A typical European card user spends roughly $2,400 a month on the small things. Groceries. Transport. Restaurants. The streaming subscriptions and the gym fees and the airline fees that nobody puts on a credit card with a sign-up bonus. Twelve months of that comes out to $28,800 a year. Two of the more visible crypto cards aimed at exactly this kind of spender are the Bybit Card and Gnosis Pay. Their landing pages promise 10% and 5% cashback respectively. Both numbers are technically accurate. Both are also misleading, in completely different ways, and the version of this comparison that floats around on social media usually compares the wrong tiers.

So here is the same exercise done more carefully.

What Bybit actually pays you

Bybit's card has six cashback tiers, named Base, Beta, Alpha, Apex, Omega, and Infinite. Rates run from 2% at the bottom up to that headline 10%. The 10% only kicks in at the Infinite tier, which requires either Supreme VIP status on the exchange or $25,000 a month of card spend. That is not the case for someone running $2,400 a month through a card.

The case that does apply looks like this. A brand-new account with no card history starts at Tier 1 (Base): 2% capped at $5 a month. At $2,400 of monthly spend you would generate $48 in cashback at the headline rate, but the cap clips you to $5. Twelve months of that comes to $60 a year on $28,800 of spend. The effective rate is 0.21%, which is the figure that gets quoted in critical reviews of the card.

The catch is that Tier 1 is, in practice, a new-user state. Tier 2 (Beta) unlocks at either VIP1 trading status on the exchange or, more importantly for our profile, $500 a month of card spend. Anyone sustaining $2,400 a month of card use clears that bar in the second month. Tier 2 is still a 2% rate, but the cap moves up to $50 a month. On $2,400 of spend, 2% × $2,400 = $48, which sits below the $50 cap. The cap stops biting. Annual cashback at Tier 2: $576. Effective rate: 2.0%.

That is the realistic Bybit number for this profile. Not $60.

Bybit pays cashback in USDT. There is no token-price exposure on the rewards themselves. A $576 cashback figure today is a $576 cashback figure when you spend it in February.

What Gnosis Pay actually pays you

Gnosis Pay runs four cashback tiers under a programme called GIP-110, currently funded by the Gnosis DAO and scheduled to run until 30 June 2026. The current top tier (Tier 4) pays 4%, with a separate 1% bonus for holders of the Gnosis Pay OG NFT. Combined, that is up to 5%, which is the figure on the landing page.

Tier 4 does not have a cap that bites at our profile. The weekly cap at Tier 4 is around €2,000, well above the roughly €510 a week that $2,400 a month works out to. So 4% × $28,800 = $1,152 a year, paid weekly in GNO directly into the user's Gnosis Pay Safe.

The catch is on the qualification side. To unlock Tier 4 you need to hold 100 GNO in your Safe. GNO traded at $131.41 on CoinGecko on 8 May 2026, putting the lockup at $13,141 of capital that has to sit in one volatile token. Lower tiers (1, 2, 3) require 0.1, 1, and 10 GNO respectively, with cashback rates of 1%, 2%, and 3%.

Cashback at Gnosis Pay is also paid in GNO, not in stablecoin. That $1,152 nominal figure assumes GNO holds its dollar value through the year. If GNO drops 30% over the year, the realised dollar value of the cashback is closer to $806 by the time a user converts it. If GNO rises 30%, the cashback is worth more, but at that point the user is making money on a token bet, not on card rewards.

The realistic comparison, side by side

The honest version puts Bybit Tier 2 against Gnosis Pay Tier 4, which is the realistic tier on each card for someone spending $2,400 a month.

Bybit Tier 2: $576 a year in USDT, with no capital requirement.

Gnosis Pay Tier 4: $1,152 a year in GNO, with $13,141 of GNO locked in the Safe.

The gross gap is 2x. Not 19x. The naive comparison that puts Bybit Tier 1 ($60) against Gnosis Pay Tier 4 ($1,152) is comparing the worst Bybit tier against the best Gnosis Pay tier, which is not a fair test.

