Best Practices for Handbag Sample Photography and Feedback

What is the best way to photograph a handbag sample and send feedback that actually gets fixed? Shoot one nine-frame set under 5500K light with an 18% grey card in the first frame, then send a numbered revision list in which every item points at one annotated image. Photographs without numbers get discussed; photographs with numbers get sewn — a numbered list tied to a single frame per item is the difference between closing a sample in one round and spending three rounds on the same bag. The whole set takes about twenty minutes to shoot and removes the 3-6 days that a vague review usually adds to a sampling window of 6-10 working days.

Sample photography parameters we recommend

Three Ways to Shoot a Sample, Compared

There are three realistic ways a brand photographs a sample: a phone on a desk, a folding lightbox with fixed lamps, and a small macro rig with a tripod and a grey card. They differ in what they can settle. The question is not which produces the prettiest picture; it is which produces a picture that settles a dispute about a seam, a shade or a strap drop.

The recommendation is not the most expensive option. It is the lightbox for the overview frames and a phone held 15-20 cm away for the macro frames, because the two together settle everything that a revision list needs to name. A dedicated rig earns its place only when a programme runs more than about twelve samples a season, at which point the repeatability starts to matter more than the setup time.

Whatever the setup, the rule that does the work is consistency: the same distance, the same light and the same backdrop every round. A revision that looks fixed in a brighter photograph probably is not, and that is how two rounds get spent on the same seam. This is the same discipline described in the sampling and approval process, where the approved sample is a reference object rather than a memory.

The Nine-Frame Set That Answers Most Questions

Nine frames covers a bag completely without turning review into a photo project. One overview establishes proportion, seven detail frames cover the joints and finishes that fail first, and one worn frame shows how the bag sits on a body. Shooting them in a fixed order means the second round can be compared frame by frame against the first.

1. Frame 1, overview with grey card. Front elevation, bag upright and empty, ruler at the base, grey card at the edge.
2. Frame 2, back elevation. Same distance and light; this is where panel symmetry and back pocket placement show.
3. Frame 3, base and feet. Bag inverted; base reinforcement and foot alignment are visible here and nowhere else.
4. Frame 4, strap attachment. Macro of the joint where the strap meets the body, because this is the joint that fails first in the field.
5. Frame 5, seam macro. 15-20 cm away, ruler flat against the seam, so stitch density can be counted off the image.
6. Frame 6, edge paint. Macro along a raw edge at a shallow angle to catch gloss and coverage.
7. Frame 7, hardware and closure. Zip puller, slider, feet and any plated part, shot straight on to show tone.
8. Frame 8, interior and lining. Bag opened flat; pocket layout, lining tension and any raw edge inside.
9. Frame 9, worn. On a body or a stand at the intended drop, showing where the bag actually falls.

How Close Should a Macro Frame Be?

Close enough that a 10 mm seam allowance fills a third of the frame, which on most phones means 15-20 cm. At that distance stitch density can be counted directly: 7-9 stitches per inch on a body seam is the commercial norm and is legible in a 3000 px frame. Further back than that and the seam becomes a texture; closer and the depth of field collapses and half the stitch line is soft.

Feedback in Three Formats: Which One Gets Fixed

Feedback arrives in three shapes: prose in an email, annotated images, and a numbered list with images attached. They are not equivalent. Prose has to be interpreted, annotations have to be collected, and a numbered list can be worked through, ticked off and returned. The format is a stronger predictor of how many rounds a sample takes than the content of the feedback itself.

A numbered list also gives the reply a structure. When the production partner answers item 3 with a measurement rather than with a promise, the exchange is finished; when it answers with we improved the strap, the exchange continues. Numbering is the cheapest control available in the whole sampling stage, and it is the one most often skipped.

Lighting and Colour Fidelity in Sample Photographs

Colour cannot be judged from a photograph that has no known light source. A lamp at 5500K with a colour rendering index of 95 or higher approximates the D65 daylight standard used in a colour approval booth, which means the sample photograph and the lab dip are being read under comparable conditions. An 18% grey card in the first frame lets the white balance be corrected afterwards, and it is the single cheapest addition to any setup.

• One light temperature. 5500K throughout; mixing a warm room lamp with a daylight lamp makes every shade judgement worthless.
• Neutral backdrop. Mid-grey or white card; a coloured backdrop reflects into the bag and shifts the reading.
• Fixed exposure. f/8, ISO 100, 1/125 s, so two rounds are comparable frame for frame.
• No enhancement. Turn off auto-enhance, portrait mode and filters; they change colour and blur edges that need to be sharp.

Even with all of that, a photograph is evidence about construction, not a colour approval. Shade should still be signed off on a physical swatch measured under D65 with a check source, as set out in the Pantone to fabric dyeing walkthrough. The photograph's job is to show whether the seam is straight and the strap sits square.

What to Photograph on Each Silhouette

The nine frames shift by silhouette, because the joint that fails differs. A tote fails at the strap-to-body junction under load; a crossbody fails at the slider and the strap adjuster; a bucket fails at the drawcord channel and the reinforced base. Photographing the generic set on a structured satchel means missing the one joint that matters.

