Selling printable Carcassonne score sheets as downloads: the sheet your buyer trusts when a farmer is argued over

A printable Carcassonne score sheet is judged at the one moment the board stops being complete evidence: the final scoring. For an hour the landscape has been public — every completed city and road was scored the moment it closed, the meeples came back, and the running totals on the fifty-space track looked settled. Then the tiles run out and the game pays its second, quieter half: the unfinished cities, roads, and cloisters at their reduced rates, and the fields, which have scored nothing all game and can now be worth more than any single feature on the board. A farmer laid down in the first ten minutes, on a field nobody else contested, suddenly pays three points for every completed city it touches — and the table has to settle questions the board cannot answer from its final state alone. Which cities touching that field were actually completed? Two fields that look separate on the finished board — were they joined by one tile in the last round, merging their farmers into one contest? That city scored during play at two points a tile — is someone now counting it again at the end because a meeple is still standing next to it? The board shows the present and proves nothing about which features were already paid. The only evidence is the sheet: a record that either separates points scored during play from points scored at the end, feature by feature and field by field, or shows a single running total that nobody can rebuild once the farmers are argued over. That is why this vertical rewards a particular kind of seller. Not the seller with the prettiest landscape artwork, because the tiles, the meeples, and the landscape come in the boxed game and your buyer already owns them. The seller whose sheet gives the end of the game its own page — an end-scoring block per feature type at the correct reduced rate, and a field map in words, one line per field, with its farmers and its completed cities counted once each — so a final total can be rebuilt from the sheet alone after the landscape has been swept back into the box. Nobody refunds a score sheet because a game was dull. They stop reordering when a farmer was argued over and the sheet could not settle it. This guide walks through what a serious printable set actually contains, the rules every buyer assumes your sheet enforces, a worked game with the arithmetic checked the hard way, what the main channels really take from a low-priced set, the delivery failures specific to a game whose biggest points are invisible until the tiles run out, and the setup that prevents each one — including the honest cases where you should stay exactly where you are.

What a printable Carcassonne set actually contains (and why the end-scoring page is the product)

Open a serious printable release — the kind a regular game group reorders instead of searching again, and a host prints without being asked — and count what is in it, because the contents are the product. A single running-total column per player is a notepad, and in this game it is worse than a notepad: it hides the exact seam, between during-play points and end-game points, where almost every dispute lives.

• A during-play log with a feature column, not just a points column: one row per completed feature, naming what closed (city, road, or cloister), how many tiles it covered, how many pennants it held, and whose meeples scored it. Carcassonne scores most of its points the moment a feature completes, and those meeples return to their owners immediately — which means by the end of the game there is no physical trace of who scored what. If the sheet stores only the running total, a double-counted city is undetectable. If it stores features, anyone can re-add the column in a minute and the argument ends at the row where it started.
• The scoring table printed on the sheet itself, in both its rates — completed city 2 points per tile plus 2 per pennant during play, 1 per tile plus 1 per pennant at the end; road 1 per tile in both cases; cloister 9 completed, or 1 plus 1 per surrounding tile at the end; fields 3 per completed city, at the end only — because the two rates for a city are the commonest scoring error in this game. A sheet that prints only one rate invites the end-game city to be scored at the during-play rate, and at five or six tiles that single mistake is worth more than most farmers.
• A separate end-scoring block, visibly not more log rows. Unfinished features are a second scoring pass with its own rules — reduced city rate, cloisters counted by neighbours present, roads unchanged — and it deserves its own labelled section below a heavy line, with one line per unfinished feature per player. The visual separation is the enforcement: a city that appears in the during-play log cannot quietly appear again in the end block, because the two sections look different and a reader auditing the sheet checks them against each other.
• A field ledger: one line per field, written in words. Fields are the only feature that scores nothing when claimed and everything at the end, and they cannot be reconstructed from the final board with confidence once tiles have merged two fields into one. Each field needs a line of its own: a plain-words location ("north of the big city, west of the long road"), the farmers standing in it by colour and count, and at the end, the completed cities bordering it, counted once each. A field that borders the same city along three tiles is still one city and three points, not nine — and the ledger line is where that gets written correctly, once, instead of argued tile by tile.
• A majority-and-tie note printed next to the field ledger. Points go to the player or players with the most meeples on a feature; tied players each score the full amount, not a share. That rule matters most exactly where feelings run highest — a shared field paying twelve points to two players at once looks like an error to anyone who assumes points are split. The sheet that prints "ties score in full" beside the field lines turns the game's most suspicious-looking correct result into a printed rule the table can point at.
• A lap box for the fifty-space track. The score track runs from 0 to 49 and then wraps, so any score past fifty is a marker that has been round once — a fact pencil-and-memory tables lose constantly ("am I on 12 or 62?"). One small "+50" tick box per player on the sheet, checked when a marker laps, removes an entire species of argument that has nothing to do with the landscape at all.
• A final build line, below the end block and visibly not another scoring row: during-play total, plus end-scoring features, plus fields, in that order. The single most expensive Carcassonne argument is a field scored from excitement rather than from the ledger — a total announced before the bordering cities were counted. The build line forces the three kinds of points to be added in writing, in order, while the landscape is still on the table to be checked against.

