Consignment and resale shop software in Baltimore: the shelf has four owners

The shops this is actually about

Start with the count, because the count is more interesting than the trade usually gets credit for. In the 2023 County Business Patterns file — the Census Bureau's census of establishments with paid employees, which is the caveat that always matters in a trade thick with one-person operations — Maryland has 288 used merchandise stores under NAICS 453310, employing 3,650 people on an annual payroll of $100,153,000. Baltimore City holds 23 of them, with 290 employees and $8,201,000 of payroll.

Twenty-three shops is not a big number, so it is worth putting next to something. Baltimore City has 38 women's clothing stores, employing 400 people on $6,231,000. So the city's resale sector is a little over half the size of its new women's-clothing sector by headcount — and it pays $28,279 per employee against the clothing stores' $15,578. That is a ratio of 1.82, and I want to be fair about what causes it, because the flattering reading is not the whole story: annual payroll per mid-March employee rewards full-time staff, and used merchandise stores include large donation-based operations with warehouse and logistics roles that a boutique does not have. But even discounting for that, the number kills a stereotype. On the Census Bureau's own figures, the average job in Baltimore City resale pays considerably better than the average job in Baltimore City fashion retail.

Baltimore City's twenty-three shops are 7.99% of the state's total against a population share of 9.10% — the city has slightly fewer resale storefronts than its size would predict, which surprised me for a city with this much rowhouse turnover, this many estates, and a vintage scene on two separate corridors. Three Maryland jurisdictions — Cecil, Charles and Somerset counties — report no used merchandise store with a single employee on a payroll at all.

Two thirds of the trade employs fewer than ten people

The size distribution is where the software argument starts. The CBP file breaks establishments into employment-size bands, and Maryland's 288 used merchandise stores sit like this:

Two thirds of Maryland's resale establishments employ fewer than ten people. That matters more here than in most trades, because of what the work actually is: a resale shop's labor is not shelf-stocking, it is adjudication. Somebody has to look at each object, decide what it is, decide what it is worth, decide whether to take it, and record who it came from. A hardware store does that once per SKU and then forever afterward buys more of the same thing. A resale shop does it once per object, permanently, at a rate of a few hundred objects a week, with six people.

There is one more split in that file worth naming, because it changes who your competition is. The CBP state file breaks establishments by legal form of organization, and in Maryland 92 of the 288 used merchandise stores — 31.9% — are non-profits, carrying 1,189 employees and $34,939,000 of payroll, which is 32.6% of the trade's employment and 34.9% of its payroll. Nationally the concentration is far heavier: 6,506 of the country's 19,466 used merchandise establishments are non-profits, and they carry 118,443 of the industry's 222,478 employees. Fifty-three percent of American resale employment sits inside charities. Maryland is unusually commercial by that measure, at a third rather than a half.

I mention it because it shapes the whole market for software. A donation-funded thrift operation and a fifty-percent-split consignment boutique have almost nothing in common operationally: one has no consignor ledger at all and a serious logistics problem, the other has a consignor ledger that is the business and no logistics problem to speak of. Vendors sell one product to both. That is the first place the seams show.

What the platforms charge, and what the meter is counting

On 14 September 2026 I went through the published pricing pages of the vendors a Baltimore consignment, resale or vintage shop would realistically shortlist, plus the general commerce platforms and the marketplaces that are the actual competition for the online half. The specialist tier behaves the way this series has come to expect: about half of it publishes nothing.

resaleworld.com/pricing answers 404, which is notable because Liberty4 Consignment is one of the oldest products in the category. consignpro.com/pricing answers 403. ricochethq.com/pricing and rose-app.com/pricing do not answer at all — no response, not a refusal. consignmentaccess.com/pricing returns a clean 200 and a body of 114 bytes, which is a page in the sense that a closed door is an entrance. bestconsignmentshopsoftware.com/pricing returns a real page with no dollar sign anywhere on it. And traxia.com/pricing — Traxia being the company behind SimpleConsign — returns a long, handsome page whose only numbers are a mock dashboard showing a fictional shop's day: today's sales $4,820, consignor payouts $1,240 queued this week, average ticket $38.40. You are shown a picture of your own business instead of a price.

Two vendors publish complete ladders, and between them they demonstrate the single most important fact about buying software for this trade.

A note on that Shopify row, because it is a methodology point rather than a criticism. shopify.com/pricing served a 200 and a complete plan ladder — in euros. The page geolocates, and a check run from a US machine through a plain HTTP client can come back with a European price card. I am not going to print dollar figures taken off a page that showed me euros. What is quotable, because it is written in dollars on the same page and applies everywhere, is the marketplace line: 50 synced marketplace orders per month free, and over 50, “1% up to $99 USD/mo.” Hold on to that one. It is the only place in this teardown where a platform meters the exact thing a resale shop does most.

The subscription is the cheap part. The payouts are not.

Here is the finding, and it is the reason I would tell an owner to ignore the sticker entirely.

ConsignCloud's Basic plan is $139 a month, which is $1,668 a year, and it is the cheapest real product in the category. Listed inside that plan, without fanfare, is the line “ACH Payouts ($2.50/payout).” A payout is what happens when you pay a consignor. That is not an edge case in this business; it is the business. It is the one transaction that distinguishes a consignment shop from a store.

