LabelCropper
← All articles

Every LabelCropper feature, explained

· 13 min read · Features

LabelCropper has grown well past “crop the blank half off”. This is the full list of what it does today, grouped by the part of the job it belongs to.

One pass, on drop

Every export at once, not one at a time

The file picker takes a multiple selection and the drop zone takes a handful together, so a day that was downloaded in six batches goes in as six files in one go. Once they are in there is no notion of which file a page came from except for sorting: pages from different PDFs interleave freely, and a page range typed against one file does not disturb the others.

There is no merge step

Whatever went in comes back as a single PDF under one timestamped name. Combining is not a second job you run after the crop — it is the same pass. The cropped pages are written into one document in picking order, so a morning’s labels print as one continuous stack whether they arrived as one export or ten.

This is worth saying plainly because the picker screen invites the opposite reading. Merge PDF sits at the bottom of the marketplace list, and it looks like the second half of a label run. It is not: it is there for PDFs that are not labels at all — a courier manifest, a set of scans, an invoice pack — where you want the page grid and none of the label handling. A label batch never needs it.

Nothing waits on a setting

Cropping, captioning, sorting and combining all start the moment the files land, with nothing to configure first. Customisation comes after the batch is already right: reorder, rotate, duplicate or delete anything in the grid, or change the caption size, page size and summary page in Settings and export again. The order is deliberate — a bench should get a printable stack before it is asked a single question.

Or nothing at all

With Download automatically on, a batch that lands with nothing flagged saves itself as soon as the sort finishes. Choose the files, and the next thing you touch is the printer.

Cropping

Automatic Meesho label cropping

A Meesho sheet is an A4 page carrying a shipping label, a Product Details table and usually a tax invoice ending in a bold Total row, followed by a disclaimer and blank paper. LabelCropper finds the drawn content box and the rule under that Total row, and cuts there — keeping the label and the invoice summary, dropping the rest. Exchange labels, which carry no invoice at all, are detected as a separate case and cropped to the label box.

Automatic Flipkart label cropping

A Flipkart (E-Kart) sheet puts the label in a box about a third of the page wide, with a dashed tear line beneath it and a full tax invoice below. There is no summary worth keeping, so the whole invoice goes and the crop narrows horizontally as well as vertically — removing roughly 85% of the sheet.

Automatic Snapdeal label cropping

A Snapdeal sheet prints the invoice beside the label rather than beneath it, so the cut runs down the page instead of across it. Everything to the right of the label goes and nothing at all is taken off the label itself — the seller address, the Snapdeal reference barcode and the notices at the foot all stay, because that is what makes it returnable. The caption is read off the invoice half, since the label prints its own copy of the SKU with the hyphens stripped and the size run onto the end. Parcels shipping on an Amazon Shipping label, where the whole label is one image with no text in it, are handled as their own case.

AJIO: matching rather than cropping

AJIO is the exception in this list, because nothing needs cutting — its label is already 4 × 6. The job is pairing. AJIO hands you two files, the labels and the A4 invoices, and both carry the same AWB number: the invoice prints it as text, the label only under a barcode. LabelCropper reads the barcode and puts each label immediately in front of its own invoice, so the bench works down one stack — label on the parcel, invoice inside. The SKU and quantity are lifted off the invoice and printed on the label, which is the only way they can be read off a pile at all. A label that matches nothing is reported and left unpaired rather than paired with a guess.

Drawing the crop yourself

Automatic geometry is the right answer for a label and the wrong one for a scan with a margin to lose or a deck with a footer on it. For those there is a crop box you drag: eight grips to resize it, one page at a time or the same rectangle applied to all of them, and a button that finds the content edges for you rather than making you drag to them. Every page keeps its own rectangle, so trimming one page on its own and giving a whole scan one margin are the same tool rather than two.

Invoice scaling rather than deletion

On Meesho pages the invoice band is kept but shrunk, so the page holds its legal content in a fraction of the height. On a real 109-page export this removed about a third of the page height before scaling, and appreciably more after.

Unrecognised pages are passed through, never guessed at

Geometry is read from each page, never hardcoded. A page that does not match is flagged and left exactly as it arrived. Picking the wrong marketplace card on the home screen costs you a warning, never a mangled label.

The SKU caption

A bordered caption table under every label

Each cropped label gets a bordered table carrying its SKU, size, colour and quantity, set large enough to read off the edge of a stack. It sits below the label rather than above, because the label’s own header is what identifies the parcel — the caption is the picking aid, and putting it at the foot stops the two competing for attention.

Two layouts, chosen by marketplace

Flipkart labels are about 216pt wide, so four stacked rows would double their height; they get a grid with headings across the top and values beneath. Meesho labels are around 571pt wide and can afford one field per row, which gives a long SKU the full width instead of a quarter of it.

A parcel that holds two garments says so

One label does not always mean one item. Where a parcel covers more than one order line, the caption names them rather than captioning the first and leaving the rest to be discovered at the packing table: two SKUs print in full, and beyond that it reads SKU +2 more, because three at a readable size will not fit. The quantity shown is the parcel’s, not the first line’s.

