Customer Lifetime Value in Singapore: How to Calculate LTV Without Fooling Yourself

Lifetime value is the most useful number in marketing and the easiest to inflate. It is useful because it is the only figure that tells you how much you can afford to pay for a customer — without it, every conversation about whether a campaign is “too expensive” is guesswork. It is easy to inflate because almost every input is a forecast, and forecasts made by the person who wants the budget approved tend to come out well.

We have seen a Singapore e-commerce brand present an LTV of S$840 to justify a cost per acquisition of S$210. The S$840 was revenue, not profit, over an assumed five-year customer life, based on eleven months of trading. On gross margin over an observed horizon the real figure was closer to S$190. They were losing money on every customer and their dashboard said they were winning.

This guide covers how to calculate a lifetime value figure you can actually plan against: which formula suits your business, the Singapore-specific adjustments (GST above all), why the famous 3:1 ratio is weaker evidence than people think, and how to tell an honest LTV from a hopeful one.

What LTV actually means

Customer lifetime value is the total profit a customer generates over the whole time they do business with you. Three words in that sentence do the work.

Profit, not revenue. If your gross margin is 40%, a customer spending S$1,000 is worth S$400 to you, not S$1,000. Using revenue is the single most common error, and it overstates value by exactly the amount of your cost of goods.

Whole time, which means you are forecasting. Any LTV for a business under three years old is an extrapolation, and should be labelled as one.

Customer, which means you need to be able to recognise the same person across purchases. If your systems cannot link a repeat buyer to their first order, you cannot calculate LTV — you can only calculate average order value and hope.

Three formulas, and which one is yours

There is no single LTV formula. There are three families, and picking the wrong one produces confident nonsense.

Our strong recommendation for Singapore SMEs is the cohort method with a fixed horizon: take everyone who became a customer in a given month, and measure the actual gross profit they have generated in the 12 or 24 months since. It is not clever, but it is observed rather than assumed, and it lets you compare January’s customers against June’s to see whether the business is getting better or worse at retention.

The churn-based formula deserves a specific warning. Dividing by churn assumes churn is flat across a customer’s life. It never is — most cohorts churn hardest in the first three months and then flatten out. Applying an early-months churn rate to the whole lifetime understates LTV badly; applying a late-months rate overstates it wildly. If your monthly churn in month two is 12% and in month twelve is 2%, the formula gives you an eight-month lifetime or a fifty-month one depending on which you feed it.

The practical consequence is that a single blended LTV hides two completely different customers: the majority who buy once and vanish, and the minority who stay for years. Averaging them tells you about neither. If you can, calculate LTV separately for first-time buyers and for anyone who has made a second purchase — the second-purchase group is usually worth several times the first, which is why the second purchase is the highest-leverage thing most Singapore retailers can work on.

The Singapore adjustments

Strip GST out. This is the big one. If your prices are GST-inclusive, as almost all Singapore consumer prices are, then 9% of every order total is tax you collect and remit. It was never your revenue and it certainly is not your profit. An LTV built from GST-inclusive order values is overstated by roughly 8.3% before you have made a single other mistake — and worse, it corrupts your comparison against CAC, because the platforms report the same inflated figure. Our guide to ROAS versus ROI works through the same arithmetic on the return-on-ad-spend side.

Use gross margin, and be honest about what is in it. For a Singapore e-commerce business that means cost of goods, inbound shipping, payment gateway fees, last-mile delivery and returns. For a services firm it means the delivery cost of the work — consultants’ time including employer CPF at 17% for staff aged 55 and below. A “60% margin” that ignores delivery labour is not a margin.

Marketplace commissions belong in margin. If a meaningful share of repeat purchases come through Shopee, Lazada or a delivery platform rather than your own store, the platform’s commission comes off every one of those orders. A customer whose repeat purchases migrate to a marketplace has a materially lower LTV than the same customer buying direct — which is a strategic argument for owning the relationship, not just a bookkeeping note.

Do not discount over long horizons unless you have to. Textbook LTV applies a discount rate to future cash flows. For most SME horizons of 12 to 24 months the adjustment is small and the added complexity costs you more in comprehension than it buys in accuracy. If you are modelling beyond two years, discount; below that, do not bother, and instead shorten your horizon.

