How to Sell Trading Cards at Scale: Channels, Fees, and Going From Hobbyist to Dealer

Gross sale price is a vanity metric. What pays your rent is net margin — what lands in your account after marketplace fees, payment processing, shipping, supplies, and your time.

A clean way to think about it: Net = Sale price − marketplace fee − payment processing − shipping − supplies − (minutes spent × your hourly value). Most sellers track the first three and quietly eat the last two. That's a mistake. At volume, a listing that takes you eight minutes instead of two is the difference between a sustainable operation and a second unpaid job.

The other trap is comparing channels on fee percentage alone. A channel with a lower fee but a smaller, slower audience can net you less per card because items sit unsold, get relisted, or eventually sell below market. Liquidity — how fast and how close to market a channel clears your inventory — is part of margin too. The right move is rarely "use the cheapest channel." It's "match each card to the channel where it nets the most, fastest."

Each channel wins on a different axis. eBay has the deepest buyer pool and the most predictable pricing for graded and high-value singles. TCGplayer is purpose-built for TCG singles and bulk, with cart mechanics that move commons efficiently. Whatnot turns selling into live entertainment and can clear large volumes of mid-grade and sealed in a single stream, but the live format and its fee structure mean you're trading margin for speed and reach. Local and LGS sales carry near-zero platform fees but the lowest prices, because the buyer is reselling. Card shows sit in between: a table fee and your weekend, but cash in hand and no shipping.

The table below is a worked example, not a rate card. Fees, promoted-listing rates, and store/seller subscriptions vary by category, country, and current promotions — always confirm the live numbers on each platform before you price. The point isn't the exact percentage; it's the relative shape of the trade-offs.

Read it as: high-value and graded singles lean eBay; TCG singles and bulk lean TCGplayer; fast liquidation of volume leans Whatnot; cash-now-no-shipping leans local and shows. Most serious sellers run two or three of these in parallel rather than betting on one.

For raw cards above a certain value, the most important pre-listing decision isn't where to sell — it's whether to grade first. A graded card in a strong slab can sell for a large multiple of its raw price, but grading costs money and weeks of turnaround, and a low grade can leave you underwater versus just selling it raw.

The logic is straightforward: grading is worth it when (expected graded value × probability of a strong grade) minus grading and shipping costs clearly beats the raw sale price. The hard part is the probability. Centering, surface, edges, and corners decide whether you get a gem or a mid-grade, and eyeballing that on a valuable card is how dealers lose money — submitting cards that come back too low to justify the fee, or selling raw cards that would have graded high.

This is exactly the gap CardGrade is built to close. Photograph the card and it predicts a PSA, BGS, or CGC grade across 47 inspection points in about 60 seconds, at 92.8% accuracy — so you can triage a stack and only pay to grade the cards likely to clear the bar. For the full break-even math, run your specific card through the grading break-even calculator at /tools/grading-calculator, and if you're deciding which company to submit to, the head-to-head at /resources/psa-vs-bgs-vs-cgc-vs-tag breaks down resale liquidity versus subgrades versus cost. The deeper economics of submission live in the grading-ROI pillar at /resources/grading-roi.

At low volume, listing by hand is fine. You photograph a card, look up the price, write a title and description, set a price, and publish. Call it five to eight minutes per card once you include identifying the exact set, edition, and variant.

The problem is that this cost is linear. Two hundred cards a month at six minutes each is twenty hours — half a work week spent on data entry, before a single card sells. And the per-card economics are worst exactly where most volume lives: the $3–$15 commons and mid-value cards where six minutes of labor can exceed the margin on the sale. You end up in a bind — the cheap cards aren't worth listing individually, but bulk-lotting them leaves money on the table.

Three symptoms tell you you've hit the wall: a backlog of unlisted inventory you keep meaning to get to, listings with thin or copy-pasted details because you ran out of time, and pricing that's stale because you set it once and never revisited it. When listing throughput — not sourcing, not demand — is the thing capping your sales, you've outgrown manual.

This is where CardDealer changes the unit economics. Photograph a card and it's identified in about 1.4 seconds, priced on a three-source blend of PriceCharting, Collectr, and TCGplayer data, and turned into a publish-ready listing for eBay, Shopify, or TCGplayer — under a minute per card instead of six-plus. It charges a flat fee and takes 0% of your sales, which matters precisely because it doesn't tax your upside as you scale.

Run the math on the labor line alone. Drop per-card listing time from six minutes to one and that 200-card month goes from twenty hours to roughly three and a half. The freed time is real margin — either more cards listed or fewer hours worked for the same output. And because each listing carries accurate identification and current three-source pricing, you stop bleeding value on the misidentified-variant and stale-price mistakes that quietly erode hand-built listings.

The honest framing: a flat-fee tool pays off above a volume threshold and not below it. If you list a handful of cards a month, keep doing it by hand. If unlisted inventory is piling up and listing is your bottleneck, the per-card cost of automation is a rounding error against the labor and the missed sales. For the full operating model — sourcing, pricing cadence, and channel mix as a business — see the card-business pillar at /resources/card-business, and the channel-by-channel selling playbook at /resources/selling-cards.

You don't have to call yourself a dealer to operate like one, but the line determines your tooling and your channels.

You're operating as a collector if you sell occasionally, mostly to fund the next pickup, and a weekend of hand-listing clears your backlog. Stay manual. Grade selectively — only your standout raw cards, and only after checking the break-even. Sell where the buyers are deepest for that specific card: eBay for graded and high-value singles, TCGplayer for TCG singles, local or shows when you want cash without shipping.

You're operating as a dealer when sourcing volume exceeds what you can list by hand, when listing throughput caps your revenue, and when the same pricing and identification decisions repeat hundreds of times a month. At that point three things change: grading becomes a triage process (predict first, submit selectively, with CardGrade doing the screening); listing becomes a pipeline (CardDealer for sub-minute, accurately priced listings at a flat fee); and channel choice becomes a portfolio (eBay and TCGplayer for steady margin, Whatnot or shows for fast liquidation of volume). The shift isn't about selling more cards — it's about removing yourself as the bottleneck so the operation scales without scaling your hours.