SimpleConsign, which is the more expensive product on the shelf, includes ACH consignor payments from its $259 Standard tier upward, with no per-payout charge published anywhere on the page. So the two ladders cross, and you can compute exactly where. ConsignCloud Basic plus payouts equals SimpleConsign Standard at 576 payouts a year — 48 a month. It equals SimpleConsign Professional at 1,056 a year, or 88 a month.

Read that again with a real shop in mind. At forty-eight consignor payments a month, the $139 product costs more than the $259 product. Forty-eight is a small shop. A boutique on The Avenue with four hundred consignors on the books and a monthly payout run is not near that line, it is nine times past it. The cheapest sticker in the category is, for most of the businesses it is sold to, the most expensive stack in the category — and the pricing page is not lying to anybody, because the fee is printed. It is just printed next to a feature name instead of next to a volume.

I want to be fair to ConsignCloud here, because the product is well regarded and the transparency is real: it publishes a complete ladder, it publishes the payout fee, it does not gate its price behind a demo, and three of its competitors will not tell you anything at all. The criticism is not of the vendor. It is of a meter that is indexed to the number of people you owe money to, in a trade whose entire growth strategy is signing up more people you will owe money to.

The introductory discount is metered in your sales

SimpleConsign's launch offer is a genuinely new shape, and I have not seen it anywhere else in two years of these teardowns. Professional is $359 a month; new customers pay $99. The fine print, in the vendor's own words: “Intro pricing applies for 6 months or until $50K in sales (whichever comes first); standard plan pricing applies thereafter.”

The discount is worth $260 a month. The clock that spends it is not a calendar — it is your till. The two conditions coincide at exactly $8,333 a month of sales, which is $100,000 a year. Below that you get the whole six months and the whole $1,560. Above it, every extra dollar you ring up shortens your own discount.

The same structure applies to the vendor-mall side of the ladder, where Vendor Mall Plus is $379 a month and the introductory rate is $199, on the identical six-months-or-$50,000 condition. The discount there is $180 a month, and the $50,000 clock runs at exactly the same speed — which is to say faster for a mall with forty booths than for a boutique with one till.

A welcome offer that gets smaller the better your store does is a defensible piece of commercial design — it is a trial, not a subsidy, and a shop doing $50,000 a month does not need six months to decide. But it is worth naming plainly, because nobody models it: the strongest shop in the cohort receives one sixth of the discount the weakest one does, and the thing that takes it away is the only thing the shop is trying to do.

The part no national platform models

Everything above is about money. This section is about ownership, and it is the reason I picked this trade. A restaurant owns its food. A hardware store owns its bolts. A resale shop stands in a building full of objects and, for a large fraction of them, the honest answer to “is this yours?” is it depends what you mean, and on which statute you are asking under.

Maryland answers that question in at least four places, and they do not agree, because they were not written to. What follows is how we read them when we build for this trade. It is not legal advice, and every one of these is worth twenty minutes with your own attorney.

Four conditions, and a consignment shop may fail the third one simply by existing

The Uniform Commercial Code, as enacted in Maryland, defines the word. Here is §9–102(a)(20) of the Commercial Law article, in full:

“Consignment” means a transaction, regardless of its form, in which a person delivers goods to a merchant for the purpose of sale and:
(A) The merchant: (i) Deals in goods of that kind under a name other than the name of the person making delivery; (ii) Is not an auctioneer; and (iii) Is not generally known by its creditors to be substantially engaged in selling the goods of others;
(B) With respect to each delivery, the aggregate value of the goods is $1,000 or more at the time of delivery;
(C) The goods are not consumer goods immediately before delivery; and
(D) The transaction does not create a security interest that secures an obligation.

Four conditions, joined by “and.” All of them must hold, at the moment of delivery, for the delivery to be a “consignment” in the sense Article 9 uses the word. And look at what condition (A)(iii) is.

It is not a fact about the goods. It is not a fact about the contract. It is a fact about what other people believe about your business. A shop that is generally known by its creditors to be substantially engaged in selling other people's goods is not, for these purposes, taking goods on consignment — which means Article 9 steps back, and ordinary law of agency and bailment decides who owns what. The statute's own definition excludes the business whose sign says “Consignment.”

That is not a drafting accident, and it is not as absurd as it first sounds. The whole point of the Article 9 consignment rules is to protect a creditor who lends against inventory it can see, on the reasonable assumption that the inventory belongs to the borrower. If every creditor in the room already knows the borrower sells other people's things, the assumption fails and the protection is unnecessary. It is a rule about appearances, in a statute otherwise obsessed with filings.

But as a software problem it is remarkable, because it is a condition no database can hold. “Is this shop generally known by its creditors to be substantially engaged in selling the goods of others?” has no column type. It is not knowable at intake, it can change over time, it is resolved after the fact by a court looking at what your lenders actually understood, and it silently governs the legal character of every row you created while it was true.

One thousand dollars, per delivery

Condition (B) is the one an owner can actually act on, and it is the one I would put on the intake screen tomorrow. “With respect to each delivery, the aggregate value of the goods is $1,000 or more at the time of delivery.”