A worked example: a Singapore e-commerce brand

A homeware brand, GST-registered, selling direct and on a marketplace.

Note what this figure is not. It is not “lifetime” — it is 24 months, stated plainly, which is the horizon they have actually observed. It is not revenue: the revenue equivalent would be 2.4 x S$109 = S$261.60, which is the number a less careful brand would have put in front of its board. And it is the number that makes the acquisition decision obvious: with a 24-month LTV of S$101.90, a fully loaded customer acquisition cost above about S$34 needs a good explanation, and above S$50 needs a rethink.

The second purchase is where the leverage is

Work the homeware example the other way and something useful falls out. The 2.4 orders per customer is an average across a cohort in which most people bought once. If, say, 62% of that cohort bought once and 38% went on to buy an average of 4.7 times, the maths still lands near 2.4 — but the two groups are worth wildly different amounts. The one-time buyer is worth S$44 of gross profit. The repeat buyer is worth around S$207.

That gap reframes the budget conversation. Moving the repeat rate from 38% to 45% — seven percentage points, achievable with a reorder email flow and a better post-purchase experience — lifts cohort LTV by roughly S$11 per customer, an increase of about 11% in what you can afford to pay for acquisition across the entire business. Winning an 11% improvement in cost per acquisition through media buying alone, in a competitive Singapore auction, is considerably harder work.

This is the argument for measuring LTV at all. It is not to produce a number for a slide; it is to reveal that the cheapest growth available to most local businesses is sitting in their existing customer list rather than in the ad auction.

The 3:1 ratio, and what it is really worth

Everyone quotes it: LTV should be at least three times CAC. It is worth knowing where it comes from, because the provenance changes how much weight it deserves.

The figure traces to David Skok’s SaaS metrics work, which observed that “the best SaaS businesses have a LTV to CAC ratio that is higher than 3, sometimes as high as 7 or 8.” Read that carefully. It is an observation about the best venture-backed subscription software companies, offered as a guideline for assessing whether such a business is ready to scale. It is not a derived threshold, it was never claimed to be one, and it does not describe a Singapore renovation contractor, a tuition centre or a dental clinic.

Three reasons to hold it loosely. First, it says nothing about cash: a 4:1 ratio with a 30-month payback is more dangerous for an SME than a 2.5:1 ratio paying back in six months. Second, it is extremely sensitive to how you calculate LTV — move from revenue to gross margin and a 3:1 becomes roughly 1.3:1 at a 44% margin, without anything about the business changing. Third, a very high ratio is not automatically good news; a 9:1 usually means you are underspending on acquisition and leaving growth on the table.

Use the ratio as a sanity check on a number you have already calculated honestly, and pair it always with payback period. Skok’s more useful observation, in our view, is the other one: that the best SaaS businesses recover CAC in five to seven months, and that profitability is “anemic” when recovery takes longer than twelve.

Five ways LTV gets inflated

1. Revenue instead of gross profit. The largest single distortion, typically overstating value by half or more depending on margin.

2. GST left in. Roughly 8.3% of a GST-inclusive order total, straight off the top, on every order in the calculation.

3. An invented lifespan. “Customers stay five years” from a company trading for two. If you have 18 months of data, publish an 18-month LTV. A shorter, observed figure beats a longer, imagined one every time.

4. Survivor bias in the cohort. Calculating average spend across “active customers” excludes everyone who left, which is exactly the population that drags LTV down. Always calculate across the full original cohort, including the ones who bought once and never returned.

5. Best-customer thinking. Founders anchor on their favourite clients. The median customer is usually worth a fraction of the mean, because a small group of heavy buyers pulls the average up. Look at the median as well, and if the two are far apart, segment rather than average.

What to do with the number once you have it

LTV is not a trophy. It has three specific uses.

It sets your acquisition ceiling. Your maximum sustainable CAC is a fraction of gross-margin LTV, with the fraction determined by how much cash you can tie up and how confident you are in the forecast. Conservative operators use a third; well-funded ones go higher.

It ranks channels properly. Different channels bring customers with different lifetime values. A channel with a higher cost per acquisition can be the better buy if its customers stay longer — something a cost-per-lead comparison can never show you. Segment LTV by acquisition source, which is where consistent UTM tagging and clean conversion tracking pay for themselves.

