What a QC photo set can and cannot prove
Three things this piece settles
- A checklist line passes only when the frame shows the thing itself, clearly enough for someone else to agree.
- A missing frame is a re-shoot request; a defect shown in a frame is a claim.
- A photo set can prove identity, size marking and visible defects, and nothing about feel or durability.
Which stop this touches
8 stops · open a stop to read its windows · “n open” = actions still available
- Stop 1: Pick the channel — 3 actions still open. A channel is a filtered view of one agent, not a shop. Learn what its row count proves and what it cannot.
- Stop 2: Hand the link over — 3 actions still open. A link carries an identity, not a product. Keep the identity intact and the order stays traceable.
- Stop 3: Pay for the item — 2 actions still open. Item payment and freight payment are two separate moments. Most closed doors close at the first one.
- Stop 4: The warehouse clock — 3 actions still open. Arrival starts a clock. Verify the window on your own account before you plan around it.
- Stop 5: QC photos — 3 actions still open. A photo set is evidence of coverage, not a verdict on quality. Ask for angles, not for opinions.
- Stop 6: Merge and pack — 3 actions still open. Merging trades a second freight charge for a bigger volumetric weight. Run both sides.
- Stop 7: Freight and customs — 2 actions still open. Freight is charged on chargeable weight. Customs is decided on declaration fields you fill in.
- Stop 8: Checking the parcel — 2 actions still open. Delivery starts the claim clock. Photograph the unopened parcel before you open it.
Four groups of checkable lines: carton, cut, detail, markings
A QC photo set is a group of frames taken at the warehouse, and what it can prove is bounded by what a camera can see in a few minutes without opening a sealed product. The checklist below is written as four groups of checkable lines, ordered from the outside of the parcel inwards: the carton, the cut and shape, the details, and the markings. Each line is something you can tick or reject from the frames themselves rather than from what you hope the frames imply. The list is deliberately shorter than it could be, because a line that cannot be judged from a photograph does not belong on it.
The carton group covers everything about the packaging the item arrived in, and it is the group people skip because it feels administrative. Check the box for crush damage on the corners, for water marks along the base, and for tape that has been cut and re-sealed. Check that the packaging matches the item type, since a rigid shoe box inside a soft bag tells you something the seller page did not. Judge this group first, because damage here makes every later group less worth reading.
The cut group is about silhouette, and it is the group where a photograph is genuinely stronger than a written description. Check that the shoulders, the sleeve length and the body width read the way they did on the listing, and check that the proportions between them are plausible for the stated size. Check the hem and the collar for symmetry, since an asymmetric collar is far easier to see than to describe. Judge this group against the listing image rather than against what you expected, because expectation is not evidence.
The detail group covers the parts a product photograph usually hides, which are stitching, hardware, lining and the inside of the item. Check the density and the straightness of the topstitching along a seam you can see clearly, and check that the hardware has weight and a consistent finish rather than a painted look. Check the inside of the item wherever the frames allow it, because an interior is where cheaper construction shows first. Judge this group last in the set, because it takes the most time per frame.
The markings group is about what is printed or attached, which means labels, tags, logos and the size marking itself. Check that the size on the physical tag matches the size on the order, and check that the label layout is coherent rather than one word pressed onto blank fabric. Check that any brand mark sits where that brand places it, which is a comparison against a reference rather than a test of memory. Our own sample shows why this group needs care: the 192 titles yield 51 distinct brand strings, and three of them, Ami, Ralph Lauren and Margiela, are also substrings of another brand string in the same set, covering 5 rows where the text alone cannot separate two labels.
How each line passes and how it fails
A line passes when the frame shows the thing itself, clearly enough that a second person could reach the same conclusion from the same image. A line fails when the frame that would settle it is missing, which is a different problem from a frame that shows a defect. That distinction is the whole point of a checklist, because a missing frame produces a request for another frame while a defect produces a claim. Record which of the two you have before writing anything to the agent.
Carton lines pass on a clear shot of all four corners and the base, and they fail on a single oblique frame that hides two of them. Cut lines pass when the item is laid flat or hung straight with the whole silhouette inside the frame, and they fail when the item is folded or cropped. Detail lines pass when the camera is close enough to resolve individual stitches, and they fail when the entire garment is in the frame at once. Markings pass when the tag or the logo fills a meaningful part of the frame, and they fail when the text is too small to read.
