A sample library is a working instrument rather than an archive, and it earns its keep only when every swatch carries its documents alongside its appearance. Build it as five linked records per material: the physical sample, a digital image, the certificate or declaration, the commercial terms, and the performance data. Start with the materials you order most often and the ones that cause the most rework, label each sample with the mill, article code, lot and date, and store leather and coated fabric in controlled humidity. A library assembled this way turns a seasonal material search into a fifteen-minute comparison, and it is the single most effective way to shorten the next development cycle.
What a sample library is, and what it is not
Most brands already possess the raw material for a library without realising it: a cupboard of swatches sent by tanneries and mills, a folder of certificates attached to old emails, and a box of approved bulk references. What is missing is the link between them. A swatch without its article code cannot be reordered, and a certificate without its scope line cannot support a claim. The library is the structure that connects the three.
It is also not a design mood board. A mood board selects by appearance; a library selects by whether a material can be bought, tested, repeated and repaired. Keeping the two separate is important, because a beautiful material that cannot be re-ordered in six months is a design risk rather than an asset, and the library is the place where that risk becomes visible before the collection is committed.
Programme parameters at a glance
| Parameter | Commitment |
|---|---|
| Minimum order quantity | 500 pieces per style / colourway |
| Sampling lead time | 6–10 working days (12–15 days for complex constructions) |
| Bulk production lead time | 35–50 days after sample approval |
| Inspection standard | AQL 2.5 (ISO 2859-1), single sampling, general level II |
| Incoterms | FOB Xiamen — EXW, CIF and DDP available on request |
| Payment terms | T/T 30% deposit / 70% before shipment; L/C at sight for container loads |
| Verified production floor space | 4,950 m² (SGS-verified) |
| Team | 137 staff across 7 production lines |
| Equipment | 149 machines including computerised saddle-stitch and edge-painting units |
| Monthly capacity | 200,000 pieces |
| Entity | Founded 2014; principal has worked in bag development since 2004 |
The five records every sample should carry
The reason most libraries fail is that they store one record and assume the rest will be found later. A sample without documents is a swatch; a document without a sample cannot be judged against a new lot; commercial terms without performance data lead to a material being selected on price and rejected on durability. The five records below are the minimum set, and the discipline is to refuse to add a sample to the library until all five exist, even if three of them are marked as pending.
| Record | What it contains | Who maintains it | What it prevents |
|---|---|---|---|
| Physical | A labelled swatch, cut generously enough to fold and to stitch | Material or programme coordinator | Selecting a material that cannot be delivered or that behaves differently at the seam |
| Digital | A photograph under standard lighting with the code visible in the frame | Programme coordinator or designer | Disagreements about which swatch was approved and what it looked like |
| Document | Certificates, audit reports, declarations and test reports | Compliance or sourcing lead | Claims that cannot be substantiated and re-testing that has already been done |
| Commercial | Price basis, minimum order quantity, lead time and validity date | Sourcing lead | Quotes that expire unnoticed and budget assumptions that no longer hold |
| Performance | Test results for abrasion, tear, flex, fastness and any functional claim | Technical or quality lead | Specifying a material that fails the application it was chosen for |
What to collect first, and in what order
A library built from everything available becomes unusable within a season, so the first task is a filter. Collect in descending order of how often the material appears in the programme and how much trouble it causes when it changes. Leather bodies, linings and hardware finishes come first in almost every bag brand; trims, threads and packaging follow once the core is stable. The sequence below is the one we recommend when a brand is starting from a cupboard of unsorted swatches.
- Audit what already exists: gather every swatch, certificate and quotation into one place before discarding anything.
- Rank materials by the frequency with which they appear in your range, and by how much rework a substitution causes.
- Start with the highest-ranked material family and build the full five records for ten to fifteen references.
- Add the lining and hardware families next, because they cause the second-largest share of late-stage problems.
- Photograph everything in one session under the same lighting so that the digital record is genuinely comparable.
- Match each document to its swatch and mark clearly any record that is missing or has expired.
- Write a one-page index that lists every reference by family, mill and status, and keep it at the front of the library.
