Picture somebody sensible. They read a comparison, picked a card advertising 4 miles per dollar, and made it the card they use for everything. Groceries, dinners, the phone bill, the flights, the lot — around S$4,000 a month. A year later they check the balance and it is roughly a third of what they expected.
Nothing went wrong. Nobody was misled. They earned exactly what the card promised.
The card promised 4 miles per dollar on the first S$1,000 each month. Everything after that earned 0.4 — a tenth as much. Across a S$4,000 month, that averages out at 1.3 miles per dollar.
That is the entire subject of this article, and if it is new to you, it is the most valuable thing on this page.
The number on the advertisement is a number about S$1,000
Every bonus card in Singapore is built the same way. It pays a headline rate — usually 4 miles per dollar — on eligible spending up to a monthly cap, and everything past that cap earns a base rate.
That base rate is the part worth memorising, because it is almost identical across the market: 0.4 miles per dollar. It arrives as one point per dollar at one bank and one unit per five dollars at another, but it lands in the same place. Roughly 0.4% back.
So there are two rates, not one. The advertisement quotes the first. Your statement earns a blend of both, and the blend gets worse the more you spend.
The advertised rate applies to the first S$1,000 a month. After that, spending earns the 0.4 base rate every major bank falls back to — roughly a tenth as much. What you actually earn is the average of the two, and it gets worse the more you spend. Each line is one more card in use, each with its own S$1,000 cap.
What this is. Arithmetic on a 4 miles-per-dollar bonus, a S$1,000 monthly cap and the 0.4 base rate that Citi, DBS and UOB all fall back to. No card is named: the shape is the same whichever you hold, and only the cap and the rate move.
The dashed red line. It marks 1.7 miles per dollar, which is where a miles card draws level with a 1.7% flat cashback card — because a mile cashed out is worth about a cent, so miles per dollar and percent back are the same number. Anything below that line loses to cashback unless the miles are redeemed for more than a cent.
The second limit. The solid lines assume every dollar up to the cap qualifies for a bonus category — the brochure case. Real spending does not arrive pre-sorted: groceries, insurance, school fees, tax and rent are largely excluded or carry an admin fee that exceeds the reward. So a second limit bites, and the dashed line shows it. When you have run out of eligible spending rather than run past the cap, adding another card changes nothing.
a theenoughpoint.com toolRead the solid red line, which is one card. At S$1,000 a month it earns the full 4. At S$4,000 it earns 1.3 — worse than an ordinary card with no categories, no caps and nothing to remember. The card did not fail. There was simply four times as much spending to spread the bonus across.
This is why “which card is best?” is close to an unanswerable question. The same card is excellent at S$800 a month and mediocre at S$4,000, and neither fact is about the card.
The obvious fix, and when it does not work
If one cap is the problem, more caps are the answer. That is what the other lines on the chart show: at S$4,000 a month, four cards hold the full 4 where one card gives 1.3. Three times the miles, same spending.
Which is where nearly every guide stops. Here is what it leaves out.
Those caps are not interchangeable. A card capped at S$1,000 of online spending does nothing for money you spend in shops. A second online card does nothing once your online spending has run out. The caps sit on categories, and the categories are narrower than they sound.
Now list where a household’s money actually goes. Rent or mortgage. Insurance premiums. School fees. Income tax. Groceries. For most families those are the biggest lines in the budget, and they are almost all either excluded from bonus categories, impossible to put on a card at all, or chargeable only through a service that takes a 1.8% to 2.3% fee — which is more than the miles are worth.
So there are two different walls, and they need opposite responses:
- Hitting the cap — you have eligible spending spilling into the base rate. Another card genuinely helps, and the calculator below will tell you by how much.
- Running out of eligible spending — you have unused bonus capacity sitting on cards you already hold. Another card does nothing at all. You would be paying an annual fee for capacity you are not using.
The dashed line on the chart is what the second wall looks like: four cards, but with only 60% of spending landing somewhere a bonus applies, earning 2.56 rather than 4. Most households, on how the exclusions actually work, are hitting the second wall while shopping for a solution to the first.
What the whole exercise costs to run
Before deciding it is worth it, price the machine. None of this appears in any earn-rate comparison.
None of these appear in an earn-rate comparison, and together they decide whether the stack was worth assembling. The two marked heavy are the ones that most often turn a winning calculation into a losing one.
Four cards at around S$196 is roughly S$780 a year — about 1.6 percentage points of rebate on S$48,000 of spending, which is more than the entire advantage of a cashback card. Most are waivable on request; none should be assumed away.
Charged per transfer, per currency, regardless of size. A four-currency stack converting once a year pays it four times. American Express and HSBC charge nothing, which is a real quality difference between two cards earning the same rate.