Two failure modes deserve naming because both of them look like passes at a glance. The first is the frame that shows a similar but different item, which happens when several rows share one photo session, and it is caught by checking the row id against the id you handed over. That check matters especially where the title is thin, since 89 of the 192 sampled titles, 46.4 percent, run shorter than 24 characters, so the id rather than the wording is what ties a frame to an order. The second is the frame taken from an angle chosen to hide the interesting part, which is caught by asking for the opposite angle rather than a better version of the same one. The first is a bookkeeping error and the second is a framing error, and they need different requests.
Counting frames is a weak test on its own, and it is worth saying so because frame counts are the easiest thing to compare between two sets. Twelve frames of the same sleeve prove less than three frames of the sleeve, the hem and the label together. The number that matters is coverage against the checklist, which is what the QC scorecard computes and which can be read in a single pass. The exception is a set that covers every line but is too low in resolution to judge, where the count and the coverage both look right and neither of them is.
Judgement calls belong in writing, and the place to write them is next to the line they belong to. If a frame is ambiguous, say which part is ambiguous and what a second frame would need to show, because a request phrased that way usually produces a usable image on the first attempt. If a frame is fine but your expectation was different, that is a note about the listing rather than about the item, and it belongs in a different column. The exception is a frame that is ambiguous and unimportant, where the honest move is to tick it and move on.
When a photo fails: the re-shoot list
A failed set produces a re-shoot list rather than a complaint, and the list is short by design. Ask for the missing or unreadable line, name the angle or the distance you need, and give the row id so the next session cannot be attached to the wrong item. Keep the request to the lines that actually failed, because a request asking for everything again usually produces more frames of the same quality. Send it before the photo window closes, since the window is the one part of this process that cannot be reopened.
The five lines worth re-shooting first are the carton base, the full silhouette, the size tag in close-up, the logo placement, and any interior the first set skipped. Each of those is a single frame and each settles a question that no other frame settles. Asking for them together costs one session rather than five, and it removes the most common reason a second request becomes necessary. The exception is a line that failed because the item itself is wrong, where the frame you want is the one that proves it rather than the one that shows it better.
Write the request in the language of the checklist rather than in the language of disappointment, because the person holding the camera has not seen your listing. A request for the size tag against a stated rule produces a usable frame, while a statement that the item looks wrong produces an opinion. Include what you already have, so the session does not repeat frames you can already judge. The exception is a genuine mis-ship, where the useful message is the row id and the mismatch rather than a list of frames.
Some things a photo set cannot prove at all, and knowing where the boundary sits saves a wasted request. It cannot prove how a fabric feels, how a garment drapes after a wash, whether a sole will stay glued, or whether a mechanism will still work in a month. It cannot prove authenticity in any formal sense either, because a photograph of a label is not a verification of a supply chain. What it can prove is narrower and more useful: that the item in the box is the item on the order, that its size marking matches, and that it carries whatever defects the frames show.
Set the checklist against the window and the order of work becomes obvious, because reading frames is slow and the window is not. Skim the carton group first, then the cut group, then the markings, and leave the detail group for the frames that survived those first passes. If time runs out, the lines that matter most are the size tag and the silhouette, since those are the two that are hardest to argue about later. Our own sample is a reminder that the text you compare against is not always clean, because eight of the 192 stored titles still carry an unescaped HTML entity in place of an ampersand. The exception is a row whose title names no brand at all, where the id is the only handle you have.
The data point behind this piece
The 192 sampled titles yield 51 distinct brand strings, and three of them, Ami, Ralph Lauren and Margiela, are also substrings of another brand string in the same set, covering 5 rows where the text alone cannot separate two labels.
Basis: Runbook desk parse of the 2026-10-06 snapshot; a substring test was run over every distinct brand string read verbatim out of the stored titles. The other pairings are Ami inside Amiri and Ami Paris, Margiela inside Maison Margiela, Ralph Lauren inside Polo Ralph Lauren.
Eighty-nine of the 192 titles, 46.4 percent, run shorter than 24 characters, so for nearly half the sample the only text a frame can be checked against is a product noun and a colour.
Basis: Same snapshot and same parse; the character count is taken on the stored title as written, including spaces and punctuation.
Eight of the 192 stored titles still carry an unescaped HTML entity where an ampersand was meant, which is why a photo should be tied to the row id rather than to the wording of the title.
Basis: Same snapshot; counted by a literal search for the ampersand entity in the stored title field of every row.
This piece answers searches about “joyagoo qc photos”. It is part of the reading and evidence cluster.
Where to go next
Links to Kakox are tracked so the destination can attribute the visit. Nothing on this site is sold by us.
Reviewed 2026-10-06 · Weekly review · next review weekly review