What capital costs

The $13,141 of GNO is real money that could be earning yield somewhere else. A 4% money-market account on the same capital pays $526 a year. So the net Gnosis Pay cashback, against a flat-GNO baseline, is $1,152 minus $526 = $626 a year.

Net Bybit Tier 2: $576.

Net Gnosis Pay Tier 4: $626.

The real-world delta in a flat market is around $50 a year on $28,800 of spend. That is a different conversation from "19 times more cashback".

Now layer in volatility on the GNO position itself. A 30% drop in GNO erases $3,942 of capital, which is several years of cashback gone. A 30% rally adds $3,942 of token gains, but at that point the cashback is a side effect of holding GNO, not a reason to use the card.

The honest framing is that Gnosis Pay's higher cashback is a small reward for accepting concentrated GNO exposure. Bybit's lower cashback is a smaller reward for accepting no token exposure at all. They are paying you for taking on different kinds of risk, not for using the card.

The Bybit tier most reviews skip

There is one more piece that the marketing-style comparison almost never includes. Bybit Tier 3 (Alpha) pays 4%, capped at $150 a month, and unlocks at either VIP3 trading status or $3,500 a month of card spend.

A user running $3,500 a month through the card generates 4% × $3,500 = $140 a month in cashback, which sits under the $150 cap. Annualised: $1,680, paid in USDT, with no capital lockup of any kind.

Gnosis Pay Tier 4 on the same $3,500-a-month spend would also pay 4% × $42,000 = $1,680 a year, but with $13,141 of GNO sitting in a Safe and exposed to token risk for the entire year.

Above $3,500 a month of card spend, Bybit's no-lockup 4% is the structurally cleaner deal. It does not get advertised that way because the card-marketing focus is on the 10% headline, which only ever applies above $25,000 a month.

What you actually decide

The honest answer to "which card is better" is a small decision tree, not a percentage gap.

If you are a brand-new user running $2,400 a month through the card and you have no spare capital to lock up in GNO, Bybit pays you almost nothing in your first month and around $576 a year after Tier 2 unlocks. That is not a reason to choose Bybit over a non-crypto card with a flat 1.5% cashback rate.

If you have $13,000 of capital to park in GNO and you are comfortable holding the token through whatever it does, Gnosis Pay's net cashback is around $626 a year flat-market, with upside on a token rally and meaningful downside on a token drop. The cashback is real money. So is the volatility.

If you spend more like $3,500 to $4,000 a month on the card, Bybit Tier 3 quietly outperforms both of the headline scenarios with no capital requirement. That tier exists because Bybit needs to give frequent card users a real rate, and it does. It is rarely the example people pick when they want to argue for or against the card.

Either way, the headline percentages on either landing page do not survive the math. They survive the marketing.

Why this matters for the rest of the market

The structural lesson here applies to crypto cards more broadly. A cashback rate always comes with at least three other numbers attached: the cap, the qualification requirement, and the asset the cashback is paid in. Card-issuer marketing tends to publish the friendliest of these and bury the others in help-centre articles. The way to compare cards honestly is to fix a realistic spend profile and look at all four numbers together.

If you want to do that exercise across more cards than just these two, we maintain a comparison of cashback rates, caps, and lockup requirements across 141 crypto cards on Sweepbase. Plug in your own monthly spend and see what survives the cap.

Sources

• Bybit Help Center, "Introduction to Bybit Card Rewards" (tier table verified 8 May 2026): https://www.bybit.com/en/help-center/article/Introduction-to-Bybit-Card-Rewards

• Bybit EEA card migration notice (Jan 2026): https://announcements.bybit.com/en/article/important-update-action-required-for-your-bybit-card-effective-jan-1-2026--blt646815aefe28332e/

• Gnosis Pay GIP-110 Cashback Programme: https://help.gnosispay.com/hc/en-us/articles/40288737327252-GIP-110-Cashback-Programme

• Gnosis Pay rewards page: https://gnosispay.com/rewards

• CoinGecko GNO price (8 May 2026, $131.41): https://www.coingecko.com/en/coins/gnosis

