Skip to content

Home textile & made-ups: find short sets and the packing spec by SKU

Ask which components an order is short of, what the buyer’s packing spec says for a SKU, or which DST file was approved. The agent answers from your set sheets, output rows and specs, shows what it used, and sends or changes nothing unless an admin approves.

Works from: Set assembly sheet, Retail-ready packing spec, Embroidery DST file registry, Buyer routing guide

Where sets, labels and cartons go wrong

Most errors start early: a cut plan, a miscount, an old sheet, an edited file. They show up later, at set assembly, at the pre-FRI audit or at the buyer’s DC. Each one traces back to a record your floor already keeps.

  • A SKU comes up 12 sets short, and no one knows where

    The playbook’s example: one SKU in a 1,500-set order is 12 sets short. A sheet set may be a fitted sheet, a flat sheet and two pillowcases, each made on its own line. The made-ups manager does not yet know if the cut was short, assembly miscounted or the FG count is wrong. If the buyer will not take sets from the next batch, the playbook’s fallback is to ship the balance later, and some buyers refuse it.

    $500 per shipment

    One US home retailer’s 2025 vendor manual: when a PO line arrives in more than one shipment, the first is marked short, later shipments against that line are refused, and $500 is charged per shipment.

    Source: Williams-Sonoma, Inc., Vendor Compliance Manual for Domestic Shipment and Domestic Dropship, Chapter 10: Vendor Non-Compliance (opens in a new tab), 2025.

  • Retail-ready packing that fails on a ticket, a poly bag or a label

    Direct-to-shelf buyers set the poly bag size and seal, tag and barcode placement, the fold and the master pack. The usual root cause, says the playbook: an operator reads the spec one way, the buyer another. It warns that buyers treat packing as compliance, not quality, so one packing failure can fail a shipment of perfect sets. Caught at the pre-FRI audit, it means a re-pack; caught at the buyer’s DC, a debit memo.

    $500 + $2 a unit

    What one US home retailer’s 2025 domestic-shipment manual charges for a missing or incorrect SKU or price ticket, or for poly bags missing or unsealed. Foreign vendors paid by wire have it deducted from their next payment.

    Source: Williams-Sonoma, Inc., Vendor Compliance Manual for Domestic Shipment and Domestic Dropship, Chapter 10: Vendor Non-Compliance (opens in a new tab), 2025.

  • SKU explosion: the right sheet for the right SKU

    The playbook warns that a “simple” duvet program can reach 120 SKUs once colour, size and pack configuration multiply. Each SKU has its own set composition, barcode, label set and master pack. Set assembly works from printed sheets that go out of date, and supervisors phone the office to ask which one is current. One old sheet becomes a short set at assembly or a wrong label at the buyer’s DC.

  • A logo stitches wrong, and no one can say which DST file ran

    The buyer signed off one DST file. A copy edited later can reach the machine, and often nobody recorded which file ran. The playbook’s first check is the file’s hash against the approved one: a mismatch is a version-control failure, so pull the right file, re-embroider on samples and restart. The embroidery in-charge loses samples and machine time, and an edited file needs a new buyer approval.

Short sets and packing specs, answered from your own records

Each job starts from a record the made-ups floor already keeps: set assembly sheet, packing spec, output sheet, DST registry. Answers name the rows or document used. Messages, sheet rows and ERP changes stay drafts until an admin approves or declines them.

  • Short components per set, summed from your output rows

    Answers with sources

    Keep one current set assembly sheet per SKU, uploaded or in a Google Sheet, and upload the stitching and hemming output as a CSV file or an Excel file’s first sheet. Ask about one component of an order and the agent sums the pieces stitched from the rows, sets it beside what the set assembly sheet it names needs, and shows the rows used. Supervisors can ask it which sheet is current instead of phoning the office.

    Ask“Order 2208, queen set in white: how many pillowcases are stitched, and how many does the set assembly sheet need?”

  • The buyer’s packing spec, quoted for one SKU

    Reads your documents

    Upload the retail-ready packing spec and the buyer’s routing guide. Ask what a SKU needs and the agent quotes the poly bag size, tag and barcode placement, carton marks and master pack quantity, naming the document it read. Keep your label plan in a sheet and it sets each planned value beside the spec’s and lists the differences it finds. A scanned spec has no text to read.

    Ask“What does the buyer’s packing spec require for SKU QS-WHT-04: poly bag size, tag and barcode placement, master pack quantity?”

  • A packing readiness summary, built as a sheet

    Creates files

    Before packing starts, name an order’s SKUs and ask for a readiness summary. For each SKU you name, the agent sums each component’s pieces from the output sheet, sets them beside the set assembly sheet and the packing spec, and builds a spreadsheet: complete, short or blocking. It adds the playbook’s pre-run checks, such as poly bag and tag stock and the first-50 audit, and marks any it found no record for.

