Knit Bag Partner Scorecard: Twelve Weighted Checks for Scoring a Knitted Bag Programme Partner
How should a buyer score a knit bag partner before placing an order? Put twelve checks into 100 points and weight them by what actually goes wrong: sampling accuracy, bulk delivery, AQL 2.5 evidence and change control. A scorecard beats a questionnaire because it forces a number against every claim, so two shortlisted partners can be compared on the same page. On a programme running 100 pcs per style and colourway, 7 working days sampling and 30 days bulk, the four heaviest checks are the ones that decide whether the goods land on the vessel or in a re-inspection.
The 100-Point Split: Where the Weight Actually Belongs
A scorecard only works if the weights reflect failure frequency, not intuition. Across bag programmes the disputes that reach a buyer's desk are overwhelmingly about three things: the sample did not match the bulk, the bulk arrived late, and the inspection found more defects than the buyer expected. Checks that predict those three outcomes deserve roughly two thirds of the available points.
Everything else is secondary and should be weighted accordingly. Certificates, photographs of a showroom and social audit ratings matter, but they rarely decide whether a specific order succeeds, and giving them equal weight dilutes the score into something that cannot discriminate between two partners. The split below is the one that survives contact with a real season.
- Sampling accuracy versus bulk
- 20 points
- Bulk delivery reliability
- 18 points
- AQL 2.5 evidence and test reports
- 16 points
- Change control and escalation
- 12 points
- Capacity honesty at peak
- 10 points
- Yarn and trim traceability
- 8 points
- Colour matching and lab dip turnaround
- 6 points
- Compliance documentation currency
- 4 points
- Communication response time
- 3 points
- Re-order behaviour after a failure
- 3 points
Ten weighted lines rather than twelve keeps the card readable; the two remaining checks, packing accuracy and post-delivery support, sit inside the bulk delivery and re-order lines respectively, because scoring them separately produces two lines that almost always move together. What matters more than the exact split is that a buyer commits to it before seeing the answers, so the weighting cannot be reverse-engineered to favour a preferred partner.
Checks One to Three: Sampling Speed, Sample Accuracy, Sample Cost
Sampling accuracy earns the heaviest single weight because it is the least expensive place to be wrong and the most expensive place to discover it. A partner that hits 7 working days but sends a sample knitted at a different gauge than the bulk will quote will produce a bulk that does not match the approved reference, and no amount of inspection recovers that. Score the sample against the specification, not against the photograph.
Concretely, ask for the knitting programme reference, the gauge, the yarn count and the finished weight recorded on the sample submission form, then verify those four things again on the first bulk piece. Where the numbers differ, the sample was a demonstration rather than a prototype, and the twenty points drop to single figures honestly rather than being argued away later.
- Confirm the sample was knitted on the same gauge the bulk will run on, for example E7 for a chunky shopper or E12 for a finer city bag.
- Record yarn count and finished grammage on the sample form; a 300D body yarn and a 600D base yarn behave very differently under load.
- Time the round trip from brief to door, not the knitting time; a 7 working day sample that sits two days in transit is a nine day sample.
- Ask how many sample rounds are included before the price moves, because a re-knit after a shape change is where schedules slip.
- Check whether the hardware on the sample is the hardware that will be bought, since a swapped buckle changes the load path entirely.
- Score colour matching separately: a lab dip returned inside the sampling window is worth more than a perfect dip returned late.
Checks Four to Six: Bulk Delivery, Capacity Honesty, Change Control
Bulk delivery reliability is the second heaviest line because a late bag is a missed season, and a missed season costs more than any unit price difference. The question to score is not whether the partner promises 30 days but whether the 30 days starts when the buyer thinks it starts. Many programmes quote from deposit, some from sample approval, and a few from yarn arrival, and the difference is frequently two weeks.