Snapdeal additionally raises “2 items on one label — check before packing” even though the caption now names them. It reads its items off the invoice half, which is the half the crop throws away, so on a three-item parcel there is nothing left on the printed page to check against. Meesho and Flipkart print their product tables inside the label itself, which survives the crop — there the caption is a convenience and the table is still the record.

Adjustable caption size

Caption size is a multiple of the base, set per marketplace, and it matters more than it sounds: scaling a Meesho label down onto thermal stock shrinks the caption with it. Raising the percentage is how you compensate. Below about 6pt the caption stops being readable, so text is trimmed rather than shrunk further.

A line of your own above the label

One line of your own text can be printed above every cropped label — the same message across all four marketplaces. It is a switch rather than an empty box, so the text survives being turned off, and it is drawn only where a caption was drawn. Without a plan it prints on the first five labels of each day, the same allowance the caption draws from; the rest of the batch comes out without it and the count starts again tomorrow.

A crop stamp on every page

Every page carries the date and time of the run it came from, so a reprint can never be confused with the original stack sitting next to it on the bench.

Printing and page size

Three page-size choices, per marketplace

A cropped Meesho label is roughly 8in wide, so it comes down to about 50% on 4 × 6 and about 37% on 3 × 5. See printing on thermal stock for how to set the caption to match.

Dispatch sorting

Courier first, then seller, then the garment

The full comparison is courier, then seller account, then design, then colour, then ascending size, with the SKU as a deterministic tiebreak. Meesho dispatches through Valmo, Delhivery, Shadowfax and Xpress Bees, so courier is the outer grouping there. Flipkart has one carrier, so that level is constant and the ordering falls through to the seller account instead — the same code path, no special case.

Choosing what leads the sort

Courier-first is the default and suits a manifest hand-over, but it is not the only shelf layout there is. A picker in Settings chooses which dimension leads: Courier, then SKU; SKU alone — design, then colour, then size, ignoring the courier, which is right when one carrier takes everything; Colour, for a rack organised by colour; or Size, for shelves stacked by it. Every mode ends with the same tail of design, colour, size and SKU, so two parcels that tie on the leading key still land beside each other rather than shuffling apart between runs. Sorting by SKU is free; the other three come with a plan.

Sizes that sort correctly

3XL and XXXL are the same size written two ways, and a naive sort puts one of them after the other. LabelCropper maps both onto one scale, sorts numeric sizes numerically, reads the leading number out of 30-32 and 30 IN, and places Free at the end of the scale rather than among the unknowns.

The dispatch summary

Optionally appended after the labels: the item list first — every distinct garment, size and colour, with how many parcels carry it and how many units that is — then tallies by courier and by seller. It counts packages rather than units, because a pickup is counted in parcels. Valmo and ValmoPlus are counted separately, since they are billed and manifested separately even though they sort into one pile.

It is off by default: most batches are printed to be packed, not reconciled, and two extra A4 pages on every run is paper spent on a report nobody asked for. Turn it on once and it is remembered for that marketplace.

Editing the stack

Edits are stored as data on the page item rather than as changes to the bytes, which is why they survive reordering and why undo works on all of them.

Exporting

Timestamped filenames

Every export is stamped with the local date and time — labels_2026-08-30_15-32-45.pdf — so a second run never silently overwrites the first and the browser never falls back to merged (1).pdf. The stamp sorts lexicographically, so files line up chronologically in Finder and Explorer without touching the sort column. Exporting repeatedly replaces the stamp rather than accumulating suffixes.

Auto-download on a clean batch

Optional: when a batch lands with nothing flagged, the export starts on its own. On a run of clean exports that removes the last click.

Settings that stick

Caption on or off, caption size, page size, auto-download and the summary page are all saved per marketplace, on your device, and survive reloads and exports. There is deliberately one saved set per marketplace rather than named presets — the only thing that varies between real batches is which marketplace they came from, and the tool already knows that. There is no Save button because nothing needs confirming.

Offline and on a phone

LabelCropper is a web page and keeps working with the network off, so a dropped connection at the bench does not stop a batch mid-way. There is nothing to install and nothing to update — you open it and it is the current version. On a phone it can be added to the home screen from the browser’s own menu, after which it opens straight into the tool rather than into a browser, and the page grid, the drop zone and the export button are all built for a touch screen rather than shrunk onto one.

Ordinary PDFs too

Three entries hand you the same page grid for any PDF at all, with the label detection left out of it entirely: Merge PDF to combine files in the order you choose, Rearrange PDF to reorder, rotate, duplicate and delete pages, and Crop PDF to draw a box and trim them. They are one workspace behind three doors, because they are the three things people search for separately.

They are for the files a dispatch desk accumulates that are not labels — manifests, scans, an invoice pack to staple together. A label batch never needs them: it is already cropped, captioned, sorted and combined by the same pass, as above.

What a plan covers

Cropping, merging, rearranging and exporting are free with no label limit, on every marketplace, for as long as you like. What a plan adds is the dispatch end of the job:

Every visitor gets seven days of all of it with no card asked for, and plans start at ₹199 a month plus GST. The table is on the pricing page.

All of it, in your browser

Free to crop with no page limit and no upload; seven days of the paid features to try, with no card.

Open LabelCropper

Keep reading

Why us Why sellers choose LabelCropper Guide Printing cropped labels on 4 × 6 and 3 × 5 thermal stock