How to Calculate Customer Lifetime Value (LTV): A Step-by-Step Setup Guide

Most brands set their ad budget by looking at one number: what the first order brought in. That’s a risky way to run acquisition. A customer who loses you money on day one can be your most profitable by month twelve, and a customer who looks great on day one may never come back.

Knowing how to calculate customer lifetime value fixes this. It tells you what a customer is actually worth over time, which tells you how much you can afford to pay to win one. Get it wrong in one direction and you overspend on customers who never repeat. Get it wrong in the other and you cut channels that quietly bring in your best buyers.

This guide is written for someone setting up LTV for the first time. You’ll learn the formulas, the data you need, four calculation methods from simple to advanced, and a nine-step build in Google Sheets. One thing to know up front: every worked example uses a hypothetical Indian skincare brand with made-up numbers, because real brands don’t publish their LTV. The math is real. The figures are not.

What is Customer Lifetime Value?

Customer lifetime value (LTV) is the total value a business can expect from one customer across the entire time that customer keeps buying. Most basic definitions count that value as the total revenue the customer generates. You’ll also see it written as CLV. Same thing, different abbreviation.

There are two versions, and mixing them up causes most of the damage. Revenue LTV is the textbook version. It counts everything the customer pays you. Profit LTV counts only what’s left after the cost of the products you sold. Take a customer who spends ₹6,000 over their lifetime on products with a 55% gross margin. Their revenue LTV is ₹6,000. Their profit LTV is ₹3,300.

Honestly, profit LTV is the only version worth running a business on. Revenue LTV flatters you, and it makes every acquisition channel look better than it is.

So why does any of this matter? Because LTV sets the ceiling on what you can spend to acquire a customer. It shows which channels bring buyers who stay. And it turns retention from a vague goal into something you can measure. The research behind retention is old but still widely cited: Reichheld and Sasser’s 1990 Harvard Business Review work found that a 5% improvement in customer retention can lift profits by 25% to 95%.

If you already track blended efficiency, LTV is the other half of the picture.

Customer lifetime value (LTV) is the total value a business can expect from one customer over the full time that customer buys from it, most simply measured as the total revenue that customer generates. Profit LTV subtracts the cost of goods sold from that revenue, and only the profit version should guide how much you spend to acquire customers. LTV sets the ceiling on affordable customer acquisition cost.

The Terms You Need Before the Math

Seven terms show up in almost every LTV conversation. Learn them once and the formulas stop looking intimidating.

• Average order value (AOV) is total revenue divided by the number of orders. If 100 orders brought in ₹1,20,000, your AOV is ₹1,200.
• Purchase frequency is how many orders a typical customer places in a set period, usually a year.
• Customer lifespan is how long a customer keeps buying from you, measured in months or years.
• Churn rate is the share of customers who stop buying in a given period. A 25% annual churn rate means one in four customers leaves each year.
• Gross margin is the share of revenue left after the direct cost of the products sold. Sell something for ₹1,200 that cost you ₹540 and your gross margin is 55%.
• Customer acquisition cost (CAC) is the total money spent winning new customers in a period, divided by the number of new customers in that period.
• Cohort is a group of customers who made their first purchase in the same month.

Everything else in this guide is built from these seven.

What Data Do You Need to Track LTV?

You need six pieces of information for every order, and most stores already have most of them. The one that’s usually missing is cost of goods.

Field	What it is	Where it usually lives
Customer ID	One identifier that stays the same across all of a person’s orders	Your store’s customer or order export
Order date	When the order was placed	Order export
Order value	What the customer paid, after discounts	Order export
Refund or return flag	Whether the order was cancelled, refunded or returned	Order export, payment gateway
Cost of goods	What the products in that order cost you	Inventory or accounting sheet
Acquisition source	The channel that brought the customer’s first order	UTM data, order notes, or your analytics tool

The field that causes the most trouble is the first one. Guest checkouts, two email addresses, a new phone number: suddenly one person looks like three customers, and your repeat rate drops for no real reason. Pick one identifier and merge duplicates on it. In India, phone numbers are often more consistent than email addresses at checkout, though that depends on your store.

Then strip out orders that shouldn’t count. Cancelled orders, refunds and RTO orders all come out. RTO, short for return to origin, means a shipment that was refused at the doorstep and sent back to you. Leaving those in is a common way to overstate LTV in a cash-on-delivery market.

