Once a year, every premium card in your wallet sends you a bill for the privilege of carrying it. Most people pay it reflexively, glance at the fee with mild irritation, and move on. Some people cancel without thinking it through. Both approaches leave money on the table.
The right answer - keep, downgrade, or cancel - depends on a simple calculation you can run in about 10 minutes. Here's the framework.
Step 1: Add Up Credits You Actually Use
Not credits that exist. Credits you personally use, in your actual life.
This is where most people overestimate their card value. The Amex Platinum has just over $3,000 in headline credits - $200 airline fee credit, $600 hotel credit, $400 Resy dining, $200 Uber Cash, $300 digital entertainment, $155.40 Walmart+ credit, Global Entry, and more. That sounds incredible. But if you don't use Walmart+, don't care about Peacock, and rarely fly airlines where the $200 credit applies cleanly, your real credit haul is much smaller.
Be ruthless. Count only what you actually extracted last year. For recurring credits, only count them if you'd have paid for that thing anyway - the Uber Cash is only "free money" if you were going to take Ubers regardless.
Some easy-to-use credits worth noting for 2026:
- Chase Sapphire Reserve: $300 travel credit (automatic, applies to basically any travel purchase)
- Amex Gold: $120 Uber Cash ($10/month), $120 dining credit ($10/month at Grubhub, The Cheesecake Factory, Goldbelly, Wine.com, and Five Guys). Those are the two easiest of four; the full stack is $424, with a $100 Resy credit and an $84 Dunkin' credit making up the rest.
- Capital One Venture X: $300 travel credit (via Capital One Travel portal only), 10,000 anniversary miles (~$100 value)
- Amex Platinum: $200 airline fee credit (incidental fees only - seat upgrades, bags - not ticket purchases), $200 Uber Cash ($15/month plus $20 in December)
Step 2: Calculate the Earn Differential vs. a No-Fee Card
Every dollar you put on a fee card instead of a no-fee card either earns you more points or it doesn't. Calculate that difference.
The baseline is the Chase Freedom Unlimited at 1.5x Ultimate Rewards on everything, no annual fee. Or the Citi Double Cash at 2x ThankYou Points. Or the Amex Blue Business Plus at 2x Membership Rewards everywhere (up to $50k/year). All three pay in transferable points rather than cash, so a penny apiece is their floor, not their ceiling.
Example: You spend $800/month on dining. The Amex Gold earns 4x Membership Rewards on dining. At our balanced 2.5¢ per point valuation, that's 10% effective return. A no-fee cash back card earns 1.5-2%. The differential is about 8-8.5 percentage points on dining, applied to $9,600 in annual dining spend = roughly $768-$816 in additional annual value from dining alone.
Run this math for each major spending category. Add them up.
Step 3: Value Perks You Genuinely Use
Some card benefits are harder to quantify but have real value:
- Airport lounge access: Priority Pass visits are worth roughly $30-50 each if you use them. United Club or Amex Centurion access is worth more. Our catalog puts a Centurion visit at $50-$100 if you're honest about food, drinks, and comfort. If you fly 20 times a year and use a lounge on half those trips, that's $300-1,000 in annual lounge value.
- Trip delay/cancellation insurance: Hard to value until you use it. Think of it like insurance - the expected value is low annually, but the peace of mind has real worth and the payouts when you need them can be $500-2,000.
- Hotel elite status: The Hilton Aspire gives automatic Hilton Diamond status. If you stay 10+ nights a year at Hilton properties, Diamond status - with room upgrades, free breakfast, and suite night awards - can be worth $300-600 annually.
- Purchase protection: Not exciting, but useful. If you buy expensive electronics or appliances on a card with purchase protection (Amex cards generally have the best), you're saving on extended warranty costs and replacement in case of damage or theft.
Step 4: Compare Total Value to the Fee
Add up Steps 1-3. If the total exceeds the annual fee, keep the card. If it doesn't, you have two real options: change the product to a cheaper card in the same family, or close it.
Let's run through a real example:
Example: Amex Gold ($325 annual fee)
The Gold's credits come to $424 at face value. This example counts $192 of that, which is the two credits from the easy list above at honest capture rates, and ignores the $100 Resy and $84 Dunkin' credits entirely on the grounds that this cardholder doesn't use them. If you do, add them back.
- $120 Uber Cash: $120 (if you use Uber regularly)
- $120 dining credit: $72 (being honest - Grubhub is a stretch, but a Cheesecake Factory or Five Guys order clears $10 most months. Conservative estimate.)
- Earn differential on $800/month dining at 4x vs. 1.5% CB baseline: $816
- Earn differential on $500/month groceries at 4x vs. 1.5% CB baseline: $510
- Total value: $1,518
- Annual fee: $325
- Net: +$1,193
For someone who spends heavily on dining and groceries, the Amex Gold is a remarkable value even at $325. The math isn't close.
Example: Chase Sapphire Reserve ($795 annual fee)
Same discipline. The Reserve's credits come to $2,202 across nine benefits; this example counts $300 of that plus lounge access, because those are the two this particular cardholder converts. Our full Reserve review works all nine, and a traveler who books prepaid hotels through Chase gets a long way past $300.
- $300 travel credit: $300 (you travel, so this gets used)
- Priority Pass (10 lounge visits/year at $40 avg value): $400
- Earn differential on $400/month of flights and hotels booked direct at 4x vs. 1.5% CB baseline: $350
- Earn differential on $300/month dining at 3x vs. 1.5% CB baseline: $184
- Total value: $1,234
- Annual fee: $795
- Net: +$439
Positive, and comfortably so at this spending level - but notice how much of it rides on two assumptions. Cut to four lounge visits a year and the net falls to +$199. Halve the travel and dining spend on top of that and the Reserve goes negative. This illustrates why the Reserve is genuinely close-call territory for anyone who isn't both traveling and dining heavily.