• GIP-131 forum proposal (programme continuation): https://forum.gnosis.io/t/gip-131-should-gnosis-dao-continue-and-update-the-gnosis-pay-cashback-programme/10377

• Bybit tier table cross-check: https://kkinvesting.io/en/posts/bybit-card-tutorial/ and https://cryptoslate.com/crypto-cards/bybit-card-review/

Per-tier weekly caps for Gnosis Pay (€200 / €500 / €1,000 / €2,000) come from cryptocardhub aggregating the help-centre. The programme-wide €20K monthly cap is verified directly. The Tier 2 trigger of $500/mo card spend on Bybit is documented in third-party tier breakdowns sourced from the issuer's help-centre, re-check the issuer page before applying these numbers.

Take a simple case. Card A advertises 4 percent cashback in the issuer loyalty token and 0 percent FX. Card B advertises 1 percent cashback in BTC and 1 percent FX. If you spend $1,000 a month, the ranked-list view says Card A wins by $30 a month.

Now read the fine print. Card A 4 percent applies only after staking $4,000 of the issuer token for 180 days. The cashback is paid in that same token, and the published 30-day average price has dropped 18 percent since the staking program launched. The token is also the only path to the 4 percent tier. There is no fiat fallback. The 0 percent FX has a footnote: foreign-currency transactions still pay a 0.9 percent crypto-to-fiat conversion fee at point of sale.

Card B 1 percent BTC cashback is uncapped. The 1 percent FX is what it says. There is no token exposure.

Now the same $1,000 a month, six months in. Card A holder paid roughly $720 in token-price drawdown on the staked position, earned 4 percent of $6,000 in cashback at average price, and lost 0.9 percent on whatever portion was foreign currency. Card B holder paid 1 percent FX on roughly half of spending and earned 1 percent BTC on all of it. The math, with reasonable assumptions, has Card B ahead by a wide margin by month four.

This is not an edge case. It is the modal case for cards advertising any cashback rate above 2 percent. I argued the general version of this in why most crypto card cashback rates mislead readers and have not changed my view. The follow-up question I keep getting is what to sort by then. I answered that one in a methodology note on which metric to compare 139 cards on, and the short version is total cost per dollar spent over six months, including conversion fees, FX, token drawdown if cashback is in a token, and any tier requirements expressed as opportunity cost.

The catalog where I track these numbers across 141 cards is at sweepbase.net/cards. I recently wrote up what I would ask any new crypto card founder pitching me a launch, which lists the five questions that almost always reveal whether the headline numbers will hold up. I also wrote up the operational lessons from running this catalog for three months for the dev.to crowd, which is a different angle on why I trust the underlying numbers.

There are three cards that survive this exercise consistently across a thousand-plus simulation runs at different spend levels. I am not going to name them here because the right answer changes depending on which currency you spend in, which region you can onboard from, and whether you actually want self-custody. The comparison calculator on the site lets you plug in your own monthly spend and see which cards beat which other cards at break-even.

If you remember nothing else from this post, remember the meta-rule: the higher the advertised cashback rate, the more likely it is paid in something other than the currency you spend in. That something is usually an issuer token whose price is correlated with the issuer marketing budget.

The Binance Visa was being recommended in articles published well after Binance shut it down. Cashback rates on Crypto.com posts were stuck at 2021 numbers from before they slashed CRO rewards. Half the "best in Germany" guides ranked cards that don't ship to Germany. One site listed a "Bling Money" card that, as far as I can tell, never launched.

So I started keeping notes in a spreadsheet. The spreadsheet got long. I cleaned it up and put it online.

That’s Sweepbase. It’s a comparison site for crypto debit and credit cards. 139 of them right now, covering the US, Europe, the UK, Canada, Latin America, Asia, the Middle East, and Africa.

What’s in there

For each card I track the things that decide whether you should order it: FX margin, ATM fee, monthly ATM limit, top-up methods, KYC tier, supported chains, cashback math (the real number, not the marketing number), which stablecoins it accepts, and whether your funds stay in your wallet or sit on the issuer’s balance sheet.