Several programmes will not move points below a 10,000-mile block — 25,000 Citi ThankYou points, for instance. Spread thin across four currencies, the tail of each balance can sit permanently below its own threshold. Points you cannot move are points you do not have.
American Express points do not expire, HSBC points run 37 months, and Citi varies by card — no expiry on some, up to five years on others. Bank points expiring is a separate clock from KrisFlyer miles expiring at three years, and a stack runs several clocks at once.
Several bonus rates require a monthly floor — S$800 or S$1,000 is common — and missing it does not reduce the bonus, it removes it. This is the asymmetry that punishes a stack you are not actively operating, and it is why the honest input is how many cards you will run, not how many you hold.
Rent, tax, insurance and school fees mostly cannot be charged directly. Routing them through a bill-payment service costs an admin fee that, at ordinary earn rates, usually exceeds the value of the miles it generates. This is the single largest category most households cannot bring into the game at all.
Sources. Currency mechanics — conversion fees, expiry and minimum transfer blocks — from The MileLion’s comparison of Singapore points currencies; category rates, caps and minimum spends from his card guide and 2026 strategy. Verified 13 August 2026. Confirm against your issuer’s own terms before acting on any of it.
a theenoughpoint.com toolTwo of those reverse the sums on their own.
Annual fees are not small against a stack. Four cards at roughly S$196 each is about S$780 a year. On S$48,000 of spending that is 1.6% of everything you spend — larger than the entire advantage of a flat cashback card. Most are waivable if you ask. “I will just get it waived” is a plan, not a fact.
Missing a minimum spend does not cost you a little. Several bonus rates require a monthly floor. Miss it and you do not earn a smaller bonus, you earn the base rate on everything that month. A system you are not actively running does not fade gently; it drops off a cliff in exactly the months you were too busy to notice.
Work out your own number
Put your real figures in. The defaults are typical of the market rather than any particular card, because the specifics will have moved by the time you read this.
Defaults are market-typical for August 2026 — a 4 miles-per-dollar bonus on the first S$1,000 each month, the 0.4 base rate every major bank falls back to after that, and 1.7% as the best uncapped flat cashback. No card is named or assumed: terms move monthly, and these are yours to overwrite with the ones on your statement.
| If you run | Average | Return | Beats cashback? |
|---|
What this is. Your own spending run through the cap arithmetic, priced at the value per mile you actually achieve, net of the fees you actually pay, against the cashback card you could have carried instead. It is the comparison the marketing never runs, because the marketing quotes the rate inside the cap.
Why the category share matters more than the card count. Bonus capacity is not a pool you can point anywhere. A card capped at S$1,000 of online spending does nothing for your contactless spending, and a second online card does nothing once your online spending runs out. So two limits apply — the caps you hold and the spending that qualifies for them — and the tool tells you which one is actually costing you, because the remedies are opposite. If categories bind, another card changes nothing.
What it still leaves out. Minimum spends, which turn one missed month into the base rate on everything. Conversion fees of roughly S$27 a transfer, and the points stranded below each programme’s minimum transfer block. Sign-up bonuses, which are real but one-off, and which flatter any first-year comparison. And the time, which is the cost nobody prices and everybody pays.
What this is not. A recommendation of any card, bank or strategy, or a claim about what any card currently offers. It is arithmetic on figures you supply. Verify every rate, cap and fee against your issuer’s own terms before deciding anything.
a theenoughpoint.com toolThe row to look at is not the one matching how many cards you hold. It is the next one down — what one more card would actually add, before you apply. On ordinary inputs that gain is often smaller than the annual fee, which is the calculation the application form does not show you.
Watch what happens if you set the category share to 60%: the gain stops after the third card. The fourth and fifth add nothing whatever, because there is no eligible spending left for them to catch. That is the second wall, in numbers.
The comparison that settles it
Here is the conversion that makes all of this legible, and it is worth remembering.
One mile per dollar, cashed out at a cent a mile, is a 1% rebate. So miles per dollar and percent back are the same number, as long as you are cashing miles out rather than flying on them.
That makes the benchmark obvious. The best flat cashback card in Singapore pays 1.7% on everything, with no caps, no categories, no minimum spends and no attention required. So your real average across all your spending has to beat 1.7 miles per dollar just to draw level with the card you could have carried without ever thinking about it.
Our sensible person at 1.3 is not drawing level. They are behind, and they have been doing homework to get there.
There is one way out of that, and it is the subject of the companion piece: redeem miles for more than a cent. A mile spent on a business-class seat you would genuinely have paid for is worth several times a mile spent in the KrisShop catalogue. Get that right and a 1.3 average becomes a perfectly good return. Get it wrong and no amount of card-juggling saves you.
