Ramp vs Amex vs Capital One: Corporate Card Math for DTC

Parker was a genuinely good product for direct-to-consumer (DTC) brands. It offered per-transaction payment terms of 15 to 90 days plus up to 2.65% cash-back on Meta and Google ad spend, backed by $157M in funding. Then on May 7, 2026 it filed Chapter 7 bankruptcy, and cardholders watched their ad-spend cards decline mid-campaign. If you ran spend through Parker, you already know. If you did not, the comparison question it was trying to answer is still sitting on your desk: for a brand spending real money on card every month, which corporate card actually costs the least once you net out fees, caps, and float? This post runs that math for a $2M brand putting $60,000/month ($720,000/year) through a card.

Parker filed for bankruptcy, and here is what brands actually felt

Parker's Chapter 7 filing listed assets and liabilities in the $50M to $100M range and 100 to 199 creditors, with failed acquisition talks cited as the proximate cause by fintech analysts. The mechanics of the failure are what matter for your decision. Parker was a fintech riding on a partner bank (Patriot Bank), and when the company collapsed, the banking partner notified customers and the cards went dark. Brands saw Meta and Google billing cards decline within days, which for an ad-heavy DTC business means campaigns throttling or stopping mid-flight.

When I talk to founders running brands at this size, the card is never framed as a risk decision. It is framed as a rewards decision: which card pays the most on ad spend. Parker's failure reframes it. Your card infrastructure is not a commodity, and a 48-hour ad blackout during a promo window costs more than any rewards-rate difference you were optimizing for.

The pattern we see again and again is brands running several cards in parallel and cycling them by billing period, often with ad spend concentrated on one primary card. One finance team we worked with was routing the bulk of its advertising through a single fintech card and moving to the next each cycle. That concentration is efficient until the card stops working. Parker's collapse is the concrete argument for keeping a bank-issued card as your base, which we will get to.

The model: a $2M brand spending $60K/month on card

To compare cards honestly you need a fixed brand to run them against. Ours is a $2M revenue DTC brand putting $720,000/year through a card, which is roughly what an ad-heavy brand at this size looks like once Meta, Google, inventory deposits, and software all route through the card. We model two spend compositions, because the answer changes for some cards. In our model we assume:

• Ad-heavy: 70% advertising ($504K), 20% inventory ($144K), 10% miscellaneous ($72K)
• Inventory-heavy: 20% advertising ($144K), 60% inventory ($432K), 20% miscellaneous ($144K)

Here is the important structural point. For flat-rate cards (Capital One Spark, Wells Fargo Signify, Ramp), the spend composition does not change the answer at all. The rate is constant across every dollar. For category cards (Amex Business Gold, Chase Ink Preferred), composition matters a lot, because the bonus rate only applies to specific categories and only up to a $150K combined annual cap. Past that cap, everything drops to 1%. That single mechanic is why the "best ad-spend card" is rarely the best card once you are spending real volume.

Net-cost math, card by card

Ranked by net annual benefit at $720K spend, the standings are clear, and they are not what most operators expect, and the spread between cards can move the needle on net margin at this revenue level.

Capital One Spark Cash Plus wins at $14,400 net. It earns 2% unlimited with no category caps. The $150 annual fee is refunded once annual spend clears $150K, which happens by late February at this volume, so the effective fee is $0. It is bank-issued, which means near-zero platform risk. This is the no-optimization-required answer.

Wells Fargo Signify Business Cash ties at $14,400 on paper: 2% unlimited, no fee, no cap. The catch is approval. We could confirm the rate but not the revenue and relationship thresholds a $2M brand needs to clear, so treat it as a strong option subject to eligibility rather than a lock.

Amex Business Gold nets $11,325 at cash redemption. It earns 4X (worth 4% at a cent per point) on your top two eligible categories, but the $150K combined cap is hit in under four months for an ad-heavy brand at $40K+/month on ads. After that, everything earns 1X. The math improves sharply if you redeem Membership Rewards points at 1.5 to 2 cents each through airline and hotel transfers, which lifts net benefit to roughly $17,325 to $23,025. Be honest about whether your team actually does that. Most DTC finance teams do not.