- Set a review date, and treat the library as a live system rather than a one-off project.
Labelling, storage and shelf life
Materials change in storage, and a library that ignores this will mislead rather than inform. Leather absorbs and releases moisture, so a sample kept in a dry office becomes firmer and lighter in tone than the same material held in a humid warehouse. Coated fabrics can hydrolyse. Uncoated brass tarnishes. Vegetable-tanned leather darkens with light exposure. A library is only comparable if all its samples have been stored under similar conditions, which is why a controlled environment is worth the small investment it requires.
| Material | Storage priority | Shelf-life consideration | What changes over time |
|---|---|---|---|
| Vegetable-tanned leather | Moderate humidity, away from direct light, flat or loosely rolled | Long-lived if protected from light and moisture | Deepens in colour and firms slightly; a stored sample will not match a new delivery exactly |
| Chrome-tanned leather | Moderate humidity, protected from heat | Stable for a long period if not compressed | Surface finish can dull; flexed areas may crease permanently |
| Coated or laminated fabric | Cool, dry, away from heat sources | Finite; polyurethane coatings degrade over time | Coating can stiffen or become tacky, particularly in humid conditions |
| Textile linings | Dry, out of direct light | Long-lived; synthetics are stable | Colour can fade under ultraviolet exposure; dust accumulates in open storage |
| Hardware | Dry, individually bagged | Long-lived, but uncoated metals tarnish | Plated finishes can dull; brass patinates; salt exposure accelerates corrosion |
| Adhesives and finishes | Sealed, within the manufacturer's date range | Strictly limited; these expire | Bond strength and cure behaviour change with age, so expired stock must not be approved |
| Thread and edge paint | Dry, sealed, away from light | Limited for reactive products | Colour and consistency shift, which affects visible stitching and edges |
Development samples and retained bulk references
There are two distinct classes of sample in a bag programme, and confusing them is the source of most disputes. A development sample shows what a material could be, and it exists to help a decision. A retained bulk reference shows what a material actually was when it was delivered, and it exists to resolve a complaint. Both belong in the library, in physically separate and clearly marked storage.
| Sample class | Purpose | When it is taken | How it is used |
|---|---|---|---|
| Mill or development swatch | Material selection and comparison | During research, before any order | Judged on hand, colour, drape and the mill's stated performance data |
| Sample-bag material | Confirm the material as assembled | At the sampling stage | Compared against the bulk delivery to detect drift |
| Approved standard | Define what the brand has signed off | At sample approval, signed and dated | The benchmark for every subsequent delivery and for incoming inspection |
| Retained bulk reference | Evidence of what was actually shipped | From every bulk lot, sealed | Used to resolve disputes and to judge whether a future lot is acceptable |
| Customer return sample | Diagnose a field failure | On complaint | Traced back through the records to establish whether the failure is material or use related |
| Archive pair | Preserve the visible ageing behaviour | One piece stored away, one handled | Shows how the material behaves over time rather than on the day it arrived |
Using the library to shorten the next season
The commercial return on a library appears in the second season, and it appears in three ways. Material selection becomes a comparison rather than a search, because the shortlist already exists and already carries its documents. Sampling becomes faster, because the construction is matched to a material whose behaviour is known. And quality becomes more predictable, because incoming deliveries are compared against a physical standard rather than a memory.
| Stage of the next programme | What the library changes | Time effect | Consequence of having no library |
|---|---|---|---|
| Brief and research | A qualified shortlist is available immediately with documents attached | Days rather than weeks of mill correspondence | The same materials are researched again by a different person |
| Sampling | Material behaviour, shrinkage and finishing response are already known | Fewer sample rounds and fewer surprises on the bench | Repeated failures that were solved on a previous collection |
| Costing | Price basis, minimum quantity and lead time are on record and dated | Faster, more reliable landed-cost models | Quotes that expire and budget assumptions that break at approval |
| Compliance | Certificates are already filed and their expiry dates are tracked | Documentation request becomes a retrieval rather than a chase | Late evidence collection that delays a purchase order |
| Bulk inspection | Incoming material is compared against a signed standard | Objective accept and reject decisions | Subjective arguments about shade and hand |
| Repair and aftercare | Spare materials and hardware are traceable to their source | Repairs become possible after the collection closes | A repair promise that cannot be kept |
Governance: who owns the library and how it stays current
A library decays without an owner. Samples are borrowed, documents expire, quotes lapse and new swatches are added without records. The governance model that works in practice is simple: one named person owns the physical and digital records, every addition follows the same five-record rule, and the whole library is reviewed on a fixed cycle. Where a brand works with a production partner, the partner can maintain the technical and performance records while the brand retains ownership of the standards.