The rules every buyer assumes (and the checks that enforce them)

A score sheet is a rules document whether its seller meant it to be or not: whatever the printed blocks imply is what the table will play. These are the rules buyers assume a Carcassonne sheet enforces, stated the way a sheet has to state them — as things to write down, at the moment they happen.

• A turn is draw, place, optionally deploy — and a meeple may only claim an unclaimed feature. The placed tile's edges must match: city to city, road to road, field to field. A follower may then be placed on a feature of that tile only if that feature, across all its connected tiles, has no follower on it already. You can still grow a feature you cannot claim, and two players can end up sharing a feature when separately claimed parts are joined by a later tile — which is exactly how shared cities and merged fields happen, and why the log records who scored, not who placed.
• Completed cities pay double rate; unfinished cities pay single rate; there is no small-city exception in current editions. A city is complete when it is fully enclosed by walls with no gaps. Completed: 2 points per tile plus 2 per pennant (the shield printed on some city tiles). Unfinished at the end: 1 per tile plus 1 per pennant. Older editions scored a completed two-tile city at the single rate; that exception was removed, and a sheet for the current edition should say so in a footnote rather than let a veteran player import it — the two-tile city is common, and the difference is two points every time it closes.
• Roads pay one point per tile, complete or not. A road completes when both ends are closed — by a crossroad, a city, a cloister, or by looping back on itself. It pays 1 per tile when it completes during play, and an unfinished road pays the same 1 per tile at the end. Roads are the one feature whose rate never changes, which makes them the sheet's control case: a road row that shows anything but its tile count is a recording error on its face.
• A cloister pays nine surrounded, or one plus its neighbours at the end. The cloister (monastery) tile pays when all eight surrounding spaces hold tiles: 9 points. Unfinished at the end it pays 1 for itself plus 1 for each surrounding tile actually present — five neighbours is 6 points, not 5 and not 9. The end block needs a neighbours-present count for exactly this, because "nearly surrounded" is where the arguments cluster.
• Farmers score nothing until the end, are never returned, and pay three per completed bordering city. A follower placed in a field lies down as a farmer and stays there for the whole game — it does not come back when anything completes, because fields never complete. At the end, for each separate field, the player or players with the most farmers in it score 3 points for each completed city bordering that field, each city counted once per field no matter how many of its tiles touch the field. Only completed cities count; the unfinished city beside a rich field pays the field nothing. Field scoring in the oldest edition was calculated from the cities' point of view at a different rate; current editions score field by field as above, and the sheet's ledger is written that way.
• Majority scores, ties score in full, and scored followers return — except farmers. On any feature, the most followers take all the points; a tie means every tied player scores the full amount. Followers on a scored city, road, or cloister return to their owner's supply at once and can be reused; farmers never return. A sheet with a small supply reminder — seven followers to deploy, plus the marker that lives on the score track — catches the quiet error of a player "scoring" with a follower they never legally had left to place.
• The game ends when the tiles run out, and the last tile still scores. Play continues until the final tile is placed and any feature it completes is scored normally; only then does the end-scoring pass happen — unfinished features at their end rates, then fields. The sheet's order is the game's order: log first, end block second, field ledger last, build line at the bottom. A table that scores fields before unfinished features will double-count at least once a night, because a city cannot be both unfinished-for-its-owner and completed-for-a-field, and the order forces someone to decide which it is, in writing, first.

A worked game, and the arithmetic, checked the hard way

Four players — Ana, Ben, Cleo, and Dan — base game, current edition. The figures below are what the sheet holds at the end; every total was re-added from its components, which is the check the sheet exists to make possible.