How much history do you need? Twelve months is the comfortable target. If you have less, cohort tables still work, and you just read the early months. Predictive models are pickier, and we’ll get to those.

If cost of goods is your gap, start with one blended margin percentage and refine it later.

How Do You Calculate Customer Lifetime Value? Four Methods, Simplest to Most Advanced

The simplest formula multiplies average order value by purchase frequency by customer lifespan. Use it for a first estimate. For any decision involving real ad spend, move to the margin-adjusted version, and then to cohort data as soon as you have it.

Method 1: The Simple Formula

The basic customer lifetime value formula looks like this:

LTV = AOV × purchase frequency × customer lifespan

Take the hypothetical skincare brand. AOV is ₹1,200. A typical customer orders 2.5 times a year and stays for 2 years. That gives ₹1,200 × 2.5 × 2 = ₹6,000.

It’s quick. It’s also crude. The formula treats every customer as identical, and the lifespan number is usually a guess. Use it as a sanity check, not as the number you build a budget on.

Method 2: Margin-Adjusted LTV

This is the version most brands should actually use. Multiply the simple result by your gross margin.

Profit LTV = AOV × purchase frequency × customer lifespan × gross margin

At a 55% margin, the skincare brand’s ₹6,000 becomes ₹3,300. Same customer, but a number you can compare against what you paid to acquire them.

Subscription businesses can use a shortcut. Divide the profit a customer generates each period by the churn rate:

Profit LTV = (average revenue per account × gross margin) ÷ churn rate

Average revenue per account (ARPA) is the revenue you earn from one paying account in a period. Say a hypothetical subscription app charges ₹499 a month, has an 80% gross margin and loses 5% of subscribers each month. LTV is ₹499 × 0.80 ÷ 0.05 = ₹7,984, which implies an average lifespan of 1 ÷ 0.05 = 20 months.

That “1 ÷ churn” step is worth remembering. A 20% annual churn rate implies a five-year lifespan. At 25% it’s four years. At roughly 33% it’s three. Small changes in churn move LTV a lot.

One caveat. These formulas assume churn stays constant, and it rarely does. New customers often leave faster than established ones, so treat the result as a rough estimate until cohort data confirms it.

There’s also a difference between subscriptions and ecommerce that trips up beginners. In a subscription, churn is visible because people cancel. In ecommerce, customers just go quiet, and you have to decide when quiet becomes lost. A common rule of thumb is no order in roughly twice your typical gap between purchases.

Method 3: Cohort-Based LTV

Methods 1 and 2 estimate. Cohorts measure.

A cohort is a group of customers who made their first purchase in the same month. Instead of guessing a lifespan, you track what each group actually spends over the following months. Say 1,000 customers made their first purchase in January. Here’s what the hypothetical brand’s cohort looks like:

Months since first order	Cumulative revenue per customer	Cumulative profit per customer (55% margin)
0	₹1,200	₹660
3	₹1,500	₹825
6	₹1,800	₹990
12	₹2,400	₹1,320

The January cohort’s 12-month profit LTV is ₹1,320 per customer. Across all 1,000 customers, that’s ₹13.2 lakh of gross profit in the first year. Nobody guessed a lifespan. You counted what customers actually did.

You don’t have to wait forever to use this. Pick a horizon, 12 months is common, and compare every cohort at the same age. A January cohort at month 6 should be compared with a March cohort at month 6, not with its own month 12.

Shopify’s Customer cohort analysis report shows this kind of view. According to Shopify’s Help Center, it groups customers into cohorts by the date of their first order and lets you choose metrics such as customer retention rate, average order value and amount spent per customer. The page doesn’t state which plans include it, so check yours.

Method 4: Predictive LTV

Predictive LTV uses statistical models to forecast how much each customer will still spend, instead of waiting for it to happen. In plain English, the model answers two questions for every customer: how many more orders will this person probably place, and how much will each one be worth?

The most common setup pairs two models. BG/NBD, introduced by Fader, Hardie and Lee in 2005, estimates purchase frequency and the probability that a customer is still active. It’s a lighter successor to the older Pareto/NBD model from 1987. Gamma-Gamma then estimates average spend per order. Combine the two and you get a forecast for each customer.