- Name a single owner for the library, with a documented deputy to cover absence.
- Apply the five-record rule without exception; a sample without documents is stored outside the library until the records arrive.
- Review expiry dates quarterly, covering certificates, audit reports, quotations and adhesives.
- Re-photograph a sample whenever the mill changes the article or the colour reference.
- Log every removal from the library, because a swatch on a designer's desk is a swatch unavailable for comparison.
- Retire references that can no longer be bought, and mark them as archived rather than deleting them.
- Keep the library accessible to the people who make material decisions, not locked in a warehouse.
Building the library with a production partner
A production partner can accelerate a library considerably, because the tanneries and mills that supply a factory already hold the swatches, the certificates and the performance data in a form the partner understands. Our programme team maintains a working material library covering the leathers, linings, fabrics and hardware finishes most often requested by brands, and it is updated as mills change their ranges. What a brand should retain is ownership: the approved standards, the signed references and the commercial records belong with the brand, not with any supplier.
The practical arrangement that works best is a joint reference set. The partner holds the breadth, covering hundreds of stock references with their technical data, and the brand holds the depth, covering the twenty or thirty materials that define its own range with signed standards and retained bulk references. Bulk deliveries are then checked against the brand's standards, with the partner's library used to propose alternatives quickly when a mill discontinues a line.
| Responsibility | Held by the brand | Held by the production partner |
|---|---|---|
| Signed approved standards | Yes; this is the definition of what the brand accepts | Held as a copy for incoming inspection |
| Breadth of stock references | A shortlist only | Yes; the working range of leathers, linings, fabrics and hardware |
| Certificate and audit records | Yes, including expiry tracking for claims the brand makes | Collected and forwarded, with verification of supplier and scope |
| Performance test data | The data for materials the brand actually uses | Wider panel data across the stock range |
| Commercial records | Final price, quantity and lead time as contracted | Current price basis, minimum order quantity and availability |
| Retained bulk references | Yes; one sealed set per lot | Retained for production traceability |
| Alternative proposals | Assessed against the brand's standards | Proposed when a mill discontinues or a material fails |
A library of this kind changes the rhythm of development. Rather than restarting a material search each season, the brand works from a qualified shortlist, samples in the standard six to ten working days or twelve to fifteen for complex constructions, and moves to bulk thirty-five to fifty days after approval, with inspection to AQL 2.5 under ISO 2859-1, single sampling, general level II, and shipment FOB Xiamen or on EXW, CIF or DDP terms as the programme requires. The library does not remove the work of developing a collection, but it removes the repetition, which is where the time was being spent.
Start the library with a qualified shortlist
Send us your current material range and the categories you plan to add, and we will return a qualified shortlist with swatches, documents and performance data attached, ready to become your first signed standards. Email service@junyuanbags.com or message the programme team on WhatsApp at +86 177 5002 0688.
Frequently Asked Questions
What is a material sample library in bag sourcing?
It is a structured set of material references in which every sample carries its article code, its supporting documents, its commercial terms and its performance data. It is a working instrument used to select, compare and inspect materials rather than an archive of interesting swatches. A library of fifty fully recorded references is worth more than a cupboard of five hundred unlabelled ones, because only the recorded samples can be reordered or defended.
How big should a leather swatch be?
Large enough to fold, to handle and ideally to stitch. A generous square is the practical minimum, and a rectangle that can be folded over a seam gives far more information about how the material bends, creases and finishes at an edge. Swatches the size of a business card show colour and little else, which is why they cause disagreements later when the material behaves unexpectedly at a fold.
