Annual Fee Break-Even Math: The One Formula for Any Card (With Worked Numbers)

Every "is this card worth the fee" article ends in a shrug — "it depends on your spending." It does, and the dependency is a single division. Net fee divided by the extra cents each dollar earns over a no-fee card gives the spending at which the fee pays for itself. Below that number the fee is a loss; above it, every dollar is a gain. This guide gives the formula, the baseline, and the break-even spend for seven common cards, so you can stop shrugging.

Quick answer

Break-even spend = (annual fee − credits you would use anyway − benefits you would pay for) ÷ (card's earn rate − 2% baseline, in cents per dollar). Against a 2% cash-back baseline and a 1.5-cent point valuation, the Sapphire Preferred breaks even at about $1,800 of 3x spending once its $50 hotel credit is counted; the Amex Gold at about $2,100 of 4x grocery and dining spending after its realistic credits; the Ink Preferred at about $3,800 of 3x business spending; the Venture X at $0 because its credits cover the fee; and the Amex Platinum at a number no one reaches on spending alone — its case rests on credits and lounges, not earn rates.

• The baseline matters: compare against a 2% no-fee card, not against zero
• Credits count only if you would have spent the money anyway
• Year one is not the test: the welcome bonus makes almost any card profitable in year one; run the math for year two
• Premium cards rarely break even on earn rate: they break even on credits, lounges, and insurance, or not at all

💡 Pro tip — Write the answer as a single number and compare it to last year's statements. "I need $2,100 of grocery and dining spending to justify the Gold" is a question you can answer in a minute; "is the Gold worth it" is not.

The formula

Three inputs, one division:

1. Net fee = annual fee − credits you would have spent on anyway − benefits you would otherwise pay for (Global Entry, lounge day passes, travel insurance you would buy).
2. Extra earn per dollar = (card's rate in the category × cents per point) − (baseline rate × 1 cent). The baseline is a 2% no-fee cash-back card, because that is what the same dollar earns with no fee at all.
3. Break-even spend = net fee ÷ extra earn per dollar.

Worked once: the Amex Gold earns 4x at restaurants and U.S. supermarkets. At 1.5 cents per Membership Rewards point that is 6 cents per dollar, against 2 cents on the baseline, so 4 extra cents. The fee is $325; a cardholder who uses the $120 Uber Cash and the $120 dining credit but nothing else has a net fee of $85. Break-even: $85 ÷ $0.04 = $2,125 of 4x spending a year — about $180 a month. Nearly everyone clears it.

Key takeaway: the same formula that says the Gold pays for itself at $180 a month says the Platinum never pays for itself on spending, because 5x on flights minus 2% is 5.5 extra cents against a net fee of several hundred dollars.

Point valuations used here

Raise the valuation and every break-even falls; lower it and every break-even rises. Sweet-spot redeemers should run the math at 2 cents; portal-only redeemers at 1.25 (Preferred) or 1.5 (Reserve).

Break-even spend for seven cards

All figures against a 2% baseline at the valuations above, using realistic credit usage for a typical cardholder.

Two patterns fall out of the table:

• Mid-tier cards break even on earn rate. The Preferred, Gold, and Ink Preferred all pay for themselves on a few thousand dollars of category spending, which is why they are the workhorses of most wallets.
• Premium cards break even on credits. The Venture X, Reserve, and Platinum each have a break-even that swings from $0 to five figures depending on whether the credits are used. Earn rate barely moves the answer. That is the real subject of premium card credits compared.

The year-one distortion

A welcome bonus makes almost any card profitable in year one. The Sapphire Preferred's bonus at 1.5 cents is worth many multiples of its fee; the Amex Platinum's bonus is worth more than two years of fees. So the year-one math is trivially positive and tells you nothing about whether to keep the card.

Run the formula for year two, with the bonus gone and the credits at what you actually used in year one. That is the number that decides keep, downgrade, or cancel — and the annual fee refund rules by issuer give you a window to act on it after the second fee posts.

Two adjustments the formula needs

Insurance and protections. A card whose trip delay, baggage delay, primary rental coverage, and purchase protection would otherwise cost you money has a benefit you should subtract from the fee. Value it at what you would actually pay: a frequent traveler who would buy a $100 annual travel policy can subtract $100 from the Reserve's fee; someone who would never buy one subtracts nothing.

Your time. Monthly credits — $10 Lyft, $25 DoorDash, $20 Resy — cost minutes to remember and use. A card whose value depends on twelve monthly coupons is worth less than the same dollars in one annual credit. The formula cannot price this, but it should make you round the coupon-like credits down.

Real use case: two cards, one household

A household spends $9,000 a year on groceries and dining, $4,000 on travel, and $20,000 on everything else.

Option A — Amex Gold plus a 2% card. Gold on groceries and dining: $9,000 × 4 × 1.5 cents = $540, minus $85 net fee, gives $455 over the baseline. Travel and everything else at 2% on the cash card. Net gain over an all-2%-card wallet: $455.

Option B — Sapphire Preferred plus a 2% card. Preferred on dining ($5,000 of the $9,000, assuming groceries are in-store) at 3x: $5,000 × 3 × 1.5 = $225, plus $4,000 of travel at 2x: $120, total $345, minus $45 net fee: $300 over the baseline. Net gain: $300.

Option C — Amex Platinum alone. $4,000 of travel at 5x (assume $3,000 is airfare): $3,000 × 5 × 1.5 = $225 plus $1,000 at 1x: $15; everything else at 1x: $29,000 × 1.5 cents = $435. Total $675 against a baseline of $33,000 × 2% = $660, plus $230 of realistic credits, minus $895 fee. Net: −$650, before lounges.

Option A wins on spending. The Platinum only enters the conversation if this household values lounges and a $600 hotel credit it does not currently use.

Decision framework

Common mistakes

1. Comparing against zero. "The Gold earned me $540" ignores that a free 2% card would have earned $180 on the same spending. The gain is $360, not $540.
2. Valuing points at the portal rate when you transfer, or the transfer rate when you use the portal. Use the rate you actually redeem at.
3. Counting credits at face value. Only the ones you would have spent anyway. Premium card credits compared sorts them.
4. Running year-one math. The bonus decides year one; the earn rate and credits decide every year after.
5. Forgetting the downgrade option. Most fee cards have a no-fee sibling that keeps the account open; the choice is rarely keep-or-cancel.
6. Using one card's break-even to judge another. The Preferred's $1,800 and the Platinum's $12,100 are not comparable — one is an earn-rate card, the other a credits card.

Editor's take

We keep this formula on one line because the industry's answer — "it depends" — is a way of not answering. Our reading of the table: the mid-tier cards are the easy yeses, breaking even on a couple of hundred dollars a month of ordinary spending; the Venture X is a yes for anyone who books one trip a year through the portal; and the Reserve and Platinum are not spending cards at all — they are subscriptions to a credit-and-lounge bundle, and the formula's job is to price that bundle honestly. Compute your number, compare it to last year, and let the division decide.

Related reading

• Premium card credits compared — the realistic credit values that feed the net fee
• Do premium cards make sense if you travel twice a year? — the formula applied to an occasional traveler
• Annual fee refund rules by issuer — the window to act after the fee posts
• Should you keep or cancel the Amex Platinum? — the Platinum's year-two math in full
• Should you keep or cancel the Chase Sapphire Reserve? — the Reserve's
• Chase Sapphire Preferred review — the easiest break-even in the table