You need a per-customer log of order dates and amounts, plus at least 12 months of history so seasonality doesn’t distort the fit, according to Perspective AI’s guide to predictive CLV. Without margin data, the output is revenue, not profit. If you want to see the mechanics, Twilio Segment publishes a walkthrough that uses SQL and a spreadsheet, built on frequency, recency, tenure and average transaction size, with future revenue discounted at 10%.

Predictive models also mislead, and the failure modes are worth knowing. History from a discount you no longer run makes lookalike customers look more valuable than they are. A model can’t tell a frustrated customer from one who simply hasn’t reordered yet. And it can’t see why behaviour changes.

Here’s the practical call: skip this until you’ve run cohorts for a few quarters. A sophisticated model on messy data gives you a wrong answer with decimals.

Which Method Should You Use?

Method	Best for	Data needed	Biggest risk
Simple formula	A quick sanity check	AOV, frequency, a lifespan estimate	Treats all customers as identical, lifespan is a guess
Margin-adjusted	Your first real budget decisions	The above plus gross margin	Still an average, hides segment differences
Cohort	Any brand with order history	Order dates, values and customer IDs	Slow to mature, small cohorts are noisy
Predictive	Brands with 12+ months of clean history	Per-customer order log, ideally with margin	Bad inputs produce confident wrong answers

Cohort-based LTV groups customers by the month of their first purchase and measures the cumulative profit each group actually generates over time, so it replaces guessed lifespans with observed behaviour. Predictive models such as BG/NBD, introduced by Fader, Hardie and Lee in 2005, estimate future purchases from each customer’s history and typically need at least 12 months of transaction data.

How to Set Up LTV Tracking From Scratch

This section shows how to calculate customer lifetime value in Google Sheets, from raw order export to a working cohort table. Plan for an afternoon. Steps 1 to 4 take most of the effort, because cleaning data is the real work.

Step 1: Decide What Counts as a Customer and a Repeat Purchase

Pick one identifier, email or phone number, and use it everywhere. Decide that only delivered, non-refunded orders count. Write both rules on the first tab of your sheet, so the numbers stay consistent when someone else takes over.

Step 2: Export at Least 12 Months of Order Data

Pull one row per order with the customer ID, order date, order value after discounts, discount code if any, refund or return status, and acquisition source. Export it in one go if your store allows. Paste it into a tab called Orders.

Step 3: Clean the Data and Remove Orders That Shouldn’t Count

Delete cancelled orders, refunds and RTO orders. Merge duplicate customers on your chosen identifier. Remove test orders and internal orders. It’s dull work, and it’s the step that decides whether your number is trustworthy.

Step 4: Add Gross Profit to Every Order

Create a column called order_profit and calculate it as order value × gross margin percentage. If margins differ by product, use each product’s cost of goods instead. If you don’t have product-level costs yet, one blended margin is fine to start.

Some teams also subtract shipping and payment-gateway fees. Whichever definition you choose, keep it the same every month.

Step 5: Tag Every Customer With a Cohort Month and Acquisition Channel

Create a second tab called Customers with one row per customer. Column A holds the customer ID. Column B holds the first order date:

=MINIFS(Orders!B:B, Orders!A:A, A2)

Column C holds the cohort month, which is just the first day of that month:

=DATE(YEAR(B2), MONTH(B2), 1)

Column D holds the acquisition channel from the customer’s first order. Then go back to the Orders tab and add two helper columns. Column D looks up each order’s cohort month:

=VLOOKUP(A2, Customers!A:C, 3, FALSE)

Column E counts how many months after the first order this order happened:

=(YEAR(B2)-YEAR(D2))*12+MONTH(B2)-MONTH(D2)

Step 6: Build the Cohort Table

Create a third tab called Cohorts. List cohort months down column A. Across row 1, put the months-since-first-order you care about: 0, 3, 6, 12. In cell B2, enter:

=SUMIFS(Orders!$C:$C, Orders!$D:$D, $A2, Orders!$E:$E, "<="&B$1) / COUNTIFS(Customers!$C:$C, $A2)

That adds up all the profit a cohort has generated up to each month and divides it by the number of customers in the cohort. Copy the formula across and down. Each cell is now the cumulative profit per customer for that cohort at that age.

Step 7: Read the LTV Curve and Pick a Horizon

Look along a row. The numbers climb quickly after the first order, then flatten. Where it flattens tells you how long customers keep adding value. Choose one horizon, 12 months is common, and use it for every report so comparisons stay fair. That figure is your LTV.