    Ask“Prepare the packing readiness summary for order 2208 as a sheet, for SKUs QS-WHT-04 and QS-WHT-02: complete, short or blocking.”

  • The short list, sent to the stitching supervisor

    Acts in your tools

    Once you know what is short, the agent drafts a WhatsApp Business or Slack message to the stitching supervisor with each short component by SKU and lot, and a row for your shortage sheet in Google Sheets. A department admin approves or declines each one. Nothing is sent or written unless approved.

    Ask“Draft a WhatsApp to the stitching supervisor listing the pillowcases and flat sheets still short for order 2208.”

  • An earlier planned start for the make-up batch

    Reads your ERP

    If the buyer accepts the short sets from the next batch, the playbook makes them up there rather than shipping a balance. Ask the agent to draft an earlier planned start for that batch’s production order in Odoo. A department admin approves or declines it, and it runs only if approved. It cannot change a quantity or a set composition. Odoo is the only ERP it can write this to.

    Ask“The buyer accepts the 12 short sets from the next batch. Draft a planned start of the 14th for production order MO-3391.”

A new SKU’s first sets, twelve sets short and a logo that looks off

  • The first 50 sets of a new SKU

    Packing supervisor

    The first 50 sets of a new SKU are packed, and the pre-FRI audit starts in the morning.

    1. Asks

      “Make the pre-FRI checklist for SKU QS-WHT-04 from the buyer’s packing spec and routing guide.”

    2. The agent

      Reads the packing spec and the routing guide, and lists each check with the buyer’s wording: poly bag size and seal, tag and barcode placement, the fold, the master pack quantity. It adds the carton marks the playbook names (SKU, size, colour, quantity, lot number, factory code, master pack ref), notes any the spec leaves out, and builds the checklist as a spreadsheet, each line citing its document.

    3. Result

      The supervisor audits the first 50 sets against the buyer’s own wording, each line traced to its document, before the rest of the SKU is packed.

  • Twelve sets short at assembly

    Set assembly supervisor

    Set assembly closes the bundles for a 1,500-set order, and one SKU is 12 sets short.

    1. Asks

      “SKU QS-WHT-04 is 12 sets short. Show the cut, stitched and packed pillowcases by lot, and draft a note to the stitching supervisor with what to make up.”

    2. The agent

      Reads the pillowcase rows for QS-WHT-04 in the uploaded cutting, output and packing sheets, lists the pieces each stage recorded by lot beside what the set assembly sheet needs, and points to the stage where a count drops, citing the rows it used. It quotes the playbook: the common cause is an assembly miscount. It drafts a WhatsApp Business message to the stitching supervisor with the pieces still missing.

    3. Approval

      The message waits for the made-ups manager, as department admin, who approves or declines it. It is sent only if approved, and the audit trail records who decided.

    4. Result

      The supervisor sees at which stage the pillowcases for the 12 sets went missing, and the make-up request is sent to the stitching supervisor only if the manager approves it.

  • A logo that looks slightly off

    Embroidery in-charge

    The first piece of a new design comes off the embroidery machine, and the logo looks slightly off.

    1. Asks

      “Which DST file, version and file size are approved for design EM-114, and do the ones on last night’s run sheet match?”

    2. The agent

      Reads the DST file registry: file name, version, file size, hash, placement and the buyer’s approval date. Sets the file name, version and size on last night’s typed run sheet beside the approved ones, and says whether they match or a value was never recorded. If the size differs, it quotes the playbook: pull the approved file, re-embroider on samples, restart, and get a new approval for any edit. It never opens the DST file itself.

    3. Result

      The in-charge knows whether the machine ran the approved file before more pieces are embroidered, with the registry line cited.

What a short set or a late PO costs a made-ups unit

Published, cited figures: one US home retailer’s charges for split, late and mis-packed orders, and a small 2008 survey of Pakistani factories. None is a Mavick result. Before a pilot, record set-mix accuracy, short-ship rate and chargebacks, then compare.

  • Fewer orders that ship short of a full set

    3.14%

    Average stitching loss reported by eight Pakistani garment factories, three of them making home textiles, in a small 2008 survey. It ran from 1.5% to 5%. Packing losses averaged 1.54%.

    How the agent helps

    A piece lost in stitching or packing leaves its set short of a component. The agent does not count pieces on the floor. When you ask, it sums the pieces made of one component for a SKU from your uploaded output rows, sets the total beside the set assembly sheet, and shows the rows it used.

    Track: Set-mix accuracy %, by order

    Source: CABI South Asia, Cotton Value Chain: Skill Gap Analysis in Garments Sub-sector (opens in a new tab), 2008.

  • Short components found while the ship date still holds

    5% of cost of goods

    One US home retailer’s 2025 vendor manual charges 5% of cost of goods on a PO 21 or more days late, and 3% at 8 to 21 days. Global orders are included. The vendor can pay air freight to the DC instead.