There’s a compare table for up to four cards side by side. There’s a calculator that estimates your yearly cost based on how you actually spend. Category pages slice the catalog by region, by feature like no-KYC or self-custody, by issuer type.

No login required. No email gate to read the data.

The data layer

The whole catalog is one CSV file in a public repo. You can read it, audit it, and yell at me if something’s wrong. There’s a “report error” button on every card page that goes straight to my inbox.

I think this matters. Most card comparison sites scrape press releases, which means they reproduce whatever number the issuer wants you to see. When Crypto.com cut staking rewards in early 2024, it took listicles 8 to 14 months to update. The CSV approach forces me to keep one source of truth, and forces anyone reading to see a date stamp.

What I’m honest about

I’m not impartial about every card. Some links are affiliate. The ones that are, I disclose. The cards I rank highest aren’t always the highest-paying affiliates. Crypto.com’s higher tiers pay more than self-custody options like Gnosis Pay, but Gnosis Pay still ranks above them in several of my self-custody categories, because that’s how the data shakes out.

The ratings are calculated, not handed out. The formula is in the open: cashback gives you points, high FX subtracts points, premium tier gives a small bump. The output gets clamped to a 2.3 to 4.9 range so nothing scores 5.0, because nothing is a 5.0.

What I’ll be writing here

This Substack will be the slower-paced companion to the site. Things I’ll actually post:

• When a card changes its terms (rate cuts, region drops, KYC tightening) and what that means in practice

• Deep dives on specific cards I’ve spent time with

• Industry shifts that affect the whole category, like Visa’s policy changes or stablecoin regulation

• Occasional reviews of cards I think are worth ordering, and ones I think aren’t

If that’s useful, subscribe. If not, the site is still free and doesn’t need an email.

sweepbase.net

A card advertises 5% cashback. You read the terms. The 5% only kicks in after you stake 15,000 dollars of the issuer's token for a year. The base cashback is 1%, and even that requires you to receive rewards in the same volatile token that cost the company half its market cap last time the market turned. Your effective rate is closer to 0.4%.

This happens across every major issuer. The numbers on the landing page describe the best-case scenario for a high-spend user who already owns a large position in the issuer's staking token. The rest of us see something else at the register.

On Sweepbase I run a fee calculator that turns those numbers into a real monthly cost. You plug in your spend, your region, and the currency you top up with. The calculator walks each card through the full fee chain. Conversion fees on top-up. Stablecoin-to-fiat conversion at the merchant. Card issuance or monthly membership. ATM withdrawal. Foreign exchange markup, which hides in the mid-market versus card-rate spread even when the marketing says zero FX fee.

The difference between the headline cashback and the real net cost can be twenty or thirty dollars a month for a 2,000 dollar spend. For smaller spenders, cashback often loses to fees entirely. You get paid 8 dollars back and spend 12 dollars in conversion costs you never saw.

A few patterns that I noticed after cleaning up the data.

The best no-fee cards are usually neobank hybrids, not crypto-native products. They have weaker cashback but cheaper conversion, and they tend to pay out in euros or dollars instead of a token.

Self-custody cards are improving but still charge more. The convenience of paying straight from your own wallet comes with higher card issuance fees. For users who value the custody model, it is worth the premium. For users who only want cashback, it is not.

Premium metal tiers almost never pay back their membership fees. To earn 200 dollars a year in extra cashback on a 200 dollar membership, you usually need to spend over 30,000 dollars on the card. Most holders of those cards spend far less.

FX fees are where the biggest surprises hide. A card with a 1% FX fee is not meaningfully worse than a card with 0% FX fee, because the 0% card almost always recovers the spread on the stablecoin-to-fiat conversion. If you ever want to check a card's real FX behavior, run a small test transaction in a foreign currency and compare the charge against the interbank rate that day.

I am keeping the Substack for notes that do not fit on the main site. Research breakdowns, regional deep-dives, occasional vendor updates, and the odd post on the architecture behind the site itself. If this is useful, you can find the full card comparison at https://sweepbase.net and the fee calculator at https://sweepbase.net/calculator. No affiliate links inside the newsletter.