Not per year. Not per consignor. Not per item. Per delivery. A local candle brand that walks in with $1,200 of stock in one trip has made a delivery that satisfies (B). The same brand bringing $600 on Tuesday and $600 on Thursday has made two deliveries, neither of which does. Same goods, same people, same shelf, same month — different legal character, decided by how many times somebody came through the door.

Every consignment system I have looked at models this as consignor → items. The delivery, the thing the statute asks about, is not an entity at all. It exists as a creation timestamp on a batch of item rows, if you are lucky, and as nothing whatsoever if the shop enters stock as it prices it rather than as it arrives. The fix is trivial and nobody has done it: make the intake event a first-class object, give it a total, and let it answer the only question the statute asks.

“Consumer goods immediately before delivery”

Condition (C) removes most of what a classic consignment boutique actually sells. “Consumer goods” is defined at §9–102(a)(23) as goods “used or bought for use primarily for personal, family, or household purposes.” A neighbor's coat, her dining chairs, the children's skis: all consumer goods, immediately before delivery, and therefore outside the Article 9 definition of consignment entirely.

So the shape of the trade determines which regime you are under. A shop taking clothes from households is largely outside Article 9. A shop taking new stock from small brands and makers is largely inside it. A shop that does both — which is most of them — is running two legal regimes across one table, and the column that decides which is not price, category, or split percentage. It is what this object was the day before it arrived, which is a fact about its previous owner's use, and which nothing captures.

If it is a consignment, the law says the goods are yours

Now the consequence, and it is blunt. §9–319(a):

Except as otherwise provided in subsection (b), for purposes of determining the rights of creditors of, and purchasers for value of goods from, a consignee, while the goods are in the possession of the consignee, the consignee is deemed to have rights and title to the goods identical to those the consignor had or had power to transfer.

Read it slowly. For the purposes of your creditors, the law deems you to own the goods on your floor. Subsection (b) restores the consignor's position only where the consignor holds a perfected security interest that would beat the creditor. Absent that filing, a lender with a blanket lien on your inventory, or a judgment creditor executing against it, is looking at property the statute has handed you.

Nobody in this trade thinks of it that way. The conversation at intake is about the split and the price drop schedule. The thing nobody says out loud is that if the shop goes under, the artist's forty pieces may be inventory of the estate.

The consignor's fix is a letter to your bank, sent before the box is opened

The Code does provide a route, and it is worth walking through it, because the route is the argument. §9–103(d) classifies what the consignor has:

The security interest of a consignor in goods that are the subject of a consignment is a purchase-money security interest in inventory.

And a purchase-money security interest in inventory has the most demanding priority rules in Article 9. §9–324(b) requires all four of the following before that interest beats an earlier secured party:

(1) The purchase-money security interest is perfected when the debtor receives possession of the inventory; (2) The purchase-money secured party sends an authenticated notification to the holder of the conflicting security interest; (3) The holder of the conflicting security interest receives the notification within five years before the debtor receives possession of the inventory; and (4) The notification states that the person sending the notification has or expects to acquire a purchase-money security interest in inventory of the debtor and describes the inventory.

Translate that into the life of a potter in Remington. To be safe when she leaves twelve pieces with you, she must: identify whether your shop has an existing secured lender; file a financing statement naming you, before you take possession; and send that lender a written notice describing the goods, which the lender must actually receive, at some point in the five years before the pieces come through your door. §9–505 at least spares her the vocabulary — it lets a consignor file using the words “consignor” and “consignee” instead of “secured party” and “debtor” — but the work is the same.

A Maryland UCC–1 financing statement costs $25 for eight pages or fewer, plus the $4.50 filing service fee the Department of Assessments and Taxation added in December 2024. Call it $29.50. Against condition (B)'s minimum qualifying delivery of $1,000, the filing is 2.95% of the goods it protects. A maker who consigns to eight shops around the city files eight statements, at $236, and has to repeat the exercise every time a shop refinances.

Approximately none of this happens. In a decade of looking at these businesses I have never seen a consignor's UCC–1 in a shop's file, and I have never seen a consignment agreement that mentions one. That is not a scandal — the risk is remote and the paperwork is real — but it is precisely the sort of thing a system should be able to tell you in one query: of the objects on my floor right now, which ones are exposed, and to whom do I owe an explanation if they are seized?

And then there is fine art, where the answer flips completely

Maryland has a second consignment statute, in an entirely different title, and it points the opposite way. Commercial Law Title 11, Subtitle 8A is called “Consignment of Works of Fine Art,” and it is four short sections that most resale owners have never heard of. They should have, because of one word in the definitions.

§11–8A–01(e) defines a work of fine art as an original art work which is a visual rendition (painting, drawing, sculpture, mosaic, photograph), a work of calligraphy, a work of graphic art (etching, lithograph, offset print, silk screen), a work in mixed media, or — at paragraph (4) — this:

A craft work in materials including clay, textile, fiber, wood, metal, plastic, or glass.