    How the agent helps

    When you ask, the agent reads an order’s delivery date in your order sheet and its rows in your output sheet, and names the SKUs it finds short of a component, citing the rows. It compares recorded quantities and dates; it does not forecast. A note to the supervisor goes out only if an admin approves it.

    Track: Daily set output against plan, by SKU

    Source: Williams-Sonoma, Inc., Vendor Compliance Manual for Domestic Shipment and Domestic Dropship, Chapter 10: Vendor Non-Compliance (opens in a new tab), 2025.

  • Less fabric lost at cutting

    5.33%

    Average cutting loss reported by eight Pakistani garment factories, three of them making home textiles, in a small 2008 survey. Losses ran from 5% to 6%, so almost every firm faced the problem.

    How the agent helps

    The agent does not plan markers or read the cutting table. Upload the cutting log as Excel or CSV, and it totals the fabric issued for a SKU from the rows, in the unit the log uses (metres, or kilos for terry). Ask again for sets cut, and it sets consumption per set beside the cut plan’s figure, showing the rows it used.

    Track: Cutting waste %, by set type

    Source: CABI South Asia, Cotton Value Chain: Skill Gap Analysis in Garments Sub-sector (opens in a new tab), 2008.

What to baseline before a pilot

Measure each one before the pilot starts, then again at its end.

Home Textile & Made-Ups: the KPIs to baseline before a pilot, and why each matters
KPI to baselineWhy it matters
Set-mix accuracy %The playbook’s target is 100%: zero short-set claims. It calls set mix the number big retail buyers care about most, and warns that the chargeback for getting it wrong can exceed the margin.
Retail-ready audit pass-rate, pre-FRIThe playbook’s target is 99% or better, and it has a failed audit raised with the QA head, the made-ups manager and Merchandising within 2 hours, with a rework plan within 24.
Barcode scan-rate at packingThe playbook’s target is 100%; anything less points to a printing or label problem. The agent reads it from a scan log you upload or keep in a sheet.
Embroidery first-pass approval %The playbook’s target is 95% or better. When a logo comes out wrong, its first check is the DST file’s hash against the file the buyer approved.
Cutting waste %, by set typeThe playbook tracks it per set type and sets no target. Read it beside the planned fabric consumption per set for each SKU.
Daily set output against plan, by SKUThe playbook reports set output daily. Reading it against plan by SKU shows a component falling behind before it becomes a short set at assembly.

What the agent works from

Records it reads

  • Set assembly sheet
  • Retail-ready packing spec
  • Embroidery DST file registry
  • Buyer routing guide
  • Cut plan per set
  • Stitching and hemming output sheet
  • Pre-FRI audit sheet

In your ERP

Reads Odoo production orders (product, quantity, status) and stock transfers (partner, status, scheduled date, source document). Drafts a production order’s planned start or assignee, or a note on a transfer, run only if an admin approves. Set sheets, output, cutting logs and ship dates come from uploads or Google Sheets. The demo shows what your Odoo holds.

One ERP connection serves the whole factory: SAP Business One, SAP S/4HANA, Odoo, Oracle NetSuite, Microsoft Dynamics 365, WFX Cloud ERP or Datatex NOW. This agent reads only the records named above; we check yours on the call.

In your tools

Reads your shortage or component sheet in Google Sheets when you ask. Prepares a WhatsApp Business or Slack message to the stitching or packing supervisor, a new row for the shortage sheet, or the short-ship email to the plant head, sent from Gmail or Outlook. Each waits for an admin to approve or decline, and none runs unless approved.

Tools named today: Gmail, Outlook, Google Sheets, Slack and WhatsApp Business.

Never made up

  • Set compositions, SKU codes or component counts
  • Retail-ready packing dimensions, labels or barcodes
  • Embroidery file versions or stitch counts
  • Production quantities or packing status

If it is not in your connected data, the agent says so.

Belongs to another department

What the Home Textile & Made-Ups playbook covers

  • SOP: set-wise cutting

    Cut to a plan per set, and take the composition from the buyer’s routing guide, never from convention: a queen set may carry two pillowcases, a twin set one. A set bundle is the exact set, never partial. Every piece is numbered and mapped to its barcode, so each set pack can be traced.

  • Decision rule: one SKU 12 sets short

    In a 1,500-set order, trace by lot to see whether the cut was short, assembly miscounted or the FG warehouse miscounted. Assembly is the common cause. Make the 12 from the next batch only if the buyer permits it; otherwise ship partial and send the balance. The plant head and Merchandising hear of it, and the buyer before shipping where possible.

  • KPI: retail-ready audit pass-rate

    Target 99% or better before FRI. The playbook calls for a 100% audit of the first 50 sets, not the first 5, because retail-ready issues compound, then an AQL audit at packing. If an audit fails on barcode placement, it says to re-do the sets to spec, retrain and add a visual jig at the station.