Capacity honesty is the check most buyers skip and later regret. A partner running 7 production lines at around 200,000 pcs per month can absorb a mid-size order comfortably, but the same partner in peak season may be booking new programmes behind orders already committed. Ask what percentage of the next eight weeks is already booked, and score the answer that gives a number rather than the one that says capacity is fine.
| Question to ask | A scoreable answer | An unscoreable answer |
|---|---|---|
| When does the 30 day clock start? | From sample approval, stated in writing on the order acknowledgement | Around 30 days from order |
| How much of the next eight weeks is booked? | A percentage, with the open weeks named | We have plenty of capacity |
| What happens if a colour is dropped? | A written change note with a revised date and a revised price | We will try to keep the date |
| Who signs off a specification change? | A named person, with a written change record | The production team will handle it |
| What is the escalation path at inspection? | Named contact, response window in hours, and a re-inspection rule | Talk to sales |
| How is a repeat order treated? | Same specification held on file, re-sampling only if yarn changes | We will start again |
Checks Seven to Nine: AQL 2.5 Evidence, Test Reports, Traceability
A partner quoting AQL 2.5 under ISO 2859-1 should be able to say which general inspection level and which sampling plan sit behind the number, because AQL 2.5 alone does not fix a sample size. The sample size comes from the lot size and the inspection level, and a partner that cannot produce both is quoting a phrase rather than running a plan. Score the ability to produce the plan, not the phrase.
Test reports are the second half of the same line. Abrasion to ISO 12947, tensile to ISO 13934-1 and colour fastness to the relevant ISO 105 parts are the three that matter for a knitted panel, and each should arrive with the laboratory name, the report date and the tested construction. A report on a swatch knitted at a different gauge is not evidence about the bag being bought.
Traceability closes the line. Yarn dyed to a specific lot, trims bought to a specific purchase order, and a finished goods record that ties the two together are what make a recall answerable in hours rather than weeks. Score the partner that can produce a lot number for the yarn in a delivered bag against the one that can only describe the yarn in general terms.
Checks Ten to Twelve: Communication, Escalation, Re-Order Behaviour
Communication is worth few points in absolute terms and many points in practice, because it multiplies everything else. A partner that answers a technical question within one working day lets a buyer resolve a specification ambiguity before it becomes a defect; one that takes a week turns the same ambiguity into a re-inspection. Score the measured response time on the last three technical questions, not an undertaking about future behaviour.
Re-order behaviour is the most predictive line on the card and the one nobody asks about. After a failure, a partner either produces a root cause and a changed process, or produces an apology and the same process. Asking what changed after the last rejected lot tells a buyer more than any certificate, and it costs one question to find out.
Sampling accuracy
Twenty points. Compare the sample's gauge, yarn count and grammage against the first bulk piece, not against the sample photograph.
Bulk delivery
Eighteen points. Pin the start of the 30 day window in writing, and score the booking level for the weeks the order actually occupies.
Inspection evidence
Sixteen points. AQL 2.5 must arrive with a general inspection level and a sampling plan, plus dated test reports on the right construction.
Change control
Twelve points. Specification changes need a written note with a revised date and a revised price, signed by a named person.
Capacity honesty
Ten points. Ask what share of the next eight weeks is already committed, and score the answer that gives a percentage.
After a failure
Three points, and the most revealing. Ask what process changed after the last rejected lot instead of asking whether it will happen again.
How the Numbers Move When the Programme Changes
The weights are not fixed in stone; they shift with what the buyer is buying. A promotional programme where the bag is given away and the deadline is immovable pushes delivery reliability up and abrasion performance down. A retail programme where the bag carries a brand for seasons does the opposite, because a faded or pilled panel returns as a complaint long after the delivery date is forgotten.
Quantity shifts the weights too. At 100 pcs per style and colourway the sampling line dominates, because the sample is a large fraction of the total effort and a wrong one is expensive relative to the order. At twelve thousand pieces the sampling line matters less and the capacity and inspection lines matter more, since the risk profile has moved from getting it right once to getting it right repeatedly.
A useful discipline is to re-weight the card at the start of every season rather than reusing last season's numbers unexamined. Two minutes spent moving points between the delivery line and the inspection line, based on what went wrong last season, is the simplest continuous improvement available to a sourcing team and it requires no new data collection at all.