That is not a narrow category. That is the entire maker section of every consignment shop in Hampden, Station North and Federal Hill: the hand-thrown mugs, the woven scarves, the turned bowls, the forged hooks, the fused-glass dishes, the resin trays. If the person who brought it in is the person who made it, and you are selling it on commission, then §11–8A–02 applies:

If an art dealer accepts a work of fine art on a fee, commission, or other compensation basis, on consignment from the artist: (1) The art dealer is, with respect to that work of fine art, the bailee of the artist; (2) The work of fine art is bailment property in which the art dealer has no legal or equitable interest until the work is sold to a bona fide third party; and (3) The proceeds of the sale of the work of fine art are bailment property in which the art dealer has no legal or equitable interest until the amount due the artist from the sale, minus the agreed commission, is paid.

And §11–8A–04, which is one sentence long:

Property that is bailment property under this subtitle is not subject to the claims, liens, or security interests of the creditors of an art dealer.

So here is the shape of the thing. Two objects on one shelf, both brought in by the people who own them, both on a fifty-fifty split, both priced at sixty dollars. One is a hand-thrown stoneware bowl from the potter who made it. The other is a factory stoneware bowl somebody found at an estate sale. Under Subtitle 8A the first is bailment property and your creditors cannot touch it. Under §9–319(a), if the delivery met all four conditions, the second may be deemed yours for exactly that purpose. Same shelf, same day, same money, opposite answers — and the deciding fact is who made it and whether that person is standing in front of you.

There is a further twist in §11–8A–03 that I find genuinely startling, and it is the cleanest example I know of a legal rule that a product table cannot express:

Notwithstanding the subsequent purchase of the work of fine art by the art dealer directly or indirectly for the art dealer's own account, a work of fine art that is bailment property when initially accepted by the art dealer remains bailment property until the purchase price, minus the agreed upon commission, is paid in full to the artist.

You can buy the piece yourself and still not own it. Ownership does not follow the sale; it follows the payment. In database terms, the transition out of “bailment” is not triggered by the transaction, it is triggered by the settlement — which in almost every system in this category happens days or weeks later, in a different module, in a batch.

The money in the drawer is not yours either

The clause in §11–8A–02(3) deserves its own paragraph, because it is the one with an operational consequence every single day rather than only in a bankruptcy. The proceeds of the sale are bailment property in which the dealer has no legal or equitable interest until the artist is paid.

Every consignment shop in the country runs one bank account. Sales go in; rent, wages and consignor payouts go out. On the reading above, some fraction of that balance — the artists' share of every craft-work sale not yet settled — is property in which you have no interest at all. It is not a debt you owe. It is somebody else's money sitting in your account.

I am not going to tell you to open a trust account; that is a conversation for your accountant and your lawyer, and plenty of well-run shops manage the risk with nothing more than discipline. What I will say is that a business in this position should be able to answer one question instantly, and almost none of them can: how much of today's balance is not mine? That is a single query over settled-but-unpaid consignor liability, split by whether the item was a craft work from its maker. It is about forty lines of code in a system that was designed for it, and it is unbuildable in a system where the consignor ledger is a report rather than a table.

You collect a tax the owner would never have owed

Move from ownership to tax, and Maryland has a regulation that is short enough to quote whole. COMAR 03.06.01.27, “Auctioneers, Agents, Brokers and Factors”:

Every factor, auctioneer, broker, or agent acting for any principal, or entrusted with possession of any bill of lading, custom house permit, or warehouse receipt for delivery of any tangible personal property or entrusted with possession of any personal property for the purpose of sale shall be responsible for the proper collection and remittance of the tax with respect to these sales, regardless of the fact that the principal or owner of the property being sold would not have been liable for collection of the tax if they had made the sale themselves.

That final clause is the whole story of consignment retail in one line. Your neighbor selling her own sofa to somebody off a classified listing is not collecting Maryland sales tax on it. Put the identical sofa on your floor and the sale becomes a taxable retail sale, because you are entrusted with possession for the purpose of sale. The shelf is what makes it taxable. Nothing about the object changed, nothing about the seller changed, and the buyer now pays six percent more.

This is, incidentally, the cleanest answer I know to the owner who asks why a customer can get the same thing cheaper on a peer-to-peer app. Part of the answer is the tax, and the tax is a consequence of the shop existing. The other part is that on a marketplace the tax is collected too — by the facilitator — but it does not feel like your price, and it does not appear on your shelf edge.

Seventy years, and the existence of a magazine

There is a second tax regulation that the antique and collectible end of this trade lives inside without knowing it. COMAR 03.06.01.14 governs resale certificates, and it opens with two definitions written specifically for this business:

“Antique” means any item of personal property that: (i) Was made in an earlier period; (ii) Has special value because of its age; and (iii) Is at least 70 years old.

“Used collectible” means an item of personal property that has been previously sold at retail as a new item and in which there is a general interest in collecting the item. A general interest in the collection of an item is demonstrated by sales of the item at antique and collectible shows and shops or the existence of associations or publications dedicated to the collection of the item.

Two tests, two different kinds of fact. The first is arithmetic: is this object at least seventy years old? As of 2026 that means made in 1956 or earlier, and it moves every January. The second is not arithmetic at all: does a club or a magazine exist for this category of thing? Maryland has written the existence of a collectors' publication into its tax regulations as a test of fact.

