Bag Tech Pack vs Sketch: Why Factories Need Both

The argument for sending both is not tidiness. It is that the two documents fail in different directions. A pack with no sketch is buildable and lifeless: every field can be filled correctly and the bag still comes out looking wrong, because proportion is not a field. A sketch with no pack is full of intent and unbuildable, because there is nothing to cut.

What buyers actually send, most of the time, is a photograph of a competitor's product with a note about the logo. That is neither document. It produces a sample that is approximately right, a price that is approximately right, and a second sample that costs another 6-10 working days.

The fix is not a longer pack. It is two documents that admit they do different jobs: one that says what the bag is for and one that says what it is made of. Buyers who adopt the discipline usually find it pays for itself on the first style, because the second sample stops happening and the price stops moving.

What Does a Sketch Carry That a Tech Pack Cannot?

Proportion and intent. The relationship between the height of the body and the depth of the gusset, the way a strap meets the top of the panel, the visual weight of a trim line — none of these is a dimension and all of them are what the customer sees. A pack can state that a panel is 240 mm by 180 mm and still never say whether the bag reads as tall or as squat.

The second thing a sketch carries is the decision about what not to build. A designer's drawing shows the two or three features that matter and quietly omits the rest; that omission is a decision, and it is lost the moment somebody tries to reconstruct intent from a materials list.

What Does a Tech Pack Carry That a Sketch Cannot?

Everything that costs money. The sketch cannot say which mill supplies the shell, what the seam allowance is, how many stitches per inch close the gusset, what pull the side-release buckle has to survive, or what the carton count is. Those are the values the quotation is arithmetic on, and none of them is drawable.

The sketch also cannot carry what a laboratory can test. A claim that the fabric meets a colourfastness grade, that a weld holds, or that a component is inside a restricted-substance limit has to be written as a standard number and a limit. A drawing of the same claim is a picture of a bag.

How Do Numbered Callouts Join the Two Documents?

By giving every component on the sketch a balloon number and every line in the materials and hardware schedules the same number. The sketch then points at the pack and the pack points back. A production team reading the pair never has to guess which buckle the drawing means, and a buyer reviewing a sample can say "callout 7 is wrong" rather than "the buckle".

Keep the numbers stable across revisions. Once callout 7 has meant the side-release buckle in two revisions it must keep meaning that, even if the buckle moves position on the drawing. Renumbering callouts between revisions is the fastest way to make an old email thread dangerous.

What Happens When Only the Sketch Is Sent?

The production team fills the gaps with what it has on the shelf, and the sample that comes back is a reasonable bag that is not the buyer's bag. The shell is a stock 600D rather than the coated cloth the buyer imagined; the binding is 20 mm rather than 25 mm; the hardware is what the line uses every day. Every one of those is a correct answer to a question that was never asked.

The price that comes back with that sample is then wrong in a specific and awkward way. It is not inflated — it is a price for a different product. When the buyer says "no, I wanted a coated shell", the price moves, and the move reads as a change of mind rather than as the correction it actually is.

There is a second cost that lands later. The stock components chosen to fill the gaps are the ones already on the shelf, which means they are the ones already used on somebody else's programme. Where the buyer then approves that sample and goes to bulk, the bag inherits a bill of materials nobody selected on purpose.

What Happens When Only the Tech Pack Is Sent?

Something subtler and, on a complicated style, worse. Every field is filled and every dimension is right, and the bag still comes out wrong, because proportion is not a field and the pack never said what the silhouette was meant to be. The panels are the specified size and the shape is not the one in the buyer's head.

This is also where a camera bag programme tends to hurt. A divider set can be specified to the millimetre and still leave a body-and-two-lens kit sitting proud of the top edge, because the proportion between the interior height and the divider height was never drawn. The numbers were right and the bag does not work.

This failure is expensive to fix because it is hard to describe. A wrong dimension can be corrected in writing. A wrong proportion produces a conversation in adjectives — too tall, too soft, too busy — and each round of adjectives costs another sample.

When Is a Sketch Enough and When Is It Not?

A sketch is enough for a conversation and not enough for a quotation. It is the right artefact for the first twenty minutes: is this the shape, is this the proportion, is this the customer. It stops being enough the moment somebody asks what it costs, because cost is arithmetic on fields that a sketch does not have.

There is one narrow case where a sketch and a sample are enough: an exact repeat of a style already in the file, where the pack exists from the previous run and the only change is a colourway. Even then, the repeat is issued against the old pack with a new revision letter rather than against the sketch.

How Do the Two Documents Behave on a Repeat Order?

Better than they did first time, provided the pair was filed properly. A repeat runs against the existing pack with a new revision letter and, where nothing about the shape has changed, without a new sketch. The golden sample from the previous run is the reference for colour and hand, and the pack already carries the mill, the shade reference and the stitch schedule.

What breaks a repeat is a silent change on either side. A mill substitution on the shell, a different foam supplier, or a hardware part bought from a second source all move the bag away from the approved pair without changing a single line of either document. Where a programme intends to repeat, the pack should state which components are single-source and what approval is needed to change one.

What Does a Good Sketch Actually Show?

Four views at a stated scale, with the strap and the trim drawn rather than described, and a note of what the bag carries. That last item is the one buyers omit and the one that changes the drawing: a kit bag drawn around wet kit needs a drained base, and the same bag drawn around dry kit does not.

The sketch should also mark what is not negotiable. If the proportion of the body to the gusset is the whole point of the design, mark it, because a production team reading a drawing without that mark will treat every dimension as equally adjustable and will happily produce a bag that satisfies all of them and none of the intent. One marked proportion on a drawing is worth a page of written explanation, because it survives being read by somebody who does not speak the buyer's language.

How Should the Two Documents Be Filed and Sent?

As one file, in one email, with the sketch inside the pack. A pack sent as a PDF and a sketch sent as an image in a separate message get separated within a day, and from that point the production team is building from half a specification. One file, one revision letter, one date.

Then hold the pair steady through the programme. Sampling runs 6-10 working days from the signed pair, bulk runs 35-50 days from written sample approval, and inspection runs at AQL 2.5 under ISO 2859-1. Where a dispute arises about what was specified, the pair together is the evidence; either alone is half of it. The reorder consistency note covers what happens when the pair drifts between runs.

Frequently Asked Questions

People Also Ask