How long do leather samples last in storage?
Chrome-tanned leather stored at moderate humidity and away from heat and light will remain usable for a long time, while vegetable-tanned leather will darken and firm gradually. The material does not expire in the way an adhesive does, but its appearance does drift, so an old sample should not be used as a colour reference for a new order. Retained references from actual bulk deliveries are more reliable for that purpose.
What documents should sit alongside each material sample?
At minimum, the certificate or declaration that supports any claim you intend to make, plus the relevant test report where a performance or safety result is asserted. Add the audit report for any facility claim, the price basis and validity, the minimum order quantity per colourway, and the lead time. Mark anything missing or expired so that the gap is visible rather than assumed to be covered.
What is the difference between a development sample and a retained bulk reference?
A development sample shows what a material could be and exists to support a selection decision. A retained bulk reference shows what was actually delivered and exists to resolve a dispute or judge a future delivery. Both belong in the library but should be stored separately and labelled clearly, because treating an aspirational swatch as though it were a delivered standard is one of the most common causes of quality arguments.
How should bag hardware be stored in a sample library?
Individually bagged, dry and away from sunlight, with the plating specification recorded against each component. Uncoated brass will patinate and plated finishes will dull, so a hardware reference is only useful alongside the supplier, the plating thickness and the date it was approved. Where a brand uses several finishes, a physical standard for each finish should be sealed and kept separately from the working samples.
How often should a sample library be reviewed?
Quarterly review of expiry dates is the practical minimum, covering certificates, audit reports, quotations and any adhesives or reactive products. A fuller annual review should confirm that each material can still be bought, that the pricing is current, and that the photographs and documents still match the physical samples. Materials that can no longer be sourced should be moved to an archived section rather than deleted, because the record remains useful.
Who should own the material library inside a brand?
One named person should own it, with a documented deputy so that the system survives holidays and staff changes. In small brands that role usually sits with the product developer or the sourcing lead; in larger ones it may sit with a technical manager. What matters is that additions follow the same rule, that the physical and digital records stay in step, and that the person making material decisions has direct access to the library.
Can a production partner build the sample library for a brand?
A production partner can accelerate it substantially, because tanneries and mills already supply swatches, certificates and performance data to the partner in a usable form. The partner is well placed to hold the breadth of stock references and to propose alternatives when a mill discontinues a line. The brand should still hold its own signed standards, retained bulk references and commercial records, because those define what it has agreed to accept.
How does a sample library shorten development time?
It converts a search into a comparison. Instead of corresponding with mills to find materials and gather documents, the brand starts from a qualified shortlist whose behaviour and paperwork are already known. Fewer sample rounds are needed because shrinkage, finish and construction response have been recorded previously, and incoming deliveries are checked against a physical standard rather than a memory of what was approved.
Should a sample library include pricing and lead times?
Yes, with dates attached. Price basis, minimum order quantity per colourway and lead time should be recorded beside the material, together with the date the quotation was issued and the date it expires. Without a validity date the library rapidly becomes misleading, and a landed-cost model built on an expired quote is worse than having no model at all.
What is the most common mistake when building a sample library?
Collecting swatches without records. The material arrives, the sample is filed, and the certificate stays in an email thread while the quotation is never written down. The library then looks comprehensive but cannot answer the questions that matter, which are whether a material can be reordered, whether it complies, and whether a new delivery matches what was approved. The five-record rule prevents this.
How many materials should a first sample library contain?
Aim for ten to fifteen fully recorded references in the material family you use most, then add lining and hardware families. That is usually enough to cover the majority of a season and to establish the working method. Depth matters more than breadth in the first year, because a small library that answers every question is more useful than a large one that answers none.
Do sample libraries need to be physical in a digital age?
They do, because material behaviour cannot be judged from a screen. Colour shifts between devices, drape and hand are invisible in a photograph, and a stitched edge cannot be assessed digitally. The efficient model is digital for storage, comparison and document management, and physical for the samples themselves, with the two linked by a consistent article code and a photograph taken under standard lighting.