Step 8: Split the Table by Channel and First-Order Type

Filter the Customers tab by acquisition channel and repeat the cohort table for each one. Then split again by whether the first order used a discount code. You’ll almost certainly find that some channels and some first-order types are worth far more than others.

Step 9: Set a Review Rhythm

Refresh the sheet monthly. Copy the finished values into an archive tab each time, so you can see whether newer cohorts are better or worse than older ones. Review budget decisions quarterly.

That’s it. You now have a measured, profit-based LTV by cohort, which beats any formula-based guess.

What is a Good LTV:CAC Ratio?

The LTV:CAC ratio compares what a customer is worth with what it cost to acquire them. For SaaS, the common rule of thumb is about 3:1. Ecommerce brands with modest repeat purchasing often run lower, roughly 1.5:1 to 3:1, according to Racklify’s 2026 guide to LTV:CAC benchmarks.

Here’s what that looks like with real arithmetic. Say the skincare brand spends ₹900 to acquire each new customer. Its 12-month profit LTV is ₹1,320. The ratio is ₹1,320 ÷ ₹900, or roughly 1.5:1. That sits at the low end of the ecommerce range.

Now look at the payback period, which is how long it takes a customer’s profit to cover what you paid to acquire them. On the first order, the brand earns ₹660 in profit against ₹900 of CAC, so it’s ₹240 short. The cohort’s cumulative profit passes ₹900 somewhere between month 3 (₹825) and month 6 (₹990), around month 4 or 5 if repeat orders arrive steadily. After 12 months, the 1,000-customer cohort has produced ₹13.2 lakh of gross profit against ₹9 lakh of acquisition spend.

Both numbers matter. A ratio tells you whether a customer is worth acquiring. Payback tells you whether you can survive long enough to collect. Racklify’s guide suggests SaaS companies in high-growth mode prefer payback under 12 months, while enterprise SaaS with multi-year contracts can tolerate 18 to 36 months. Ecommerce has no neat rule, but months, not years, is the realistic target for most brands with limited cash.

You can also flip the ratio into a spending ceiling. Divide your LTV by the ratio you want:

Target LTV:CAC	Maximum CAC at ₹1,320 LTV
3:1	₹440
2:1	₹660
1.5:1	₹880

At a ₹900 CAC, the brand is already above even the loosest ceiling. That doesn’t mean it’s doomed. It means the next job is either lifting LTV through repeat purchases or lowering CAC, and now you can see which.

One more thing. LTV:CAC and ROAS answer different questions. ROAS asks whether an ad paid for itself on the first purchase. LTV:CAC asks whether the customer will pay back what you spent over time. [INTERNAL LINK: ROAS vs MER and customer value → performance marketing metrics guide]

Treat every benchmark here as a starting point. These figures come from industry guides without a named primary study behind them, and your margins and cash position matter more than any rule of thumb.

CAC compares the profit a customer generates with what it cost to acquire them. Racklify’s 2026 benchmark guide puts the SaaS rule of thumb near 3:1 and ecommerce at 1.5:1 to 3:1 where repeat purchasing is modest, but the ratio only tells half the story without CAC payback time, which shows how long it takes cohort profit to cover acquisition cost.

How Should You Segment LTV?

Blended LTV hides more than it shows.

Picture a brand whose 12-month profit LTV is ₹1,200 across all customers. That looks healthy. Split it by how customers arrived and the picture changes: customers who came through the email list are worth ₹1,800 each, and customers who came through a discount code are worth ₹600. A 50/50 mix gives you the blended ₹1,200. Shift the mix to 30% email and 70% discount code and the blended number drops to ₹960, without anything obviously going wrong in your dashboard. These numbers are hypothetical, but the pattern is real.

Start with these splits:

• Acquisition channel, such as paid social, search, email, organic, marketplace or referral.
• First product bought, since customers who start with a hero product often behave differently from those who start with a low-margin add-on.
• First order type, full price versus discount code or sale.
• Geography, metro versus smaller cities, where delivery costs and return rates can differ.
• Payment mode, prepaid versus cash on delivery, which is worth testing in India because RTO rates can differ by mode.

One caution. Each segment needs enough customers behind it. Small cohorts swing wildly from month to month, so treat very small segments as directional, not as proof.

The real question isn’t whether your LTV is good. It’s which channel deserves your next rupee.

