Say a DTC brand spends $40 per acquisition. It cannot tell whether that is profitable until it calculates what a customer is worth. The customer lifetime value formula answers this in one number: the total revenue (or profit) you'll earn from an average customer over your entire relationship. Three formulas exist, each with different data requirements and accuracy levels. Here is how to calculate CLV using each one. The worked examples use made-up figures, and one shows each step.
Why CLV Matters for Acquisition Spend
Many growth decisions depend on one question: how much can we spend to acquire a customer? Without CLV, the answer is a guess. With CLV, it is closer to arithmetic. If $340 is the profit an average customer brings over their life, the stated 3:1 ratio sets a limit on what you can spend to get a customer. If $340 is sales, it does not show that you will make a profit. If the figure is $80, a $113 acquisition cost is more than the customer brings in.
To forecast what your own customers are worth from here, straight from a transaction file, try the free customer lifetime value tool.
CLV also exposes which customer segments deserve investment and which are quietly draining resources. In a made-up case, a store might find that customers acquired through Instagram have a CLV of $210 while Google Shopping customers average $85. Same product, same price, different lifetime economics. Without customer lifetime value analysis, you'd treat them identically.
The metric goes by multiple abbreviations: CLV, LTV, and CLTV all refer to the same thing. The formulas below work regardless of which label your team uses.
The Simple CLV Formula
The simplest customer lifetime value formula requires three numbers you can pull from any order export:
CLV = Average Order Value × Purchase Frequency × Average Customer LifespanAverage Order Value (AOV) is total revenue divided by total orders. Purchase Frequency is total orders divided by unique customers over a period (typically 12 months). Average Customer Lifespan is how long a customer remains active, measured in the same time unit as purchase frequency.
Worked Example: Simulated data, seed 2407
Question: What CLV does this made-up order data give?
Inputs: Simulated data, seed 2407: 1,000 simulated customers, with orders and order values for one year drawn at random in the script. The two-year customer life is set by hand in the script, and the year's order rate is assumed to hold for both years.
Method: Find the average order value, then the average number of orders per buyer who ordered in the year. Multiply both by the set life of two years.
import numpy as np
seed = 2407
rng = np.random.default_rng(seed)
customer_count = 1000
orders_per_customer = rng.poisson(2.5, size=customer_count)
active_counts = orders_per_customer[orders_per_customer > 0]
order_values = rng.gamma(shape=2.5, scale=24.0, size=int(active_counts.sum()))
revenue = float(order_values.sum())
order_count = int(active_counts.sum())
active_customers = int(active_counts.size)
aov = revenue / order_count
frequency = order_count / active_customers
lifespan_years = 2.0 # set by hand, not estimated from the data
clv = aov * frequency * lifespan_years
print(f"Simulated data, seed {seed}")
print(f"Simulated customers: {customer_count}")
print(f"Customers with no order in the year: {customer_count - active_customers}")
print(f"Active customers: {active_customers}")
print(f"Orders in one observed year: {order_count}")
print(f"Revenue in that year: ${revenue:.2f}")
print(f"Average order value: ${aov:.2f}")
print(f"Orders per active customer in that year: {frequency:.3f}")
print(f"Assumed average customer life: {lifespan_years:.1f} years")
print(f"Simple CLV estimate: ${clv:.2f} per active customer")Printed output:
Simulated data, seed 2407
Simulated customers: 1000
Customers with no order in the year: 77
Active customers: 923
Orders in one observed year: 2574
Revenue in that year: $155105.37
Average order value: $60.26
Orders per active customer in that year: 2.789
Assumed average customer life: 2.0 years
Simple CLV estimate: $336.09 per active customerResult: Simple CLV is $336.09 per buyer who ordered in the year ($60.26 x 2.789 x 2 years). 77 of the 1,000 simulated customers had no order and are left out, so the figure is for active buyers only.
Limit: The two-year life is a guess, not measured from the data. The figure is revenue, and leaves out costs and the lower value of cash due later.
