The Real Cost of Credit Card Annual Fees: Should You Keep Your Cards?

A $695 annual fee is not a price. It is a wager — that you will use enough of the card’s benefits, in the right months, on the right purchases, to come out ahead of the cash you handed the issuer in January. Most people never run the math. The card sits in the wallet, the fee posts every year, and the question of whether it is earning its keep is quietly never answered.

This is the math, the framework, and the part most people miss when they decide to keep — or downgrade — a premium card.

The only formula you need

Card ROI is the difference between two numbers:

Captured value is what you actually used. Not what was offered. Not what the marketing page says you could theoretically earn. What you, specifically, redeemed in the last twelve months. If you paid $695 for the Platinum and captured $400 of credits, your real ROI is −$295. The card cost you money.

The trick is being honest about the first number — because card issuers deliberately make it hard to know.

A worked example: the Amex Platinum

The Platinum charges $695 a year and advertises around $1,500 in annual statement credits. Here is what the math actually looks like for a typical cardholder who travels a few times a year and eats out occasionally:

Run that table for your own usage and the answer becomes obvious. Two cardholders with the same Platinum can have opposite ROI: a frequent traveler who pre-buys airline gift cards in November, books a hotel through FHR every year, and uses Uber daily can clear $1,200+. A weekend traveler who forgets the airline credit and doesn’t book hotels through Amex captures maybe $400. Same card. Same fee. Different verdicts.

Three classes of cards to evaluate differently

1. Fee-stack cards (Platinum, Reserve, Venture X, Bonvoy Brilliant)

These cards have annual fees above $400 and survive on benefit stacks. Your ROI is dominated by whether you actually usethe credits, not by your spending pattern. If you do not travel, do not stay at hotels through the issuer’s portal, and do not eat at restaurants that take Resy, these cards lose money for you no matter how much you spend on them.

Honest test: write down what you redeemed last year. If captured value − fee is negative, downgrade or close before the next fee posts.

2. Earn-rate cards (Gold, Sapphire Preferred, Capital One Savor)

Mid-fee cards in the $95–$325 range justify themselves on category multipliers — 4x on dining, 3x on travel, and so on. ROI here is about category spend, not credit usage. If you put $20,000 a year through a 4x dining card, you earn 80,000 points. At conservative redemption values (~1.5¢ per point in transferable currencies), that is roughly $1,200 of value against a $325 fee. Worth it.

Honest test: pull your category spend from last year. Multiply by the card’s rate. If the points value comfortably exceeds the fee, keep.

3. No-fee anchor cards (Freedom Unlimited, Double Cash, Active Cash)

No annual fee means no ROI math to do — there is no fee to recoup. These cards exist to catch the spending that does not fit any 3x/4x bonus category. The question for these cards is not should I keep this, it is am I routing the right purchases to it. Our no-fee landscape guide covers the signals for when this changes.

Red flag patterns: when to downgrade

The signs that a card is no longer earning its keep:

• You did not use the airline / hotel / dining credit last year. A $200 credit you did not use is $200 of pure loss against the fee. Two of them and a $695 card is suddenly costing you $295 net.
• You forgot the card was in your wallet. If a card sits in a drawer because it does not have a meaningful 3x+ bonus on any category you spend in, the fee is paying for benefits you forgot existed.
• Your travel pattern changed. A Reserve made sense when you flew United monthly. If you now fly twice a year, the lounge access value collapses and the fee does not.
• The issuer added a new card with a better benefit stack. Card issuers refresh products. The card you got in 2022 may be the worst version of its product line by 2026.

When premium cards genuinely pay for themselves

Premium cards are not a scam. They are a leverage instrument: the people who come out ahead have a behavior pattern that lines up with the credit stack. The clearest signs a $400+ fee card is worth keeping:

• You will book at least one hotel stay through the issuer’s portal this year (captures the hotel credit cleanly).
• You will use the airline credit before December 31 — even if it means pre-buying $200 in gift cards in November.
• You actively use at least three of the smaller monthly credits (Uber, dining, streaming). The math gets easier when you treat each monthly credit as a recurring discount on something you would buy anyway.
• You fly enough to use the lounge access. Two domestic round-trips with lounge stops is roughly $100–$200 of real value. Twelve is significantly more.

The work no one wants to do — and the tool for it

The reason most cardholders do not run this math is that doing it manually is tedious. You have to remember every credit, every reset date, every benefit you used, and then subtract — for each card, every year. The point of Cardier is to do that bookkeeping continuously instead of in a yearly panic: every credit gets tracked, every reset date is watched, and the per-card ROI is a single number you can look at before the fee posts.

You do not need an app to do the math. You need an honest accounting once a year, before the renewal date, of what you actually used. The point of the app is that you do not have to remember to do it.