Why it matters commercially: section D of the same regulation lets a buyer of an antique or used collectible give you a valid resale certificate without a Maryland sales and use tax registration number, if they are an out-of-state vendor and supply another state's registration number and license — or, if their state has no sales tax, a copy of a trader's license. That is the rule that makes it possible to sell across a dealer table at a show without charging tax. And section C is the limit that catches people: a vendor may not accept a resale certificate for a purchase of less than $200 paid by cash, check, card or electronic transfer, unless the goods are delivered to the buyer's own retail place of business.

So inside a single antique-mall transaction there are now two dollar thresholds doing different jobs — $1,000 deciding whether Article 9 calls the delivery a consignment, and $200 deciding whether you may accept a resale certificate at all — plus a seventy-year age test and a question about whether a hobbyist association exists. A product table with name, price and category cannot answer any of them.

The four-condition thrift shop

One more, for the third of Maryland's trade that is charitable. The default rule is in COMAR 03.06.01.22C(1), and it is not what most volunteers assume: “Generally, a sale of tangible personal property, a digital code, a digital product, or a taxable service by an exempt organization is subject to tax.” Being a non-profit exempts your purchases, under section B. It does not exempt your sales.

Then come the exceptions, and one of them is written for exactly one kind of store. Sales are not subject to the tax when they are:

Sales made by an independent hospital thrift shop that is operated by all volunteer staff, sells only donated articles, contributes the profits from sales to the hospital with which the shop is associated, and is not operated in conjunction with a gift shop or other retail establishment.

Four conditions joined by “and,” exactly like the UCC definition, and every one of them is an operational fact rather than a status. Hire one paid clerk and the first fails. Accept one consigned item, or buy one case of new tote bags to sell at the register, and the second fails. Share a counter with the hospital's gift shop and the fourth fails. There is also a separate exception at C(2)(a) for “sales made by a bona fide church or religious organization when made for the general purposes of the organization,” which is the rule most parish thrift shops are actually relying on.

The software point is small and sharp: whether a given sale is taxable can depend on a fact about how the shop was staffed that day and where the stock came from. Those are fields. They belong on the item and on the shift, not in a volunteer coordinator's memory.

The e-commerce half, which is where the trade is actually growing

Every post in this series has an online half, and in most trades it is the optional part. Here it is not. A resale shop's catalog is the most search-friendly inventory in local retail — every item is a specific, named, wanted thing — and it is simultaneously the hardest catalog in local retail to put online, for one structural reason.

Every row is a quantity of one

A hardware store's online store sells a 10 mm bolt. If two people buy one, that is fine; there are four hundred in the bin. A resale shop's online store sells that jacket. If two people buy it, one of them is getting an apology.

That single fact breaks more assumptions than anything else in e-commerce software. Inventory sync that runs every fifteen minutes is a fifteen-minute window in which you can sell the same coat twice. Cross-listing — putting the coat on your own site, on eBay, and on a peer-to-peer app at once, which is what the trade is told to do — multiplies that window by the number of channels. Abandoned-cart recovery emails invite somebody back to buy a thing that no longer exists. Reserve-on-add-to-cart, which nearly every platform treats as an optional setting for high-demand drops, is the only correct default in this trade and is almost never on.

And then the shop floor, which no marketplace models at all: the coat is also hanging on a rail, where a customer can carry it to the counter while its online listing is live. The correct behavior — a single stock ledger, locked at add-to-cart across every channel including the physical one, released on abandonment — is not a feature you can buy as an add-on. It is an architectural decision that has to be made at the beginning, which is exactly why platforms built for repeat SKUs cannot retrofit it.

This is also the honest reason so many shops stay off the internet, and it is worth saying to any owner who has been made to feel behind: the reason your online store keeps embarrassing you is not that you are bad at it. It is that you were sold a tool built on the assumption that inventory is a number, when in your business inventory is a set of objects.

The recall list is free, machine-readable, and the duty is already yours

Here is the federal layer, and it is the one I would want any resale owner in Baltimore to read twice. 15 U.S.C. §2068(a)(2) makes it unlawful for any person to sell, offer for sale or distribute in commerce any consumer product that is, at subparagraph (B):

subject to voluntary corrective action taken by the manufacturer, in consultation with the Commission, of which action the Commission has notified the public or if the seller, distributor, or manufacturer knew or should have known of such voluntary corrective action.

“Knew or should have known.” That is the standard, and it applies to a person selling a used object exactly as it applies to a manufacturer selling a new one. The Consumer Product Safety Commission has said for two decades that its rules reach thrift stores, consignment stores, charities and flea markets.

What has changed, and what almost nobody in the trade has noticed, is that the list is now a free, public, machine-readable file. The Commission publishes every recall it has ever issued as JSON at saferproducts.gov, with no key and no rate limit worth mentioning. I downloaded the whole thing on 14 September 2026 and counted it.

The number that reframes the problem is 4.6%. Only 458 of the 10,002 recalls carry a UPC at all, and across fifty-three years the file contains just 1,427 distinct UPC strings. So the standard advice — scan the barcode, check the recall — fails in more than nineteen cases out of twenty, and most donated goods have no scannable barcode in the first place. The match that would actually work is on brand, model, description and period: a drop-side crib, a particular jogging stroller, a style of window blind with a looped cord. That is a text-matching problem, which is to say it is now a cheap problem, and it is the single clearest case I have seen in this series for putting a language model behind an intake screen rather than in a marketing email.

