Lifetime value when the product costs less than a coffee

Finance will release the sampling budget on one condition: tell them what a newly acquired household is worth before you hand out 400,000 free sachets. That number is lifetime valuelifetime valueLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → (LTV), the profit a household delivers across its whole time in the category. In FMCG you have to build it from shelf behaviour, because nothing about the relationship is written down. No account, no renewal date, no address. Just a bottle that costs less than a latte, bought again and again for a decade.

Why subscription LTV math doesn't work here

The SaaS formula is clean:

LTV = (Average Revenue per Account × Gross MarginGross MarginGross margin is the share of revenue left after subtracting the direct cost of producing goods or services, expressed as a percentage of revenue.View full definition → %) / Churn RateChurn RateChurn rate is the percentage of customers or revenue lost over a period. It measures how fast a business loses its existing customer base.View full definition →

It needs one contract, one renewal date and one cancel event. A detergent buyer gives you none of them. She drifts: still buying you on two trips in three, quietly giving a rival the rest, and the churn proxies the retention lesson benchmarks are the closest thing to a cancellation signal you will ever see.

So rebuild LTV from quantities that exist on a shelf:

• Purchase frequency: how often the household buys the category (laundry detergent, roughly every 4 to 8 weeks per US household, estimate)
• Share of requirements: the slice of that household's category purchases you win versus rivals
• Relationship length: how many years the household stays in the category at all
• Repeat rate, taken here as an input at whatever level the retention benchmarks say is realistic for your category

The FMCG LTV formula

LTV = (Annual Category Spend per Household × Share of Requirements × Gross Margin %) × Relationship Length (years)

Inputs (illustrative, US market, estimates):

• Household annual laundry detergent spend: ~$70
• Your share of requirements in a loyal-leaning household: 60%
• Gross margin after trade spend: ~45%
• Relationship length before the household switches away or leaves the category: 8 years

Annual brand revenue per household = $70 × 60% = $42
Annual brand profit per household  = $42 × 45% = $18.90
LTV (8-year horizon, undiscounted) = $18.90 × 8 = $151.20

Against a $6 bottle, the single-purchase view understates value by a factor of twenty-five.

Then discount it. $18.90 arriving in year eight is not worth $18.90 today, and FMCG horizons are long enough for this to bite. At a 10% discount rate, eight years of $18.90 is worth about $100, not $151. A third of the headline figure is the time value of money, and it is the third finance will strike out of your business case.

Why household LTV beats individual LTV

The buying unit is rarely one person, and its composition changes. A shopper loyal to a diaper brand has a horizon capped by how long her child wears diapers, two to three years. The same household, if it stays with the parent company through wipes, then toddler snacks, then kids' toothpaste, is worth a multiple of that.

This is the arithmetic behind a portfolio the size of Unilever's, a few hundred brands across food, ice cream, personal care and home care: the unit being valued is the household's total spend with the company, not the SKU (stock keeping unit). One consequence is that a divestment looks different in this model. Selling a brand removes its own margin and also removes an entry point into households you were counting on for cross-buying later.

It also explains why basket-share data from retailer loyalty cards and consumer panels beats your own shipment numbers. Shipments tell you what you sold; panel data tells you what fraction of the household's category spend you took, and that fraction is the term in the formula you are least able to guess. Kantar's Worldpanel division publishes methodology notes and reports that are the industry reference for these numbers.

When a brand can legitimately buy a loyalty loop

A large LTV licenses aggressive first-purchase spend only if something holds the household in place after the introductory price disappears. Three mechanisms do that without a contract, and each carries a cost.

Hardware. Nespresso put a machine on the counter and made its capsule the default purchase for years afterwards, with the machine priced to spread and the margin sitting in the coffee. The loop loosened once patents on the capsule format expired and compatible capsules from other roasters reached the same shelves. Treat a hardware lock as finite: put an end date in the model instead of assuming the installed base pays forever.

