Most shop owners importing from 1688 know they have defective goods. They do not know which supplier is causing it, how bad the rate is, or exactly how much it costs each month. When you measure defect rates per supplier, that vague awareness becomes actionable data: enough to negotiate compensation, push prices down, or switch sources with a clear reason instead of a gut feeling.
Why per supplier, not in aggregate
If you import from several suppliers on 1688 and only track total defects, you know you are losing money but not where the loss comes from. This month you rejected 30 units for quality failures. Was that from supplier A, supplier B, or spread evenly? That question is unanswerable with a single combined number.
Splitting the data by supplier solves three concrete problems:
- Locating the source. A supplier with consistently high defect rates across multiple batches has a systemic problem, not bad luck. A supplier with one poor batch followed by several clean ones had an isolated incident you can accept.
- Creating negotiating leverage. When you message a supplier saying "the last batch ran 4.2% defective, equivalent to X yuan of damaged goods," they cannot push back with a feeling. A specific number opens a conversation that a vague complaint cannot.
- Making decisions from evidence. Instead of switching suppliers because it "seems like they keep sending bad stuff," you switch because three months of data shows the rate has exceeded your tolerance threshold.
What kinds of defects to record
Not every return is a supplier defect. Before you start measuring, separate three categories so you do not miscalculate:
- Manufacturing defects: goods that fail quality on arrival at your warehouse. Broken, cracked, wrong spec, missing parts, not functional. These count against the supplier.
- Transit damage: goods that were fine when shipped but damaged in transit because of weak packaging or rough handling by the carrier. Responsibility here typically falls on the carrier or the packaging spec.
- Customer returns: returns where the product was not actually faulty but the customer did not like it, bought the wrong item, or your listing description was inaccurate. These do not count against the supplier.
When you receive and inspect each incoming batch, record which category each rejected unit falls into. Skipping this step means you will accidentally blame the supplier for problems that came from elsewhere, and miss the chance to hold them accountable for problems that genuinely are theirs.
Building a simple tracking sheet
You do not need specialized software. A spreadsheet with the following columns is enough to start:
- Supplier name or 1688 store ID
- SKU ordered from that supplier
- Date stock arrived
- Units received
- Units with manufacturing defects
- Defect rate for that batch (defective units divided by total received)
- Notes on defect type (broken, wrong spec, missing component, etc.)
One row per incoming batch. After three or four batches you have enough to calculate an average defect rate per supplier.
How to calculate the average correctly: add up total defective units across all batches from that supplier, then divide by total units received from them in the same period. Do not average the batch-level percentages, because a large batch and a small batch would carry equal weight and skew the result.
What defect rate is acceptable
There is no universal threshold that applies across all product types. Small electronics like phone accessories are high-risk if the rate hits 1 to 2 percent. Ceramic kitchen goods that take rough shipping may run 3 to 5 percent defective without it being the factory's fault if packing is minimal. You need to set your own thresholds based on two inputs:
- Margin per unit. Thin-margin products tolerate less. A batch of 500 units earning roughly VND 20,000 each in profit, running a 3% defect rate, means 15 broken units. That is VND 300,000 in lost product plus handling cost, enough to wipe a full week of margin on that SKU.
- Defect handling cost. Each defective unit costs extra time to inspect, plus returns handling if it already shipped to a customer, plus disposal or liquidation. Factor this in when assessing the real impact, not just the product cost.
- Platform rating exposure. Defective goods that reach customers pull in bad reviews, push your shop score down, and eventually affect listing visibility. A shop with a high return rate on imported goods gets suppressed by the platform algorithm. That damage is hard to measure in the moment but very real.
A practical approach: set a warning threshold and an action threshold separately. For example, if defect rate exceeds 3% across two consecutive batches, raise it with the supplier; if it exceeds 5% or climbs across three consecutive batches, evaluate switching sources. The exact numbers are yours to decide based on category and margin, but you need thresholds or you will never know when to act.
Using the data to negotiate
When a supplier's defect rate crosses your threshold, the next step is a conversation. Well-documented data changes that conversation entirely.
Instead of messaging "your goods keep having problems," you send: "Across the last three batches, 420 units received, 19 manufacturing defects, a 4.5% rate. Defect type is mostly X. Photos attached." That kind of specificity forces a serious response. A supplier cannot say "the goods are fine" when you are holding numbers.
Common negotiation directions:
- Compensation in the next batch. The supplier adds replacement units to the next order or discounts it. Most common when there is an existing relationship.
- Process improvement request. If defects cluster around one type, for example packaging failure, ask them to change that step and confirm with photos on the next batch.
- Lower unit price. If the defect rate persists, the handling cost you are absorbing is effectively a subsidy for their quality failure. That is a reasonable basis for negotiating a lower order price.
Long-term supplier relationships on 1688 are mostly built through consistent orders and reliable payment. But when quality slips, clear data protects you and applies pressure in the right place.
When to switch suppliers
Negotiation does not always work. Some suppliers improve after one conversation. Others keep delivering the same rate regardless of how many times you raise it.
Signs that switching makes sense:
- The rate does not drop after two or three separate complaints. They know about the problem and have not fixed it, which means they are not prioritizing quality for your account.
- The rate is trending upward over time. A first batch at 2%, second at 3%, third at 4.5% is a deteriorating pattern, not random variation.
- Defective goods are reaching customers. If bad units slip past your own inspection and land with buyers, the real cost now includes bad reviews, platform refunds, and shop reputation damage. That kind of loss is substantially heavier than warehouse loss.
- A better option exists on 1688. If another supplier sells the same item with stronger review scores and higher repeat-purchase signals, staying out of loyalty is costing you money.
When switching suppliers on an important SKU, do it carefully: place a small test batch with the new supplier, compare real quality, then shift volume. Do not burn the old relationship while the new one is unproven.
Inspection cadence
Tracking defect rates only works if you inspect goods consistently when they arrive. A simple process:
On arrival: count the full quantity first, then open a sample inspection. You do not need to check every unit in every batch, but you need to be consistent. Checking 10 to 15 percent of each batch and recording the result is a reasonable baseline. For small batches under 50 units, inspect everything. For larger batches from 200 units up, a random sample of 30 to 40 units is usually enough to detect systemic problems.
Photograph defective units on the day they are found. This is the most important evidence you have when following up with a supplier. Photos with a date stamp and a clear view of the defect type are hard to argue against. Without photos, suppliers can question whether the damage happened in your warehouse.
Record the data the same day stock arrives. Leaving it for later usually means it does not get done. Data recorded late also loses precision about timing and batch sequence.
Bottom line
Defect rate is a hidden cost that comes directly out of the margin on every SKU. Tracking it per supplier turns a vague loss into something you can act on: negotiate compensation, decide whether to stay or switch, and understand exactly what you are paying for poor quality. Three months of clean data is worth more than three years of instinct.