One more artifact from that file, which I did not go looking for. Seventeen recalls carry this exact remedy text:

Firm no longer in business. Recall remedy no longer available. Discard the product. Do not donate or resell.

Fourteen more carry a variant of it. That is a federal agency writing an instruction addressed to precisely two audiences: people cleaning out a house, and the shop they are about to bring the box to.

The fee is charged on the tax

Now the marketplace arithmetic, because for most Baltimore resale shops eBay is not a channel, it is the second half of the business. eBay publishes its rates in full, which I appreciate, and one sentence in that page is worth more attention than it gets:

The total amount of the sale includes the item price, any handling charges, any shipping costs collected from the buyer (some exceptions apply), sales tax, and any other applicable fees.

The final value fee is a percentage of a total that includes the sales tax eBay itself collected and remitted as marketplace facilitator. eBay's own worked example on that page uses a six percent rate, which is Maryland's: a $400 item, $24.00 of tax, a $424.00 total, a final value fee of 13.6% plus the $0.40 per-order fee, which is $58.06. Of that, $3.26 is charged on the tax — money that was never yours and never theirs.

Six percent of a price you set, then thirteen point six percent of the six percent. It is not a large number on one sale. On $36,000 a year of eBay sales, if every order carried Maryland's six percent, it is about $294 — two months of your consignment software, paid as a commission on somebody else's tax.

The cheapest thing in the shop pays the highest rate

The other thing eBay's rate card reveals is that the fee percentage moves in the opposite direction to the value of the goods, and the resale trade sits at the wrong end of it.

A six-dollar paperback — the most characteristic object in American resale — pays an effective 20.3%. A five-thousand-dollar ring pays 15.0%. The rate is not merely regressive by accident of the flat fee; the headline percentage for books, film and music is itself 15.3%, the highest general rate on the card, higher than jewelry's.

One cent, at a hundred and fifty dollars

And then there is the athletic-shoe rule, which is the most consequential single line in the whole card for anybody who takes in sneakers. eBay charges 8% at $150 and above, with no per-order fee, and 13.6% below $150.

The effective rate falls from 13.9% to 8.0% across one cent. Which means any pair you would have priced in the high $130s or the $140s belongs at exactly $150.00, and a shop that lists a hundred pairs a year in that band is leaving something close to $880 on the table by not knowing it. This is not a legal subtlety or a compliance risk. It is a pricing rule, published by the marketplace, that no consignment system in the category will enforce for you — and enforcing it is one conditional in a pricing function.

Two reporting thresholds, ten times apart

Last piece, and it is fresh enough that most accountants are still writing about it. On 4 July 2025 the reconciliation act at Public Law 119–21 changed both of the information-reporting thresholds that a consignment shop lives between, and it moved them in opposite directions.

Section 70433 deals with the form you file when you pay a consignor:

(a) In General.—Section 6041(a) is amended by striking “$600” and inserting “$2,000”. … (f) Effective Date.—The amendments made by this section shall apply with respect to payments made after December 31, 2025.

So calendar year 2026 — the one running now — is the first year at the higher figure, reported in early 2027. Two details inside that section matter to anybody who has to implement it. Subsection (b) indexes the threshold for inflation for calendar years after 2026, rounded to the nearest $100, so it will not stay at $2,000. And subsection (e)(2) strikes “taxable year” from §6041(a) and inserts “calendar year,” which means a shop with a June fiscal year now aggregates its consignor payments on a calendar basis regardless of its own books.

Section 70432, four paragraphs earlier, deals with the form a marketplace files when it pays the same person:

A third party settlement organization shall be required to report any information under subsection (a) with respect to third party network transactions of any participating payee only if—(1) the amount which would otherwise be reported under subsection (a)(2) with respect to such transactions exceeds $20,000, and (2) the aggregate number of such transactions exceeds 200.

That restores the pre-2021 rule, retroactively. Put the two next to each other and the asymmetry is stark: you must report a consignor once their calendar-year payments pass $2,000; a peer-to-peer marketplace must report the same seller only once they pass $20,000 and more than two hundred transactions. Ten times the money, and a transaction count on top.

I am not going to pretend this is an injustice; the two provisions are aimed at different problems and the marketplace rule has its own long political history. But it is a real competitive fact, and it is worth understanding before you assume a consignor left you for a better split. Some of them left because your channel reports them and the other one does not.

The build consequence is specific and unglamorous: your payout ledger has to aggregate per consignor, per calendar year, against a threshold that changed on 1 January 2026 and will move again in 2027, and it has to have collected a taxpayer identification number from anybody who might cross it — ideally at intake, when the person is standing in front of you, rather than in February, when they are not.

What custom actually costs

We publish our prices for the same reason I have just spent several thousand words on other people's: a number you can put in a budget is worth more than a demo. These are fixed, not estimates, and they are the same numbers on our pricing page.

Now set that against what the stacks cost. I have modeled one shop — 450 active consignors, $480,000 a year of sales, of which $96,000 is online: $60,000 through its own storefront and $36,000 on eBay at an average order of $42 — assembled two different ways from the prices above.