Numbering the Revision List

A revision item is only actionable if it names a place, a measurement and a direction. Move the strap 8 mm toward the centre is actionable; the strap feels wrong is not. Each item should carry a frame number from the nine-frame set, so the reader goes straight to the right picture, and each should say whether it blocks approval or is a preference.

1. Place. Which panel, which seam, measured in millimetres from a stated datum edge.
2. Observation. What the frame shows, stated as a measurement rather than an adjective.
3. Target. The number wanted: 10 mm seam allowance, 7-9 stitches per inch, strap drop 250 mm.
4. Priority. Blocks approval, or nice to have. Say which.
5. Frame reference. Frame 4 of 9, so nothing has to be searched for.

Two habits make the list faster to action. Group items that touch the same pattern piece, because one pattern change can close three of them. And separate construction items from colour items: construction goes back to the bench, colour goes back to the dye house, and mixing them in one list means neither moves quickly.

Turning Photographs Into a Specification Change

Every accepted revision should end as a line in the specification, not just as a change to one bag. If the strap drop moves 8 mm on the sample, the tech pack moves with it, because the bulk run is built from the document rather than from the sample. A revision that lives only in an email thread is a revision that will be re-litigated at inspection.

• Update the measurement table. Body width, gusset depth, strap drop and pocket positions, each with a tolerance.
• Update the construction notes. Stitch density, seam allowance, reinforcement, edge finish.
• Update the colour references. Body, lining, thread, binding and hardware, each with its approved code.
• Re-issue with a version number. v1.1 dated, so nobody is working from the earlier file.

This is also the point at which the number of rounds becomes a decision rather than an accident. A programme that has revised three times on the same silhouette is usually not facing three problems; it is facing one unclear specification. The rounds a custom bag really needs is the honest answer to how many are normal before the brief, rather than the bag, needs rewriting.

When to Stop Photographing and Start a Call

Photographs stop being efficient at a predictable point: when the same item has been revised twice, or when the disagreement is about hand feel, drape or stiffness rather than about a measurement. Those properties do not survive compression, and another nine-frame set will not settle them. A fifteen-minute video call with the sample in hand on both ends will.

Use the call to agree a physical reference instead of a description. Send a cutting of the felt, interlining or foam that matches the intended hand, or a competitor bag already owned, and let the bench match an object. Then confirm in writing with one frame and one measurement, so the agreement does not evaporate on the next round. Where the prototype and pre-production sample distinction applies, hand feel should be frozen at the pre-production stage rather than at the prototype, because interlining and lining change it.

Six Failure Signatures a Still Frame Shows

A photograph cannot measure a seam, but it can show nearly every construction fault early enough to act on it. The frame has to be close enough to count stitches and flat enough to show geometry, which is what the 15-20 cm macro distance and the fixed 90 cm overview were for. Reading a set against a named list rather than against an impression catches faults that would otherwise surface at final inspection, where correcting them costs weeks rather than a round.

Naming Files So Three Rounds Stay Comparable

Three revision rounds produce 27 frames and one argument about which set is current. A naming convention removes it. Every file carries the style code, the round number, the frame number and the date, and nothing else; anything the receiver renames falls outside the sequence and can be ignored safely. The same convention should be used by whoever photographs the reply, agreed in the first email, so the fourth frame of round two can be placed beside the fourth frame of round one without opening both folders.

• One scheme, written once. STYLE-R2-F4-2026-03-11 states everything needed later; a file named IMG_2287 states nothing.
• Retain the raw frame. Keep the untouched file at 3000 px or more and edit a copy, so a second person can re-read the same image.
• Fix size and colour space. sRGB throughout at 3000 px on the long edge, so a stitch count means the same thing in both rounds.
• Keep every round. Deleting round one removes the evidence that a fault existed; hold all three sets until 90 days after bulk is accepted.

Where the Days Go in One Review Round

Photography itself takes under half an hour. The delay is everything around it: selecting frames, writing the numbered list, waiting for a second opinion, and the courier legs that carry the bag out and back at 3-5 days each. A round that could close in eight calendar days routinely takes eighteen, and the ten extra days almost always belong to somebody on the buying side rather than to the bench.

Booking the courier before the shooting session rather than after the list is finished removes two days from almost every round, because the parcel leaves on the day the list is signed instead of the day somebody remembers to arrange it. Rounds stack quickly once they slip: two rounds at eighteen days each is five weeks of a calendar that was supposed to hold sampling inside 6-10 working days. The sample revision sequence sets out the full round.

What a Usable Reply Contains

The reply has a format too, and asking for it at the outset saves a round. An answer given item by item should re-shoot the same frames under the same settings, quote a measurement rather than a promise for each item, and say which items were left unchanged and why. Where a revision touched a pattern piece, a physical counter sample should follow by courier; where it was photographic only, images settle it and three days are saved.

• Same frame numbers. Round two frame five beside round one frame five, at the same distance and light source, so the comparison is immediate.
• A number per item. Item three answered with nine stitches per inch is closed; item three answered with improved is not.
• Declared exceptions. Anything left unchanged should be stated explicitly, with the reason, rather than disappearing from the reply.
• Courier only where it earns it. Returning the unit costs USD 35-70 and 3-5 days; still frames settle most construction items without it.

Frequently Asked Questions

People Also Ask