Two Worked Scores: A Large Tote Run and a Small Capsule
Consider two partners scored for the same buyer. Partner A returns a technically excellent sample knitted at the correct gauge, gives 30 days from sample approval, quotes AQL 2.5 with a stated general inspection level, and answers technical questions inside a working day. Partner B returns a beautiful sample knitted at a finer gauge than it intends to run, promises around 30 days, and answers questions in three to four days.
On a twelve thousand piece tote run, Partner A scores high on the three heaviest lines and Partner B loses most of its twenty sampling points at the gauge check. The gap is decisive and the buyer places with A. On a four hundred piece capsule where the buyer wants a fast, low-risk first order, the same card still favours A, but the gap narrows because the sampling line is a smaller share of the risk.
The point of the worked example is that the card does not change the decision so much as it makes the reason for the decision written down. A buyer who has to justify a sourcing choice to a colleague, or revisit it two seasons later, benefits from a record that says which check moved the score rather than a memory of a general impression.
What a Scorecard Cannot Tell You
A scorecard measures declared behaviour and documented evidence, and both can be presented well by a partner whose actual floor performance is ordinary. It cannot verify that the yarn in the delivered bag is the yarn in the test report, that the subcontracted finishing stage was assessed, or that the photographs of the line were taken this year. Those need verification, and verification needs a visit or a third-party inspection with an explicit brief.
It also cannot price risk. Two partners may score within a few points of each other while one of them is quoting FOB Xiamen and the other CIF to a European port, and the scorecard says nothing about which is better value once freight, duty and the cost of a delayed season are included. Score the partners first, then price the shortlist, in that order, so the commercial comparison happens on a filtered field.
Turning the Score into Contract Language
The score earns its keep when the heaviest lines become contract clauses. The sampling line becomes a clause requiring the bulk to be knitted at the gauge and yarn count recorded on the approved sample form. The delivery line becomes a clause naming the event that starts the 30 day window. The inspection line becomes a clause naming the general inspection level behind AQL 2.5 and the test methods that must be reported.
The one clause worth writing first
- Require that the bulk is knitted at the gauge and yarn count recorded on the approved sample form, and that any change is issued as a written change note with a revised date and price.
Score a knit bag programme before you commit
Send the bag type, quantity per colourway and target delivery week. We will return a completed scorecard against sampling, bulk delivery, AQL 2.5 evidence and change control, and quote FOB Xiamen at 100 pcs per style and colourway with 7 working days sampling and 30 days bulk. Start on the contact page.
Frequently Asked Questions
What is a knit bag partner scorecard?
It is a weighted evaluation tool that turns around twelve sourcing checks into 100 points so two shortlisted partners can be compared on the same page. Unlike a questionnaire it forces a number against every claim, and the weights are set before the answers arrive so they cannot be reverse-engineered.
How many checks should a scorecard have?
Around twelve, grouped into sampling, delivery, inspection evidence, change control and communication. More than that and the card stops discriminating between partners; fewer and it misses the failure modes that actually cost a season. Ten weighted lines plus two folded into others works well in practice.
Which check should carry the most weight?
Sampling accuracy, at roughly twenty of the hundred points. A sample knitted at a different gauge or yarn count than the bulk will run produces goods that do not match the approved reference, and no amount of inspection recovers that position once the knitting has started.
How should a buyer score bulk delivery reliability?
By pinning the event that starts the window rather than by the number of days quoted. From deposit, from sample approval and from yarn arrival can differ by two weeks on the same programme, so the clause naming the start date is worth more than the figure itself.
What does AQL 2.5 actually require a partner to state?
The general inspection level and the sampling plan, because AQL 2.5 alone does not fix a sample size. Sample size comes from the lot size and the inspection level under ISO 2859-1, so a partner quoting only the phrase is quoting a slogan rather than running a plan.
Which laboratory tests belong on the evidence line?
Abrasion to ISO 12947 with the end point agreed in cycles, tensile strength to ISO 13934-1 or ASTM D5034 with the method named, tear resistance to ISO 13937 where the structure suits it, and colour fastness to the relevant ISO 105 parts. Each report should carry the laboratory name, the report date and the tested construction rather than a bare pass statement.
How can a buyer check capacity honesty?