Ramp nets $10,800 on the card alone (1.5% flat, $0 fee). But the card rate understates Ramp, and we will return to that.

Chase Ink Business Preferred nets $10,105 at cash value. It earns 3X on social and search advertising, shipping, internet/phone, and travel, inside the same $150K combined cap. At a cent per point it sits below Ramp; redeemed for travel at 2 cents per point it climbs to roughly $20,305 (applying $0.02/pt consistently to all points, the same methodology used for Amex Gold above). It is a good fit for shipping-heavy brands with a team that redeems for travel.

Amex Blue Business Cash ($7,700) and Chase Ink Business Cash ($7,250) are both non-competitive at this volume. Blue's 2% caps at $50K/year, and Ink Cash's bonus tiers cover office and telecom categories that barely register for DTC. These are early-stage first cards, not $720K cards.

Ramp's real value is the software, not the 1.5%

Reading the ranking above, Ramp looks like a middle-of-the-pack card, and on card yield alone it is: $10,800 versus $14,400 for the flat 2% cards, a $3,600/year gap. But that framing misses where Ramp actually earns its place.

Ramp bundles expense management, receipt automation, integrations, and policy enforcement at $0 on the Free tier. If your brand was paying $8K to $15K/year for a legacy expense tool (an illustrative range; your actual cost depends on the tool and team size), Ramp erases that line entirely. Add it back: $10,800 in cash-back plus $10,000 in avoided software cost is $20,800 in total economics, which beats Capital One Spark's $14,400 on a fully-loaded basis. That is the Ramp pitch, and it only works if you were genuinely paying for expense software before.

Two cautions. First, the bill-pay fees are real. Standard ACH runs about $0.59 per transaction, with same-day ACH and wires higher, unless you pay from a Ramp Checking balance. A brand cutting 100 vendor payments a month via standard ACH loses roughly $708/year, trimming Ramp's effective net from $10,800 to about $10,092. Second, Ramp is a fintech riding on a partner bank (Celtic Bank). It is far better capitalized than Parker was, but the category tail risk is not zero, which is exactly why we frame it as the platform layer on top of a bank-issued base, not the base itself.

The float question: what Parker was really selling

The rewards rate was never the whole Parker story. The extended payment terms were, and they are worth putting a real number on. Float value is the working-capital benefit of paying later instead of sooner, and you price it at your effective borrowing cost. The formula is simple: monthly spend times (days of float / 365) times the prime rate.

At $60,000/month and today's 6.75% prime rate (FRED DPRIME, effective December 11, 2025), a standard credit card's roughly 25-day grace period is worth about $3,325/year in float. Parker's net-60 terms were worth about $7,990/year, a $4,665 annual advantage. Net-90 terms were worth about $11,990. That is a real, recurring economic benefit, not a marketing claim.

One nuance worth stating plainly: not all of Parker's terms were free. The shorter rolling card terms were largely interest-free, funded by interchange, while its longer-dated 60 to 90-day credit offering (a separate product from the basic card) may have carried explicit interest or fees. The float above is the value the terms created, not a claim that Parker gave it away for nothing.

When we talk to operators about this, the framing that lands is the one a fractional CFO gave us: right now, cash is the most expensive resource on the balance sheet. When cash is expensive, float on your largest recurring spend is not a rounding error, which is the same logic that drives your DTC working capital playbook. A brand carrying a $40K/month Meta budget on net-60 terms was effectively financing $80K of ad spend at close to zero cost, instead of paying inside 25 days and pulling on a line of credit to cover the gap. That is the benefit that disappeared for DTC brands when Parker failed, and no live bank card replaces it.

Calculator: model your own card spend

Plug in your numbers to see which card wins for your actual spend mix.