Frequency plus delivery. Yakult sells a small daily dose, so frequency is counted in days, not weeks, and in Japan its home-delivery network of Yakult Ladies has been reordering on the household's behalf since the 1960s. High frequency compresses payback: a household buying weekly repays a $15 acquisition costacquisition costCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → inside a year, which changes what you can afford at trial.

Someone else's replenishment rail. Amazon's Subscribe & Save turns a repurchase decision into a default in exchange for a standing discount. Two costs are easy to leave out of the model: the discount is a permanent reduction in gross margin in every year, not a one-off promotion, and the reorder cadence, the address and the data sit with a retailer that also sells private-label goods in most of these categories. Withdrawing the discount later reads as a price rise, so it is close to a one-way door.

The counter-example is more common than all three: nothing holds at all. When the only reason for the second purchase was the launch price, share of requirements falls back as soon as the deal ends, and the honest model is not eight years of $18.90 but one purchase and a residue. Deal-acquired cohorts carry lower share of requirements than organically acquired ones, so plugging the loyal-leaning 60% into a coupon-acquired cohort inflates its value.

Some households are worth less than nothing. A shopper who buys you only on deep multibuys, after coupon redemption and the trade funding behind the display, can contribute close to zero gross profit per unit; add the sampling that recruited her and the cohort is negative. Averages hide this, so calculate LTV by acquisition source rather than once for the brand.

What this means for CAC and marketing spend

With a defensible household LTV you can set a ceiling on acquisition cost, in the per-incremental-household sense the foundations lesson establishes. $100 of discounted household value makes $15 to recruit a first-time buyer cheap, even though it is several times the margin on one bottle. That is the case for in-store demos, first-purchase discounts and paying a retailer for placement inside its app.

Two ways this goes wrong at scale:

• Judging the campaign on first-purchase return. The first purchase is a cost of acquisition, not a sale that has to pay for itself.
• Counting promoted volume as acquisition. A big multibuy mostly pulls volume forward from households already loyal to you and fills their cupboards. Units rise, incremental households barely move, and the LTV case was built on households.

🎬 [VIDEO: "How Procter & Gamble Thinks About Brand LoyaltyBrand LoyaltyYour customers' propensity to repeatedly purchase from you and resist competitive offers, driven by satisfaction, habit, trust, and switching costs.View full definition →" - youtube.com - search for P&G or Kantar Worldpanel talks on repeat purchase and brand loyalty measurement in CPG, useful for seeing loyalty-card data used in practice]

A quick sanity-check framework

1. What horizon? Diapers give you two to three years, detergent decades.
2. Discounted or not? An undiscounted eight-year LTV overstates value by roughly a third.
3. Panel or guess? Share of requirements comes from a panel (Kantar, Circana, Nielsen) or it is fiction.
4. Household or shopper? Mixing the two pushes per-person numbers up in larger households.
5. Does the margin include trade spend, coupon redemption and slotting? List-price margin flatters every line below it.

Key Takeaways

• FMCG LTV is built from purchase frequency, share of requirements, relationship length and margin after trade spend, not from a churn rate.
• LTV = Annual Category Spend × Share of Requirements × Gross Margin × Years. A $6 detergent bottle sits inside roughly $150 of undiscounted household value over eight years, closer to $100 once discounted at 10%.
• Value the household, not the shopper or the SKU. Cross-buying across a portfolio the breadth of Unilever's outlasts any single category tenure, and divesting a brand costs you an entry point as well as its margin.
• Heavy acquisition spend is only justified when something holds the household after the deal ends: hardware (Nespresso, with an expiry date once compatible capsules arrive), frequency plus delivery (Yakult), or a retailer's replenishment rail (Amazon Subscribe & Save, paid for in permanent margin and in data you no longer own).
• Compute LTV by acquisition source. Deal-acquired cohorts show lower share of requirements and can be gross-margin negative once coupons, trade funding and sampling costs are counted.