The $139 product produces a $21,355 stack. The $359 product produces a $10,495 stack. The cheaper subscription is 2.03 times the more expensive one, a difference of $10,860 a year, and the entire gap is the payout meter.

Against those, the paybacks are honest rather than flattering. On Stack A, a $6,000 online store is paid back in about five months and a $12,000 build in about ten. On Stack B — the shop that is already on the better-shaped contract — the same $6,000 build takes about twenty months and the $12,000 build about forty. That second column is a real answer, and it is the answer more often than owners expect: if your consignor payments are already included in a flat subscription, the financial case for replacing your consignment system is weak, and the case for owning only the storefront is the one to look at.

What we would actually build

The design follows from everything above, and it starts with one decision that sounds pedantic and is not: the unit of this business is an object, not a product, and the object has a provenance. Four things carry structure a generic retail schema does not give them.

An intake event is a first-class record, not a timestamp. It knows who delivered, when, what the aggregate value of that delivery was, and what the goods were immediately before it — household use, studio stock, wholesale inventory, donation. Those four facts are exactly the inputs to §9–102(a)(20), and recording them costs nothing at the counter because somebody is already typing a name and a number. It is the cheapest compliance artifact in this entire article, and it does not exist in any product I have looked at.

An object carries its own ownership state and its own settlement state, separately. Consigned, donated, bought outright, bailment property, sold-but-unsettled, settled. The distinction between the last two is the one §11–8A–03 forces, and it is the difference between a balance you can spend and a balance you are holding. From that one pair of fields you get the query that no shop in this trade can currently run: how much of today's cash is not mine, and to whom.

A listing is a projection of an object onto a channel, with a single stock ledger underneath and a reservation lock that every channel respects — including the counter. One object can have five listings; it cannot have five buyers. That is the whole of the cross-listing problem, solved once, at the beginning, instead of patched forever with sync jobs.

A consignor is a party with an agreement, a split schedule, a payout method, a calendar-year running total and a taxpayer identification number captured on day one. Payouts are a first-class ledger with their own history, not a report that is regenerated, because the question people actually ask is “what did you pay me in March” and the answer has to be the same in June.

On top of those, the two screens that earn their keep. The first is intake: photograph an object, and have the system propose a title, a category, a condition, a comparable price from your own sales history, and a recall screen against the CPSC file — matched on brand, model and description rather than a barcode that four point six percent of recalls carry. The second is pricing, which knows the rules that are otherwise in somebody's head: the price-drop schedule, the split, the athletic-shoe cliff at $150, the $200 resale-certificate floor, and the seventy-year line that moves every January.

None of this is exotic engineering. It is ordinary software with an honest data model, and the reason it does not exist off the shelf is that a vendor selling one product to a donation-funded warehouse operation and a four-hundred-consignor boutique cannot afford to model either one properly.

Build, or keep paying

I would rather you kept a subscription that works than paid us to rebuild it. The honest test is short, and it is the only list in this article:

• Keep buying when your consignor payments are already inside a flat subscription rather than metered per payout, when your online sales are a small tail on a healthy shop floor, when nobody in the building is retyping the same object into two systems, and when your stock is nearly all donated or bought outright so the ownership questions above simply do not arise.
• Build when you are paying a per-payout fee on hundreds of consignors a month, when you have sold the same item twice on two channels in the last quarter, when you cannot say what share of your bank balance is consignor money, or when a maker has asked you a question about what happens to her work if the shop fails and you did not have an answer.

Most shops we talk to land in the middle: keep the point of sale, keep the accounting, keep the card processing, and own the object record and the storefront — which is where this trade's actual peculiarities live. That is a $6,000 to $12,000 decision, not a rip-and-replace.

When you should not call us

If you run a donation-funded thrift operation with no consignors, your problem is logistics and volunteer scheduling, and there are good, cheap tools for both; we would not be the right spend. If you are a single-person vintage seller doing everything through one peer-to-peer app, stay there — the marketplace's audience is worth more to you than any site we could build, and we will say so. If your shop floor is losing money, software will not fix the rent. And if you are three weeks into a migration onto a system you have already paid to configure, finish it; a half-migrated consignor ledger is worse than either end state, and in this trade it is worse in a way that involves phoning four hundred people.

How we work

Fixed price, agreed before we start. You talk to the people writing the code — there are two of us and no account layer. You own every line, in your repository, on your infrastructure, from the first commit. We ship in weeks, not quarters, and we start with the one screen that is costing you the most, which in this trade is almost always intake or settlement. If a week of work would tell us both whether the rest is worth doing, that is what the $3,500 Prototype Sprint is for, and it comes off the price of the build.

Questions we get from resale and consignment shop owners

Do consigned goods in my shop belong to me or to the consignor?