During play, Ana completes a six-tile city with one pennant: 6 × 2 + 1 × 2 = 14. She closes a five-tile road for 5, and surrounds a cloister for 9. Her log totals 14 + 5 + 9 = 28. Ben completes a four-tile city with no pennants for 4 × 2 = 8 and a seven-tile road for 7: 15 in the log. Cleo closes a six-tile road for 6 and a three-tile city with one pennant: 3 × 2 + 1 × 2 = 8, for 14. Dan completes a cloister for 9 and a four-tile road for 4: 13. At the last tile the log reads Ana 28, Ben 15, Cleo 14, Dan 13 — close enough that the end of the game, not the log, will decide the order behind Ana.

The end block pays the reduced rates. Ana has an unfinished three-tile road, worth 3, and a cloister with five neighbours present, worth 1 + 5 = 6: 9 more, running total 37. Ben has an unfinished five-tile city with one pennant — at the end rate that is 5 × 1 + 1 × 1 = 6, not the 12 the during-play rate would have invented — and an unfinished four-tile road for 4: 10 more, running total 25. Cleo has an unfinished four-tile city for 4 and a cloister with seven neighbours for 1 + 7 = 8: 12 more, running total 26. Dan has an unfinished two-tile road for 2 and an unfinished two-tile city for 2 — scored at the end rate without argument, because the removed small-city exception only ever applied to completed cities: 4 more, running total 17.

Then the fields, from the ledger, one line at a time. Ana is alone in a field bordering three completed cities: 3 × 3 = 9, final 46. Ben is alone in a field bordering two completed cities: 2 × 3 = 6, final 31. The third field is the one the ledger earns its keep on: Cleo and Dan each have two farmers in it — two farmers apiece because a late tile joined two separately claimed fields into one, which the ledger recorded when it happened — and it borders four completed cities, each counted once even where a city touches the field along several tiles. Tied majority scores in full, so Cleo takes 4 × 3 = 12 and Dan takes 12 as well: Cleo finishes on 38, Dan on 29. Ana 46, Cleo 38, Ben 31, Dan 29.

Read the movement, because it is the product's whole argument. Before the fields, the order was Ana 37, Cleo 26, Ben 25, Dan 17. The field pass moved Cleo past Ben into second — on a tie rule that pays both players in full, in a field whose merger was written down one round before the end — and it gave Dan, in last place, the joint-largest field score of the game without moving him off the bottom. Every one of those facts is an argument at a table playing from a totals pad: the tied field "must be split," the merged field "was two fields," the five-tile city "was worth twelve." On this sheet each claim is settled by a line written during the game, at the rate printed beside the line. That is what the buyer is actually buying.

What the channels actually take from a low-priced set

A printable Carcassonne score sheet is a low-priced consumable, typically in the same few-dollars band as the other score sheets in this series, and at that price the channel arithmetic — not the sticker — decides what you keep. Work it on a representative sale, and put your own current figures in before you list, because these rates change and your product page should never quote this guide's numbers as the channel's promise.

On a marketplace sale, the buyer finds you, and you pay for being found: a listing fee where the marketplace charges one, a transaction percentage on the total including any shipping-equivalent or tax handling the marketplace applies, and a payment-processing percentage plus a fixed amount. On a five-dollar sheet those fixed amounts are not rounding — the fixed processing piece alone can take a larger share than any single percentage line, which is why bundling matters more in this vertical than in almost any other: a pad-length file, a base-game and a large-print pair, or a score-sheet plus field-ledger set sold as one listing spreads every fixed fee across more value without changing your per-game price in the buyer's head. On your own storefront, there is no marketplace transaction percentage, but there is still processing on every order, and there is the delivery cost the next sections are about — a failed download on a two-dollar-margin product does not cost two dollars, it costs the support exchange, and support time is the only input in this business that does not scale.

Two honest consequences follow. First, never price a single bare sheet so low that one refund — issued, as refunds usually are, without the platform fees coming back to you — wipes out several good sales; price the bundle, and let the single sheet be the entry listing if you list one at all. Second, a score sheet is a reorder product only if the file is the same file every time: a buyer who reorders and receives a sheet with the field ledger moved will not complain, they will simply learn not to trust the new pad at the final scoring, and the reorder after that goes to someone else. Version discipline — one canonical file, changed deliberately, announced on the listing — is a pricing decision as much as a quality one.

The three delivery failures specific to a game whose biggest points are invisible until the tiles run out

Digital delivery fails in generic ways for every product — a link that expires, a file that will not open — and those are covered by every platform's help pages. A Carcassonne score sheet has three failures of its own, and all three happen after a technically successful download, which is why sellers who only check "did the file arrive" never see them coming.