The earning side sets your rate. The redemption side sets your multiplier. You need both, and most people optimise only the first.
What to actually do
-
Look at where your money goes, not at any card. Split last month into what a bonus card could reward and what it could not. Rent, insurance, tax, school fees and most groceries fall in the second pile. That split sets your limit, and no product changes it.
-
Work out which wall you are hitting. Unused capacity on cards you hold means another card is pointless. Eligible spending spilling into the base rate means another card helps.
-
Count the cards you will actually run — the number you will still be routing correctly in month eleven, after a busy fortnight and a holiday. That is a smaller number than the one you would like it to be.
-
Take off the running costs before deciding, and treat fee waivers as requests rather than rights.
-
Then compare it honestly to doing nothing. If your average is under about 1.7 miles per dollar and your miles end up cashed out rather than flown, a single flat cashback card would have paid you more and asked nothing of you. For a great many households that is the right answer, and there is no shame in it.
The version of this hobby that pays is real — our own redemption ledger is 1.5 million miles of evidence. So is the version where somebody earns 1.3% while believing they earn 4. The only thing separating them is the arithmetic above.
No commercial relationship
We hold no affiliate, referral or advertising relationship with any bank or card issuer, and nothing here is a paid placement. No card is recommended, and no card's current terms are asserted — issuers revise rates, caps and fees continually, and the figures here are typical market examples for August 2026 rather than an offer summary. This is general information about how reward structures work, not financial advice, and not a view on what any individual should apply for, hold or spend.
Sources and verification — verified 13 August 2026. Every load-bearing figure below, one line each.
- The 0.4 miles-per-dollar base rate that spending above a bonus cap falls to, consistent across Citi (1 ThankYou point per S$1), DBS and UOB (1 unit per S$5): confirmed against The MileLion’s Citi Rewards review and independent market comparisons. It is the most important constant in this article and the one card marketing never leads with.
- The S$1,000 monthly bonus cap used throughout as the typical figure, and the fact that caps tightened during 2026 — one card’s cap was cut from S$1,500 to S$1,000, and contactless sub-caps of S$600 to S$750 are now common: The MileLion’s 2026 credit card strategy and his card guide by category.
- Category rates and their caps — 4 miles per dollar on online and contactless spending, with higher narrow-category rates existing (6 on certain e-commerce platforms, 10 at selected grocery and transport merchants on a minimum spend): The MileLion’s card guide. We model 4 rather than the outliers, because the outliers carry minimum spends and merchant lists that make them unrepresentative of a household’s average.
- The four-card “minipack” for spenders below roughly S$3,800 a month — two cards covering online spending and two covering contactless — is Aaron Wong’s framing in the 2026 strategy, as is the observation that most people will not need the full twelve-card stack he documents. Our contribution is the arithmetic underneath that advice, not the advice.
- Currency mechanics — conversion fees of roughly S$25–30 per transfer with American Express and HSBC free, minimum transfer blocks of 10,000 miles at several issuers, and expiry ranging from none to 37 months: The MileLion’s comparison of Singapore points currencies.
- Annual fees of S$196.20 to S$261.60 on the main general-spend and bonus cards, first year commonly waived and later waivers at the issuer’s discretion: issuer terms and The MileLion’s card reviews.
- Bill-payment admin fees of roughly 1.79% to 2.25% for routing rent, tax and similar through a payment service: The MileLion’s analysis of earning miles through cashback and bill payments.
- The 1.7% cashback benchmark, the highest uncapped flat rate available in Singapore as at August 2026, with 1.5–1.6% alternatives and exclusions that apply to all of them: compiled from published card terms and market comparisons. Rates change; check the issuer’s page before treating the hurdle as settled.
- The averaging arithmetic and the two-wall model are ours, set out in full in the module behind the calculator. The second wall — that bonus capacity is limited by eligible categories and not only by caps — is the correction that separates this from a simple cap calculation, and an earlier draft of this article got it wrong by treating capacity as a single pool.
- Rewritten 13 August 2026 for clarity, on the reader’s objection that the original was hard to follow. The argument and every figure are unchanged; the jargon is not. The piece now leads with what the reader experiences rather than with the mechanism, and states miles per dollar alongside percentage-back throughout, since the two are the same number at a one-cent redemption.
A standing credit. As with the redemption piece, most of what Singapore knows about this subject was worked out in public by Aaron Wong at The MileLion. His card guide and annual strategy are the reference works; where this article differs is in asking what the structure earns on an ordinary household’s spending rather than on an optimiser’s. For the card-by-card detail this piece deliberately avoids, his guide is where to go.
