • Cumulative CM2 per customer, RUB
  • CAC, RUB
Data as a table
MeasureCumulative CM2 per customer, RUBCAC, RUB
09301,100
11,1321,100
21,2581,100
31,3561,100
41,4351,100
51,5001,100
61,5551,100
71,6011,100
81,6421,100
91,6791,100
101,7111,100
111,7391,100

Cumulative customer contribution and CAC. The curve crosses the CAC level in the payback month and then grows ever slower as customers drop out. Axis: months after the first order.

LTV/CAC by acquisition channel
Bar chart: Search ads 1.58, Social 0.99, Creators 0.66, CRM and email 5.96, Affiliates 1.350123456Search adsSearch adsLTV/CAC: 1.581.58Threshold 1: 11SocialSocialLTV/CAC: 0.990.99Threshold 1: 11CreatorsCreatorsLTV/CAC: 0.660.66Threshold 1: 11CRM and emailCRM and emailLTV/CAC: 5.965.96Threshold 1: 11AffiliatesAffiliatesLTV/CAC: 1.351.35Threshold 1: 11ratio
  • LTV/CAC
  • Threshold 1
Data as a table
MeasureLTV/CACThreshold 1
Search ads1.581
Social0.991
Creators0.661
CRM and email5.961
Affiliates1.351

LTV/CAC by acquisition channel. CRM and email are best (5.96); creators (0.66) and social (0.99) are below the threshold of 1: they do not pay back within a year.

First-order CM3 and buyout
Line chart: First-order CM3, RUB: −690 → 466−800−600−400−20002004006005060708090First-order CM3, RUB, 50: −690First-order CM3, RUB, 60: −401First-order CM3, RUB, 70: −112First-order CM3, RUB, 80: 177First-order CM3, RUB, 90: 466₽
Data as a table
MeasureFirst-order CM3, RUB
50−690
60−401
70−112
80177
90466

First-order CM3 and buyout. Each 10 pp of buyout gives about RUB 290 per customer; CM3 crosses zero at a buyout near 74%. Axis: buyout, %.

Specifics by category

Apparel
Size spread and try-on give high returns (especially dresses and tops), so buyout is the key figure; repeat orders follow seasons rather than months.
Footwear
The highest returns by size and last and the costliest reverse logistics; in-store try-on and size guidance affect CM2 most; the break-even buyout is higher than for apparel.
Accessories
Returns are low, the ticket is small and the logistics share of price is high: CM2 is sensitive to delivery cost; gift orders repeat once a year, not monthly.
Kids
Sizes change fast so a repeat purchase by the same customer is expected but in another size; size returns are higher, school orders peak; the customer is a parent and the recipient a child.
Men
Buyout is higher and choice on basics quicker: CM2 is steadier; repeat orders on basics are regular so the survival curve is flatter.
Women
A wide choice and fashion items raise returns and the share of “try and return”; CM2 is lower but order frequency is higher, so the key is the retention share, not a single order.

Common mistakes

  • CM is computed on orders rather than kept ones → profit overstated by returns → compute from kept orders with losses.
  • LTV on revenue instead of CM2 → LTV/CAC looks good for a loss-making channel → LTV on contribution only.
  • Average CAC is used → scaling goes through expensive clicks → look at marginal CAC by channel.
  • Payment fees, packaging and return handling are left out → CM2 overstated → the full list of variable fulfillment costs.
  • LTV is forecast from early cohorts as fact → tail error → show the observed and extrapolated parts separately.

Russian specifics

  • Revenue and cost for CM are computed excluding VAT; the 22% VAT from 2026 must not enter contribution, and marketplace agency schemes (commission, deductions) follow the platform documents. Rule: Tax Code Art. 164: Tax Code, Art. 164 (VAT rates).
  • In distance selling the buyer may return a quality item within the set period; the period and rules come from the sales rules and the consumer protection law and affect buyout and the returns provision. Rule: Law 2300-1 of 07.02.1992: Law of 07.02.1992 No. 2300-1 on consumer rights protection (rule check not completed).
  • Customer personal data in LTV and cohort calculations is processed with consent and in anonymized form; a card identifier is not passed to contractors without a contract. Rule: Law 152-FZ of 27.07.2006: Federal Law of 27.07.2006 No. 152-FZ on personal data (rule check not completed).

Open interactively →