It redirects effort to retention. This is the one most Singapore SMEs underuse. If a second purchase raises a customer’s value several times over, then an email flow, a reorder reminder or a genuinely better unboxing experience may return more than another dollar of ad spend. Reviews matter here too, both for retention and for acquisition — our guide to getting Google reviews covers the mechanics without breaking the rules.

One PDPA note. Calculating LTV means holding purchase histories against identifiable individuals, which is personal data. Keep it to what you need, secure it, and remember that consent for marketing communications is separate from your right to hold transaction records — the legitimate interests exception does not cover direct marketing. Reusing a customer list for advertising audiences is a marketing use and needs the appropriate consent, a point our custom audiences guide goes into.

Frequently asked questions

How do I calculate customer lifetime value for a Singapore business?

Take a cohort of customers who first bought in a given month, measure the actual gross profit they generated over the following 12 or 24 months, and divide by the number of customers in that cohort including the ones who never returned. Use GST-exclusive figures, deduct cost of goods, shipping, payment fees and any marketplace commission, and state the horizon in the label — “24-month LTV”, not “lifetime value”.

Should LTV use revenue or profit?

Gross profit, always. Revenue-based LTV overstates customer value by the whole of your cost of goods, and at a typical 40% to 50% margin it roughly doubles the figure. Every decision that compares LTV against acquisition cost is meaningless unless both sides are on a profit basis.

Do I need to remove GST from my LTV calculation?

Yes, if your prices are GST-inclusive. The 9% is collected on behalf of IRAS and was never your revenue. Divide GST-inclusive order values by 1.09 before doing anything else — leaving it in overstates LTV by about 8.3% and, because platforms report the same inflated figure, quietly corrupts your return-on-ad-spend numbers as well.

Is a 3:1 LTV to CAC ratio the right target?

Treat it as a rough sanity check, not a rule. It comes from David Skok’s observation that the best SaaS businesses exceed 3 and sometimes reach 7 or 8 — a guideline drawn from venture-backed subscription software, not a threshold derived for local service businesses. It also ignores cash timing entirely, so always pair it with your CAC payback period, which for most Singapore SMEs is the more urgent number.

How much data do I need before LTV is meaningful?

Enough for at least one cohort to be as mature as your reporting horizon — so 12 months of trading for a 12-month LTV. Below that you are extrapolating, which is legitimate if you say so. A business trading nine months should publish a nine-month LTV and revise it as cohorts mature, rather than modelling a five-year lifespan it has no evidence for.

Why is my average LTV so much higher than my median customer?

Because a small number of heavy repeat buyers pull the mean up. This is normal and it matters: if you plan acquisition spend against a mean inflated by your best twenty customers, you will overpay for the typical one. Report both, and if the gap is large, segment your LTV by customer type rather than averaging across all of them.

Where to start

Pick one cohort — customers who first bought 12 months ago — and calculate their actual gross profit to date, GST stripped, including everyone who never came back. That single number will be lower than you expect and more useful than anything you have used before.

Then put it next to your fully loaded CAC. If the gap is thin, the answer is usually retention rather than cheaper media, because a modest lift in repeat rate moves LTV further than a modest lift in click cost moves CAC.

If you would like this built properly — cohorts constructed from your real sales data, LTV segmented by acquisition channel, and both numbers put on a dashboard you can check — our performance marketing team does this at the start of every engagement. The wider framework sits in our performance marketing guide, the attribution caveats in attribution models, and real outcomes in our client case studies. If you sell online, our Singapore e-commerce guide covers the marketplace-versus-own-store decision that shapes LTV more than almost anything else.

Last updated 1 August 2026. Written by Adrian Tan and the SDM team. Sources: IRAS GST guidance (9% since 1 January 2024), CPF Board contribution rates from 1 January 2026, PDPC advisory guidelines on consent for marketing, and David Skok’s SaaS metrics work on LTV to CAC and CAC payback.

Related measurement guides

• PDPA-compliant marketing tracking in Singapore — holding purchase histories to calculate lifetime value is processing personal data, and this is what the PDPA requires of it.