Sources
- Fader P, Hardie B, Lee K (2005), "Counting Your Customers" the Easy Way, Marketing Science, doi 10.1287/mksc.1040.0098. Cited only for who published the BG/NBD model and when.
When to Use the Simple Formula
Use this formula for quick back-of-envelope calculations, board presentations, and setting acquisition budget ceilings. It gives a rough figure but doesn't account for discount rates, varying purchase patterns, or the probability that a customer has permanently churned versus being temporarily inactive.
The Historical CLV Formula
The historical CLV formula calculates the actual value of each customer based on past transactions: no predictions, no averages, just what each customer has spent to date.
Historical CLV = Σ (Transaction Value × Gross Margin) for all transactions by customerIf you want profit-based CLV rather than revenue-based, multiply each transaction by your gross margin percentage. For revenue-based CLV, simply sum all transaction values.
Worked Example: SaaS Company
A made-up SaaS company tracks three customer segments by spend. These figures are invented to show the arithmetic:
| Segment | Avg Monthly Spend | Avg Tenure (months) | Historical CLV |
|---|---|---|---|
| Starter | $29 | 4.2 | $122 |
| Professional | $99 | 14.8 | $1,465 |
| Enterprise | $499 | 28.3 | $14,122 |
In this example, Enterprise customers are worth about 116 times as much as Starter customers, not about 17 times (which is what the monthly price ratio would suggest). The lifespan difference widens the gap. If your acquisition cost is the same for both segments, a gap like this could change where you put sales and marketing effort.
The Predictive CLV Formula (BG/NBD Model)
The simple and historical formulas look backward. The predictive customer lifetime value formula looks forward: how much will this customer spend in the future? This is where statistical modeling enters the picture.
The BG/NBD (Beta-Geometric/Negative Binomial Distribution) model can help predict CLV in non-contractual settings: businesses where customers can purchase at any time without a subscription (e-commerce, retail, marketplaces) (Fader, Hardie and Lee, 2005). It estimates two things simultaneously:
1. Is this customer still "alive"? A customer who bought 5 times in their first month then nothing for 6 months may have churned. A customer who buys once per quarter and last bought 2 months ago may well still be active. The model calculates a probability of being alive for each customer.
2. How many future purchases will they make? Given their historical pattern (recency, frequency), the model predicts expected transactions over a chosen horizon (30 days, 90 days, 12 months).
The inputs are minimal. You only need three columns from your orders data:
Required columns:
- Customer ID
- Transaction Date
- Transaction AmountFrom these three columns, the model derives recency (time since last purchase), frequency (number of repeat purchases), and monetary value (average transaction amount), the RFM variables that drive the prediction.
Worked Example: E-commerce Store with Repeat Buyers
An example online store has 18 months of order data (14,200 orders, 4,800 customers). These are made-up figures. A BG/NBD-based model could produce a table like this:
| Segment | Customers | P(Alive) | Predicted 12mo CLV |
|---|---|---|---|
| Champions | 380 | 94% | $842 |
| Loyal | 720 | 78% | $395 |
| At Risk | 1,100 | 42% | $118 |
| Lost | 2,600 | 8% | $14 |
In this made-up table, the 380 Champions represent 8% of customers but 42% of predicted future revenue. The model gives the 2,600 Lost customers a low chance of being active. Weigh the gain from keeping them against the cost before you set retention spend. The 1,100 At-Risk customers are the intervention opportunity: they have purchase history but are drifting away.
This is the level of insight that separates the simple CLV formula from the predictive one. The simple formula gives one average figure, such as $336.09 in the example above. A predictive model can show which customers might be worth $842 and which $14, in this made-up table.
CLV vs CAC: The Ratio That Determines Profitability
Customer lifetime value only becomes actionable when paired with Customer Acquisition Cost (CAC). The CLV:CAC ratio is one useful health check for a customer-driven business.
