Knitted Bag Decision Tree: Five Gate Questions and Their Thresholds

Five Gates, in This Order

A decision tree is only useful if the gates are ordered so that the fastest disqualifier comes first. For bag sourcing the order is volume, refresh rate, load, artwork and timeline. Volume eliminates routes on economics, load and artwork eliminate them on physics, and timeline decides between whatever survives. Running them in any other order wastes meetings on routes that a later gate would have removed.

Each gate has a threshold rather than a preference, because a threshold can be checked against a number in the brief. Write the number down at each gate and move on. By gate three most briefs have one surviving route, and by gate five the specification is half written.

Gate One: Annual Volume Per Style

Below about 5,000 pieces per style a year, knitting wins on minimum fit. At 100 pcs per style and colourway, a range can test six shades for the quantity a cut and sew route needs for one, and a woven route carrying a pattern, grading set and cutting die at roughly USD 640 per style has to amortise that over a larger run to make sense.

Above about 50,000 stable pieces a year of a single design that never changes, the woven route wins on unit price and there is no argument to be had: the setup cost is amortised to nothing and the per-unit material saving compounds. Between those two thresholds the gate does not decide, and the answer comes from gate two. Note that the threshold is per style, not per programme — a 60,000-piece programme across twelve styles is 5,000 per style, and knitting wins.

Gate Two: How Often the Design Changes

Count the change events in a year: new shades, new sizes, new artwork, new hardware. Below three, the woven route usually keeps its unit-price advantage. At three or more, knitting overtakes it, because each change on a woven route costs about USD 380 and about 12 working days in a new marker, a new screen and a fresh cutting trial, while a knitted change is a yarn carrier swap.

This gate is the one most buyers get wrong, because change events are invisible on a purchase order. A range that looks stable in January often refreshes four times by November as the retail calendar moves. Ask the merchandising team rather than the sourcing team how many changes to expect, and add one to whatever they say. The threshold sits at three, and it is rarely a close call once the number is honest.

Gate Three: Static Load and Load Character

Below 15 kg, a knitted body with a zoned base and a continuous strap is the right answer and reaches about 780 N on the handle root to ISO 13934-1. Between 15 and 20 kg it still works with a 900D construction and an inserted 0.8 mm base sheet. Above 20 kg, or wherever the load is sharp or irregular, the gate closes and a woven shell with a boarded base takes over.

The reason is puncture rather than tensile strength. An uncoated 900D knit fails at about 40 N on a 1 mm probe, where a 1680D woven nylon needs about 120 N. A bag carrying tools, tent pegs or camera equipment will be punctured from the inside, and no yarn count prevents it. This is the first of the two gates knitting loses outright, and it should be recorded without hedging.

Gate Four: What the Artwork Demands

Two colours or fewer, in a bold mark, and a knitted-in logo works well with a practical minimum stroke near 1.5 mm. A heat transfer extends this to about 0.4 mm but adds a polymer layer that complicates a monomaterial claim. Photographic or four-colour process artwork closes the gate: a flat woven panel resolves detail to about 0.2 mm and a knitted surface cannot.

This is the second gate knitting loses outright, and it is the one that most often splits a range. The usual resolution is per SKU rather than per range: the shopper and the tote go to knitting, and the gift item carrying the photographic seasonal image goes to a woven panel. Forcing one route across a range with mixed artwork requirements produces a compromise that satisfies nobody.

Gate Five: The Launch Window

Add the sampling window to the bulk window and compare it with the shelf date. A knitted route runs 7 working days for sampling and 30 days for bulk, about 37 working days end to end. A woven route needs 12 to 15 working days for a revised pattern and marker and a similar bulk window, roughly 47 working days at the short end. Under a 45-working-day window, knitting is the only route that fits.

The clause that matters is the second sampling cycle. Most programmes need two, because the first sample surfaces a change. Two knitted cycles cost about 14 working days and still fit inside 45; two woven cycles cost 24 to 30 and do not. Ask for the sampling window in writing before promising a launch date, and book machine capacity at the same time, because 7 production lines at about 200,000 pcs a month are a finite resource.

Recommended Scenarios by Gate Combination

Four combinations cover most briefs. Low volume and frequent change is the clearest knit case: a seasonal range of six shades at 800 pieces each, refreshing four times a year. Tight window and low volume is the second: a campaign bag needed in six weeks. Both are decided by gate one or gate two and rarely need gates three to five at all.

High volume and no change is the clearest woven case: 60,000 pieces of one shopper in one shade, sold for three years. Heavy or sharp load is the second: a trade kit above 20 kg. Both are decided at gate one or gate three. The interesting briefs sit in the middle, and for those the honest answer is usually a split range rather than a single route.

When Two Gates Disagree

Disagreements are normal and informative. A brief at 3,000 pieces a year with photographic artwork fails gate one in favour of knitting and gate four in favour of woven. The resolution is not to split the difference; it is to ask which gate carries the cost. Here gate four does, because a knitted surface cannot deliver the image, so the range goes to woven and absorbs the higher minimum.

The other common conflict is high volume with frequent change. Above about 50,000 pieces a year the woven unit-price advantage is large, but at four changes a year the refresh cost partly erodes it. Run the numbers rather than the gates in this case: the crossover typically sits near three changes a year, and the total cost sheet rather than the tree should decide.

Recording the Decision So It Survives Handover

Write the gate outcomes into the specification pack: the volume per style, the expected change count, the rated load, the artwork type and the launch window, together with the route each gate pointed to and the final decision. Twelve months later, when a colleague asks why the range went to knitting, the answer should be in the file rather than in someone's memory.

Then convert the decision into numbers: yarn count and gauge, finished mass with a ±5% tolerance, abrasion target to ISO 12947-2, tensile target on the handle root to ISO 13934-1, and inspection to AQL 2.5 under ISO 2859-1. Sampling runs 7 working days and bulk 30 days, FOB Xiamen with CIF and DDP options. A decision tree chooses the route; the specification is what makes it repeatable.

Frequently Asked Questions

People Also Ask

Two Things the Tree Cannot Decide

The tree does not decide taste. A brand that has built its identity on canvas texture may choose woven regardless of what the gates say, and that is a legitimate decision as long as it is made explicitly rather than dressed up as a sourcing conclusion. Write it down as a brand decision so nobody later tries to justify it with the wrong numbers.

The tree also does not replace a physical sample, and it does not tell you how the market will receive the product. Two routes can both pass every gate and still differ in how the bag looks on a shelf, which is a merchandising question rather than a sourcing one. Every gate threshold here is a prediction, and one sampling round at 7 working days confirms or corrects it for less than the cost of a wrong bulk run. Sample the winning route and the runner-up where the development allowance permits, because the comparison is worth far more than the saving. For the wider picture, see the knit versus traditional bag guide.