Common LTV Mistakes That Make the Number Useless

Most LTV errors push the number up, which is exactly why they survive. Nobody questions a flattering result.

Guessing a lifespan that’s too long. Assume customers stay five years when they really stay two, and you’ve overstated LTV by 2.5 times. Incendium’s analysis of LTV pitfalls lists this as one of five common errors, and suggests using the median lifespan from historical cohorts instead of an optimistic guess.

Using revenue instead of profit. A customer spending ₹500 on products with a 60% margin is worth more than one spending ₹800 on products with a 25% margin. Revenue LTV ranks them the wrong way round.

Ignoring one-time buyers. In many ecommerce brands a large share of customers buy once, and a smaller loyal group pulls the average up. Look at the spread, not just the mean.

Reporting one blended number. We covered this above. A single LTV hides the channels that are quietly losing you money.

Assuming the past repeats. Customers you acquire at scale through paid ads often behave differently from the early ones who found you through word of mouth. Historical LTV is a guide, not a promise.

Counting orders that never completed. Returns, cancellations and RTO orders belong outside your LTV data. In a cash-on-delivery market, this one error can distort the whole picture.

Mixing horizons. Comparing one channel’s 6-month LTV with another’s 12-month LTV makes the older cohort look better for no real reason. Fix the horizon and keep it fixed.

The most common LTV errors are assuming a longer customer lifespan than cohort data supports, using revenue instead of gross profit, ignoring churn and one-time buyers, reporting one blended number instead of segments, and counting returned or RTO orders as revenue. Overstating lifespan alone is costly: assuming five years when customers really stay two inflates LTV by 2.5 times.

How Do You Increase LTV Once You’re Tracking It?

You raise LTV by moving one of four levers: order value, purchase frequency, customer lifespan or margin. The formula you’ve been using shows you exactly which one.

For most brands, the cheapest win sits in the gap between the first and second order. According to Omnisend data cited by LTV.ai, the likelihood that a customer buys again rises from 27% after a first purchase to 49% after a second and 62% after a third. Once someone has bought twice, they’re far more likely to keep going.

That points to a few practical moves. A post-purchase email flow that arrives while the product is still in use. Replenishment reminders timed to when a typical customer runs out. Subscription options for consumables. Bundles that lift order value. And less reliance on blanket discounting, which protects margin.

Which Tools Should You Use to Track LTV?

Start in a spreadsheet. Move up when the spreadsheet starts hurting.

• Google Sheets or Excel handles Methods 1 to 3 comfortably, as the setup steps above show. The cost is manual refreshing.
• Shopify’s Customer cohort analysis report gives you cohort views of retention rate, average order value and amount spent per customer, plus details such as top marketing channels. The closest metric to LTV is amount spent per customer, which is sales, not profit, so apply your margin on top.
• GA4’s User lifetime exploration shows how users behaved across their lifetime, including lifetime revenue and engagement, and adds predictive metrics for purchase and churn probability, according to Google Analytics Help. Lifetime data is available only for users active after August 15, 2020, and because it reports revenue, you’ll need to apply margin yourself.
• A data warehouse and dashboard, such as BigQuery with Looker Studio, make sense once you have large order volumes, multiple data sources or want to run predictive models.

Whatever you use, don’t take your LTV from ad platforms. Their attributed revenue tells you what they claim credit for, not what a customer is worth. To find out whether a campaign genuinely raised LTV, compare the customers it reached against a holdout group that didn’t see it, a test design LTV.ai recommends for measuring incremental lift.

Start With One Cohort Table

Three things matter more than everything else here. Use profit, not revenue. Measure cohorts instead of trusting a guessed lifespan. And judge LTV against CAC and payback time together, because a good ratio that takes two years to collect can still sink a brand with limited cash.

You don’t need a data team to start. This week, export twelve months of orders, clean out the refunds and RTO orders, and build the cohort table from Step 6 for one channel. Even a rough version will tell you something your ad dashboard can’t: what a customer is worth after the first order.

Do that once and every budget conversation changes. You stop asking whether the ad paid for itself today. You start asking whether the customer will pay for themselves by month six.

Setting up the tracker is a one-afternoon job. Deciding what to do with the number is the harder part, and it’s what we work through in the Crystal Clear Newsletter: practical marketing and analytics breakdowns for people who own their numbers. Join if you want the next one in your inbox.

Frquently Asked Questions on How to Calculate Customer Lifetime Value