What a Retention Offer Actually Is
Somewhere in this process you'll read that you should call and ask for a retention offer. Here's the honest version of what that means.
A retention offer is a discretionary incentive an issuer may extend to keep an account open - usually bonus points or a statement credit, sometimes with a spend condition attached. Whether one exists for your account is decided by the issuer's own internal model, which weighs things like how much you put on the card and how long you've held it. It isn't a published benefit, it isn't a negotiation, and it isn't something you're owed.
Two things follow from that, and they're the whole of our advice on the subject.
A retention offer can't rescue a card that isn't earning its fee. It's a one-year patch on a recurring cost. If the four steps above came out negative, an offer that closes the gap this year leaves you in exactly the same position twelve months from now, having renewed a card you'd already concluded wasn't worth it. Run the framework first, on the card's own merits. An offer can improve a "keep." It shouldn't convert a "cancel."
Plenty of accounts have nothing available. Some issuers extend them far more readily than others, and low-spend accounts frequently get nothing at all. Anyone quoting you a success rate or a typical amount is reporting anecdotes from a system that's deliberately opaque, and building a plan on those numbers is how disappointment happens. We don't publish figures on this, because we don't have data we'd stand behind.
If you want to ask, ask once, honestly, and let the answer be the answer. Don't build a year around it.
Product Change: The Smarter Cancel
Canceling a card means losing your account age, which can ding your credit score slightly. A product change eliminates the fee while keeping the account, and for a card that's simply outgrown its price it's usually the cleanest outcome available.
Four things hold across every issuer we track. The account generally needs to be at least 12 months old before a change is possible. It doesn't count as opening a new account, so it doesn't add to your 24-month count and doesn't come with a hard pull. Your points balance survives the change. And downgrades are the easy direction - upgrade requests are a different animal, and at Capital One in particular they're frequently declined unless you accept a hard pull.
Common product change paths worth knowing:
- Chase Sapphire Reserve → Chase Freedom Flex or Freedom Unlimited: Keeps your UR balance and your account age, no annual fee. You lose the lounge access and the premium earning rates - and, unless you hold another Sapphire or the Ink Business Preferred, you lose airline and hotel transfer access entirely. See the warning below.
- Chase Sapphire Preferred → Chase Freedom Unlimited: Same logic and the same transfer-access caveat.
- Amex Platinum → Amex Green: Drops from $895 to $150, keeps the MR currency and 3x on travel and dining. Loses most of the credits, but if you weren't using them, it's a rational move.
- Amex Gold → Amex Green: Fee drops from $325 to $150. Loses the 4x dining and grocery categories, but keeps the MR ecosystem.
- Capital One Venture X → Capital One Venture: $395 → $95, loses the Capital One lounges and the anniversary miles. Capital One is the friendly case here: moving down to Venture or VentureOne keeps full transfer access.
The transfer-access question is the expensive one, and it works differently at each issuer. Chase is the harsh case and it's binary: downgrade out of a Sapphire or the Ink Business Preferred with no other gateway card and your Ultimate Rewards lose partner access completely, becoming portal-and-cash-back points. Citi is gentler - leaving Strata Premier doesn't strip transfer access outright, it just moves you to a worse rate on most airline partners. Capital One keeps it. Check which of those you're in before you downgrade, because it's a quiet devaluation of a balance you already hold.
Two more things people get wrong about bonus eligibility. Product changes don't reset it: change from one Sapphire to another and you won't earn the new card's welcome bonus, because each Sapphire product is once per lifetime. And at Amex, changing into a card permanently burns that card's welcome bonus - the bonus you lose is on the card you're moving to, not the one you're leaving. Capital One's Venture-family moves carry the same forfeit on that family's 48-month clock.
What a change does not do is close off a bonus you'd otherwise have had. Closing a card doesn't reset once-per-lifetime language either, so the choice between downgrading and canceling isn't a bonus-eligibility question at all - it's an account-age and points-access question.
The Grace Period: Your Cancellation Window
This is not widely known: close the card within about 30 days of the annual fee posting and most issuers will refund it. Chase is the outlier in your favor, at around 41 days.
Those are the working estimates we hold for nine issuers, and we hold them as estimates deliberately. Only Chase's and Amex's are well documented. Citi's has been reported anywhere from 30 to 60 days, so we use the conservative end. Most of the rest are around 30 days on thinner evidence, and a grace period can vary by card product, by how long you've held the account, and by whatever the issuer's policy is this year.
So treat 30 days as the planning assumption and confirm the real number for your own card, with the issuer, before you rely on it. The failure mode here is expensive and entirely avoidable: assume you have 45 days, discover you had 30, and you've bought another year of a card you'd decided against.
What the window buys you is time to decide properly rather than reflexively. The fee posting is the prompt to run the four steps, not a deadline to react to on the spot - you have a few weeks, and a few weeks is more than enough.
The One Situation Where You Should Just Cancel
If you've had the card for 2+ years, you've extracted all the easy value, your spending patterns have shifted so you're not hitting the bonus categories, and a product change still leaves you with a card you'd never use - cancel it.
The credit score impact of closing an account is usually modest and temporary, especially if you have other long-standing accounts. A card that's costing you $400-800 per year in fees without generating commensurate value is just burning money.
One exception: if the card is your oldest account, closing it has a larger impact on your average account age. In that case, product-change to a no-fee card instead of canceling outright.