Spend-mix tipping points and the platform-risk verdict

So where does each card actually win? A few thresholds hold up across the spend compositions:

• Flat 2% wins at essentially all volumes above $50K/year. The Amex Blue cap wipes out its 2% advantage above that, and no category card sustains a bonus rate past the $150K combined cap for a high-ad-spend brand.
• Amex Business Gold only beats flat 2% if you value points at $0.015+ each and keep bonus-category spend under $150K/year combined (roughly $12,500/month). Above that, the rate reverts to 1X and Gold underperforms.
• Ramp beats bank cards on total economics only if you were paying $8K to $15K/year for expense software. On card yield alone it stays third.
• Chase Ink Preferred beats Ramp for shipping-heavy brands with a team that redeems for travel: 3X on shipping at 2 cents per point is a 6% effective rate against Ramp's 1.5%.

The platform-risk lesson from Parker changes the default. The failure path was mechanical: a bank-partner-dependent fintech collapses, the partner bank notifies customers, cards go dark, and ad-spend cards decline mid-campaign. Bank-issued cards (Amex, Chase, Capital One, Wells Fargo) carry effectively none of that risk. For a $2M DTC brand that cannot absorb a 48-hour ad blackout, the defensible base is a bank-issued flat 2% card. Capital One Spark Cash Plus is the no-optimization winner: $14,400 net, fee refunded, bank-issued. Then add Ramp alongside it for expense management if you were paying for a legacy tool. You get most of the yield, the software, and a card that will not vanish when a startup's funding runs out.

The mistake is optimizing your corporate card for rewards rate the way you would optimize a personal card. At $720K of spend the rewards spread between the best and worst live card is real money, but it is smaller than the float Parker was selling and smaller than the cost of your ad cards going dark for two days. Pick the bank-issued base first, then layer the platform card for software. Yield is the third decision, not the first.

Sources and methodology

Card terms verified from published product pages, July 2026. Rates, fees, and caps were taken directly from each issuer's public materials: Capital One Spark Cash Plus (2% unlimited, $150 fee refunded above $150K spend), Amex Business Gold (4X on top-2 categories, $150K combined cap, $375 fee), and Chase Ink Preferred and Cash (3X and tiered bonuses, $150K and $25K caps). Points are modeled at $0.01 for cash redemption; travel-transfer upside is noted separately where it changes the ranking.

Ramp pricing from published support materials. The 1.5% flat card rate and $0 Free-tier fee are confirmed via Ramp's pricing overview. Bill-pay fees (standard ACH ~$0.59, higher for same-day ACH and wires, waived from a Ramp Checking balance) are drawn from the same source. The marketed "up to 5% savings" is a full-platform estimate and is deliberately excluded from the card-yield model.

Interest rates from FRED. The 6.75% US prime rate used for all float calculations is FRED series DPRIME, effective December 11, 2025 and unchanged as of June 2026, corroborated by the JP Morgan Chase historical prime rate table. The effective federal funds rate (3.63%, June 2026) is FRED series FEDFUNDS.

Parker bankruptcy from dated press. The Chapter 7 filing date (May 7, 2026), asset and creditor ranges, banking-partner detail, and the cause (failed acquisition talks) are reported by TechCrunch, May 9, 2026. Parker's pre-bankruptcy rates and terms are reconstructed from its archived DTC product page and are cited as historical, not current.

Card-market context. Small-business card ownership (~79%) and card share of B2B payment volume (8 to 12%) come from JP Morgan Chase Institute and Clearly Payments B2B statistics, 2026.

Limitations. The model assumes a single $60,000/month spend level and two illustrative spend compositions; your mix will shift the category-card numbers. Wells Fargo Signify approval thresholds and some Parker interest-on-terms details could not be fully verified and are flagged inline. This is a net-cost model, not underwriting advice, and approval for any card depends on your revenue, credit profile, and issuer relationship.

Frequently asked questions