Ask what share of the next eight weeks is already committed and request the open weeks by name. A partner that answers with a percentage can be scored; one that answers with a general reassurance about plenty of capacity cannot, and that difference is itself informative.
Should social audit status be scored on the card?
Yes, but with a low weight of around four of the hundred points. Audit status is a gate rather than a differentiator: it must be current and must cover every production stage including dyeing, finishing and any subcontracted work. It rarely decides whether a specific order ships on time, so it should not compete with the lines that do.
How should change control be scored?
On whether a specification change produces a written note with a revised date and a revised price, signed by a named person. Verbal assurances that the date will be kept score zero, because the record is what a buyer relies on when the date slips.
What is the most predictive question on a scorecard?
The most predictive question is what process changed after the last rejected lot. A partner that names a root cause and a changed procedure is demonstrating a management system capable of learning. A partner that offers an apology and the same procedure is demonstrating a recurrence waiting for the next season to arrive.
Do the weights change with order size?
They do. At 100 pcs per style and colourway the sampling line dominates because one wrong sample is expensive relative to the order. At twelve thousand pieces the capacity and inspection lines dominate, because the risk has moved from getting it right once to getting it right repeatedly.
Can a scorecard replace a factory visit?
No. A card measures declared behaviour and documented evidence, and both can be presented well by a partner whose floor performance is ordinary. Verifying yarn identity, subcontracted finishing stages and current working conditions still needs a visit or a third-party inspection working to an explicit written brief.
How often should the weights be reviewed?
At the start of every season. Moving points between the delivery line and the inspection line based on what went wrong last season takes about two minutes and is the most accessible continuous improvement available to a sourcing team, because it needs no new data collection and no additional approval.
What should happen after the scores are in?
Turn the heaviest lines into contract clauses and price only the shortlisted partners. Score first, then compare landed cost, so the commercial comparison happens on a filtered field rather than quietly influencing which partners survive the scoring stage in the first place. Written clauses also make the next dispute short.
People Also Ask
What is a bag partner scorecard?
A weighted evaluation sheet that converts around twelve sourcing checks into 100 points. It is designed so two shortlisted partners can be compared on the same page, with the weights fixed before any answers arrive.
How do you score a bag production partner?
Weight sampling accuracy highest, then bulk delivery, then inspection evidence under ISO 2859-1 and ISO 12947. Score answers that contain numbers, dates and named owners, and score general reassurances at zero, because an answer without a figure cannot be compared.
What should be on a bag sourcing checklist?
Gauge and yarn count on the approved sample, the event that starts the bulk window, the general inspection level behind AQL 2.5, dated laboratory reports, and the escalation path with a named contact and response window.
Why does sample accuracy matter more than sample speed?
Because a fast sample knitted at the wrong gauge produces a bulk that does not match the approved reference. Speed is recoverable inside the schedule; a mismatched reference is not recoverable without re-knitting.
How do you compare two shortlisted bag partners fairly?
Fix the weights before collecting answers, demand numbers rather than reassurances, and require the same evidence set from both. Then price only the survivors so cost does not quietly decide which partner was scored favourably.
Sources and further reading
Every figure on this page is either our own programme data or attributed below. Open the originals before you write them into a specification.
- ISO 2859-1:2026 Sampling procedures for inspection by attributesISO The AQL-indexed sampling schemes behind the AQL 2.5 figure, including the general inspection levels that fix the sample size for a given lot.
- ISO 12947-2:2016 — Textiles, determination of abrasion resistance by the Martindale methodISO The abrasion method named on the inspection evidence line of the scorecard and the end-point definition it uses.
- ISO 13934-1:2013 — Textiles, tensile properties of fabrics, Part 1: maximum force using the strip methodISO The tensile method used to verify that the tested construction matches the gauge and yarn count being bought.
- Incoterms rulesInternational Chamber of Commerce The framework behind FOB Xiamen and the alternative CIF and DDP terms referenced when pricing the shortlisted partners.
- amfori BSCI social audit platformamfori The social audit route scored at a low weight on the card, treated as a gate covering all production stages rather than as a differentiator.