CLV:CAC Ratio = Customer Lifetime Value ÷ Customer Acquisition Cost| Ratio | What It Means | Action |
|---|---|---|
| < 1:1 | Losing money on every customer | Fix urgently: cut CAC or increase LTV |
| 1:1 to 3:1 | Thin margin if CLV is profit | Improve retention or raise prices |
| 3:1 to 5:1 | A common planning range | Check profit and payback time before scaling |
| > 5:1 | May be underinvesting in growth | Test more acquisition spend |
Calculate this ratio per acquisition channel, not just as a company average. For example, your email list might have a 6:1 ratio while paid social runs at 1.5:1 (made-up figures). Channel-level CLV:CAC analysis can show where to test a budget shift.
Common CLV Calculation Mistakes
Using revenue instead of gross profit. A $100 order with 30% margins is worth $30 in CLV, not $100. Revenue-based CLV overstates actual value and leads to overspending on acquisition. Use gross margin for profit-based CLV whenever possible.
Ignoring the discount rate. $100 received today is worth more than $100 received in three years. If you lower the value of cash due later, state the rate you use and show its effect on the result. This matters most for long-lifespan businesses like SaaS or financial services.
Averaging across all customers. CLV often varies a lot between customers, and some customers may generate much more value than others. Company-wide averages can hide this. Always calculate CLV by segment: acquisition channel, product category, customer tier, or cohort.
Confusing alive and dead customers. A customer who bought once 18 months ago may not be coming back. Including them in "average lifespan" calculations inflates the number. The BG/NBD model handles this automatically by estimating probability of being alive, but the simple formula can't distinguish.
Not recalculating regularly. CLV changes as your product, pricing, and customer mix evolve. A formula calculated last year with different pricing is stale. Recalculate on a regular schedule.
Which CLV Formula Should You Use?
| Formula | Best For | Data Needed | Main limit |
|---|---|---|---|
| Simple | Quick estimates, board decks | AOV, frequency, lifespan | Rough; the life is a guess |
| Historical | Backward-looking segment analysis | Full transaction history | Past spend only |
| Predictive (BG/NBD) | Future value, individual targeting | Customer ID, date, amount | Needs a fitted model; check it against later buying |
Start with the simple formula if you've never calculated CLV. It gives you a quick baseline. Move to historical CLV when you want segment-level analysis. Graduate to predictive CLV when you need per-customer scores for targeting, retention, and budget allocation. MCP Analytics runs a BG/NBD predictive model from a simple CSV upload.
Working on this for your business? See the customer lifetime value analysis.
Your turn
Bring your own data and the question you actually need answered.
MCP AnalyticsData science by email Send us your data and your question, and we will reply. ds@mcpanalytics.ai
Frequently Asked Questions
What is the simplest customer lifetime value formula?
The simplest CLV formula is: CLV = Average Order Value x Purchase Frequency x Average Customer Lifespan. For example, if a customer spends $50 per order, buys 4 times per year, and stays for 3 years: CLV = $50 x 4 x 3 = $600. This gives a quick estimate but doesn't account for discount rates or varying purchase patterns.
What is the difference between CLV and LTV?
CLV (Customer Lifetime Value) and LTV (Lifetime Value) are the same metric, different abbreviations used interchangeably. Some companies use CLTV as a third abbreviation. All refer to the total revenue or profit a business expects from a single customer over their entire relationship.
How do I calculate CLV for a subscription business?
For subscription businesses, CLV = Average Monthly Revenue per Customer / Monthly Churn Rate. If your average subscriber pays $49/month and your monthly churn rate is 5%: CLV = $49 / 0.05 = $980. To account for the time value of money, use: CLV = ARPU / (Churn Rate + Discount Rate).
What is a good CLV to CAC ratio?
A CLV:CAC ratio of 3:1 is a common planning guess, not a rule for all firms. Ratios below 1:1 mean each customer costs more to acquire than the value counted for them. Check profit, cost, and time to earn the cost back.
How often should I recalculate customer lifetime value?
Recalculate CLV on a regular schedule that fits how fast your business changes. Recalculate after major pricing changes, new product launches, or shifts in customer acquisition channels, as these events change the underlying purchase patterns that drive CLV.