It depends on which of Maryland's rules the delivery falls under, and the answer can change from one item to the next. Under Commercial Law §9–102(a)(20) a delivery is a consignment for the purposes of secured transactions law only if four things are all true at once: the merchant deals in goods of that kind under another name, is not an auctioneer and is not generally known by its creditors to be substantially engaged in selling the goods of others; the aggregate value of that delivery is $1,000 or more at the time of delivery; the goods were not consumer goods immediately before delivery; and the transaction does not create a security interest securing an obligation. If all four are true and the consignor has not perfected, §9–319(a) deems you to have rights and title identical to the consignor's for the purposes of your creditors. If the consignor is an artist and the object is a work of fine art, §11–8A–02 makes you a bailee instead and §11–8A–04 puts the object beyond your creditors entirely. This is how we read it for the shops we build for, and it is not legal advice.

What is the $1,000 consignment threshold in Maryland?

Maryland Commercial Law §9–102(a)(20)(B) requires that, with respect to each delivery, the aggregate value of the goods is $1,000 or more at the time of delivery. It is a per-delivery test, not an annual or per-consignor one. The same maker bringing you $1,200 of stock in one trip has made a consignment under Article 9; the same maker bringing you $600 on Tuesday and $600 on Thursday has made two deliveries, neither of which clears the threshold. Nothing in a standard consignment system records the delivery as an object with its own total, which is exactly what the test asks about.

Does Maryland treat handmade craft differently from other consigned goods?

Yes, and much more broadly than most owners expect. Commercial Law §11–8A–01(e) defines a work of fine art to include, at paragraph (4), “a craft work in materials including clay, textile, fiber, wood, metal, plastic, or glass.” When an art dealer accepts such a work on consignment from the artist on a commission basis, §11–8A–02 makes the dealer the bailee, makes the work bailment property in which the dealer has no legal or equitable interest until it is sold to a bona fide third party, and makes the proceeds of the sale bailment property too until the artist is paid. §11–8A–04 puts bailment property beyond “the claims, liens, or security interests of the creditors of an art dealer.” A hand-thrown mug and a mass-produced mug can sit on the same shelf under opposite rules.

Do I charge Maryland sales tax on consigned items?

Yes, and COMAR 03.06.01.27 says so in a way worth reading closely. Every factor, auctioneer, broker, or agent acting for any principal, or entrusted with possession of any personal property for the purpose of sale, is responsible for the proper collection and remittance of the tax on those sales, “regardless of the fact that the principal or owner of the property being sold would not have been liable for collection of the tax if they had made the sale themselves.” A neighbor selling her own sofa privately is not collecting Maryland sales tax. The same sofa on your floor is a taxable retail sale because you are the one entrusted with it. Your shelf converts a private transfer into a taxable one.

What counts as an antique for a Maryland resale certificate?

COMAR 03.06.01.14A(2)(a) defines an antique as an item of personal property that was made in an earlier period, has special value because of its age, and is at least 70 years old — which in 2026 means made in 1956 or earlier. A “used collectible” is defined separately, at A(2)(b), as an item previously sold at retail as new in which there is a general interest in collecting, demonstrated by sales at antique and collectible shows and shops or by “the existence of associations or publications dedicated to the collection of the item.” For those two categories only, section D lets an out-of-state buyer give a resale certificate without a Maryland registration number. Section C is the limit that catches most shops: a vendor may not accept a resale certificate for a purchase of less than $200 paid by cash, check, card or electronic transfer unless the goods are delivered to the buyer's own retail place of business.

How much does consignment software cost?

The published monthly prices in September 2026 run from $139 to $379 before you pay anybody. ConsignCloud prints Basic at $139 a month and Pro at $189 a month per location, with ACH payouts at $2.50 each. SimpleConsign prints Basic $159, Standard $259, Professional $359 and Vendor Mall Plus $379, with ACH consignor payments included from Standard upward, under an introductory offer of $99 or $199 a month that runs for six months or until $50,000 in sales, whichever comes first. ConsignPro answered 403 at its pricing URL, Resaleworld answered 404, Ricochet and Rose did not answer at all, and Consignment Access served a 114-byte page. The sticker is not the number that matters, though: at $2.50 a payout, a shop paying 450 consignors monthly spends $13,500 a year on payouts against a $1,668 subscription.

Am I required to check whether a used item has been recalled before selling it?

Federal law makes it unlawful to sell a consumer product “subject to voluntary corrective action taken by the manufacturer, in consultation with the Commission, of which action the Commission has notified the public or if the seller, distributor, or manufacturer knew or should have known of such voluntary corrective action.” That is 15 U.S.C. §2068(a)(2)(B), and the phrase that matters for a resale shop is “knew or should have known.” The Commission publishes the whole list as a free JSON feed at saferproducts.gov. Counted on 14 September 2026 it held 10,002 recalls going back to 8 June 1973, of which only 458, or 4.6%, carry any UPC at all. A barcode scan cannot do this job; matching brand, model and description can.

Is it worth building custom software for a consignment or resale shop?

It depends almost entirely on how you pay your consignors. Two modeled Baltimore stacks for the same 450-consignor shop, with $480,000 of sales and $96,000 of it online, come to $21,355 and $10,495 a year, and the difference is the per-payout meter. The replaceable layer is $14,496 in the first case and $3,636 in the second, net of the ACH we would still pay ourselves. Against those, our $6,000 fixed-price online store pays for itself in about five months on the first stack and about twenty on the second. Our packages are published: a one-week Prototype Sprint is $3,500, a custom online store starts at $6,000, and a custom app, internal tool or operations system starts at $12,000.