  • Edge cases: mixed lots and drifting poly bags

    Two lots of one SKU packed together make the buyer’s ASN reconciliation fail, so pull and segregate them before shipping. Poly bags from a different supplier batch can drift from spec across cartons and make the buyer’s DC conveyor fail, so the playbook insists on one poly supplier per shipment.

Written as general textile practice and reviewed by four independent AI models. Not yet signed off by a human subject-matter expert. Upload your own SOPs and standards, and every answer shows which source it used.

Ten questions it ships with

The template’s starter questions. Your team can edit them, or ask in its own words.

  • Set assembly status for an SKU

    For this SKU, what does the set assembly sheet require, and which components are ready and which are short?

  • Retail-ready packing requirements

    What retail-ready packing does this buyer require for this SKU, and does the current packing plan comply? Cite the packing spec.

  • Set-mix accuracy for an order

    Check set-mix accuracy for this order: what was required per set, what has been produced, and which components are short.

  • Approved embroidery file and placement

    Which embroidery file and placement are approved for this design, and what version is currently in use?

  • Stitching output against plan

    Show stitching and hemming output against plan for the last shift, by SKU.

  • Packing readiness summary

    Prepare a packing readiness summary for this order from the set assembly and packing records: what is complete, what is short, and what is blocking.

  • SKUs at risk this week

    Which SKUs in the current order book are at risk of missing their date, and what component is holding each one?

  • Rejections by SKU

    How many pieces were rejected this week by SKU, and what reasons were recorded?

  • Label and barcode compliance

    Do the labels and barcodes planned for this SKU match the buyer’s specification? Show both and flag any difference.

  • Fabric consumption per set

    Show actual fabric consumption per set for this SKU against the planned consumption, and explain the variance.

Home Textile & Made-Ups: questions before a demo

Can the agent open our embroidery DST files?

No. A DST file is a machine file with no text in it, so the agent cannot open it, work out its hash or compare a stitched logo with the approval. Keep the DST file registry as a Google Sheet or an uploaded Excel file, with file name, version, file size, hash, placement and approval date. Ask the operator to copy the file name, version and size from the machine’s file list onto the run sheet, kept in a sheet rather than on paper. The agent sets those beside the registry and says when they differ or a value is missing.

Can it check the barcodes and labels on our packed sets?

Not by looking at them. Mavick does not read barcodes, look at images or inspect cartons. It works from text: it sets the barcode numbers and label positions in your label plan beside the buyer’s packing spec and lists the differences it finds. If your barcode scanners export a log as Excel or CSV with a result column, upload it and the agent counts the failed reads for a SKU from the rows. The packed sets still need your pre-FRI audit.

Our buyer’s routing guide is a long PDF. Can the agent read it?

Yes, if the PDF carries text and each file is 10 MB or less; split a larger guide into chapters before you upload it to the made-ups department. The agent searches the guide and quotes the clause it used for master pack quantity, carton marks or poly bag size, naming the file. A scanned guide has no text to read, so ask the buyer for the original. Where the wording is unclear, it quotes the clause and says that interpreting the buyer’s packing guide, and getting the buyer’s approval, belongs to Merchandising.

Can the agent change a production order in our ERP when sets come up short?

Only in narrow ways, and only with approval. In Odoo it can draft a new planned start for an existing production order, such as the batch that makes up short sets, or a change to who the order is assigned to. A department admin approves or declines it, and it runs only if approved. It cannot change quantities, set compositions, stock or a ship date in any ERP. A ship date that has to move goes to Merchandising as a drafted email, sent only if an admin approves it. On any other ERP it changes nothing there, so the make-up request goes to the stitching supervisor as a drafted WhatsApp Business or Slack message, sent only if approved.

Our output and set counts are on paper boards. Can the agent still find short sets?

Not from the boards. The agent does not count pieces on the floor or read a hand-written board, and a photo of a register has no text to read. Once the output or bundle count by SKU, component and lot is typed into an Excel or CSV file you upload, it sums those rows and lists the ones it used; from a Google Sheet, it reads the rows and cites them. Start with the set assembly count: it is where a short set first shows.

Can the made-ups agent look up a fabric lot in the dye house’s records?

No. The dye house’s batch cards and shade records belong to Dyeing & Processing, and only that department’s agent can read them. The separation is part of how Mavick keeps each department’s data. Ask the made-ups agent why a fitted sheet and its pillowcases differ in shade, and it names Dyeing & Processing as the department to ask. It can use your own fabric receipt log, with the lot number on each roll, kept in the made-ups department.

Bring Home Textile & Made-Ups. We’ll build it on the call.

Book a 30-minute demo (opens in a new tab)
Email
support@mavick.si