• The wrong-edition delivery. The buyer ordered the current-edition sheet and received one built for an older scoring edition — with the two-tile city exception and the old field method printed as rules — or the reverse, and their table's sheet enforces scoring their edition does not play. The rates differ exactly where it matters, and the sheet gets blamed for a picking error that happened in your fulfillment settings. It happens when several near-identical files sit behind similar listings and the mapping between listing and file was set once and never re-checked after an edit. The symptom in reviews is never "wrong file" — it is "not as described," which costs more.
• The print-scale failure. The buyer's printer dialog shrinks the sheet to fit a different paper size, the log rows compress, the paired entries ("city — 6 tiles, 1 pennant") no longer fit inside the row, and the scorekeeper starts writing across row boundaries — recreating, in the buyer's kitchen, exactly the unauditable record the product exists to replace. A file cannot control a printer dialog, but it can defend itself: the print specification on the sheet itself, a row height chosen with margin rather than at the minimum, and a one-line test in the instructions — print one sheet, enter a completed city and a full field line in it, and check both stay inside the lines before printing the pad.
• The version-drift failure. You improve the sheet — you add the counted-once note to the field ledger after one review too many — and you update the listing's preview image but not the delivered file, or the file but not the preview. The next buyer plays a different sheet from the one they thought they bought, and when their fields merge in the last round, their copy genuinely does not have the rule your support reply tells them to use. Every element of the product — preview, description, delivered file — has to change in the same edit, or not change at all.

None of these three is a platform bug. All three are mapping and versioning discipline, which is unglamorous and entirely within the seller's control — and that is good news, because it means a careful seller can be visibly better than the field at the exact moments buyers remember.

Editions and occasions are different products

The fastest way to ruin a good Carcassonne sheet is to keep adding to it. The base game, played with the standard table and farmers scored once at the end, wants one during-play log, one end block, one field ledger, and nothing else on the page. Everything else a buyer might want is a different product for a different table, and should be sold as one, with its own listing, its own preview, and its own delivered file.

A river-start edition is the mildest case and still a separate file: the start changes, the scoring does not, so it wants the standard sheet with a short start note, labelled as that variant everywhere rather than swapped silently into the base listing. An expansion edition — for the well-known variants that add traders, builders, inns on roads, or cathedrals in cities — is a genuinely different product, because its rates genuinely change (a road with an inn or a city with a cathedral can pay double, or nothing if unfinished), and it should never be bundled silently with the base file. A classroom or club edition wants the opposite trade — more games per page, a smaller log, and a results strip a teacher or club host can collect — because its buyer is buying throughput, not audit depth, and should be told so honestly on the listing. And a large-print edition is a real product, not a setting: bigger rows, heavier rules lines, fewer rows per page, for players for whom the standard log is genuinely hard to write in — its buyers are loyal precisely because most sellers treat their need as a printer setting.

What does not belong in any edition is landscape artwork pretending to be the game. A score sheet seller does not sell Carcassonne, does not sell tiles or meeples, and should never imply the boxed game's maker endorses the pad. Name the game because buyers search by it and because nominative use is how a compatible accessory is honestly described; do not copy the box art, the logo, or the tile art into your preview. The sellers who get into trouble in game-adjacent verticals are almost never the ones who wrote their own scoring block — they are the ones who made their listing look official.

The setup that prevents all three (and when to stay put)

The setup is boring, and it is the product. One canonical file per edition, named so the edition is in the filename itself. The listing, the preview image, and the delivered file changed in the same edit or not at all, with the preview re-exported from the delivered file rather than kept as a separate image that can drift. A delivery check you actually run: place a test order for each edition, download what the buyer downloads, and confirm the file in hand is the edition the listing promised — the wrong-edition failure survives every proofread and dies to a single test order. Keep the file light, the rows honest sizes, and the print specification printed on the sheet, so the buyer's first print is the test print.

For delivery itself, what you need is unglamorous: a fixed price you can read before your best month starts, expiring download links, and a file that arrives the same way every time. That is the shape to demand from any delivery tool, and it is the shape FileFreight is built around — flat-rate delivery with license keys for sellers who need them. Be honest about the limits before you move anything: it does not handle sales tax or merchant-of-record duties, it has no importer yet, and it is not a Shopify app, so if your store lives on Shopify it is not a drop-in replacement today. If your current setup already delivers the right edition, every time, at a cost you can state in one sentence, stay put — a score sheet business runs on reorders, and reorders run on the file never changing underneath a returning buyer. Move when the mapping, the versioning, or the bill is the thing you cannot state in one sentence, and not before.