On this page
You send the same tech pack and the same spec sheet to two factories and get two garments that fit differently. This is one of the most common — and most underdiagnosed — sources of size-related returns for growing apparel brands. Most merchants track late shipments and fabric defects closely, but almost none systematically measure fit consistency across factories, even though it directly drives return rate and customer trust.
Why fit consistency across factories breaks down
A spec sheet lists target measurements with tolerances, but tolerances are interpreted differently depending on a factory's equipment, pattern-grading software, and quality culture. Two factories cutting the same "size medium" chest measurement can land half an inch apart and still be technically within their internal QC bounds. Multiply that across a multi-factory supply chain and you get a product line where "medium" means something different depending on which production run a customer receives.
This is the essence of supplier fit variance: it is not that any single factory is doing something wrong, it's that "correct" is being measured against slightly different internal baselines. Understanding what tolerance actually means in garment manufacturing is the first step to diagnosing where the variance is coming from.
The measurement gap
Most brands only find out about fit variance when return reasons spike for a specific SKU or batch. By then, the inconsistent batch may already be sold through, and the damage to customer trust — plus the return costs — is already booked. Waiting for return-reason tickets is a lagging indicator; you need a leading one.
Building a factory sizing consistency benchmark
The fix is to treat factory sizing consistency as a KPI, not an assumption. That means comparing actual delivered garment measurements — and actual customer-reported fit outcomes — factory by factory, on the exact same style and spec.
Step 1: Standardize measurement points
Before you can compare factories, you need identical measurement points recorded the same way for every batch: chest, waist, hip, inseam, shoulder width, sleeve length. If factory A measures chest flat and factory B measures chest stretched, your comparison is meaningless from the start.
Step 2: Pull return and exchange data by factory
If your production management system tags each SKU batch with its originating factory, you can cross-reference that against your returns platform's reason codes. A [return reason] tagged "too small" or "too large" that clusters heavily around one factory's batches — while the same style from another factory shows low return rates — is a strong signal of supplier fit variance, independent of whether the garment technically passed inline QC.
Step 3: Normalize for style and demand effects
Not every fit-related return is a factory problem. Some styles simply run generously and attract more "too small" complaints regardless of who cut them. To isolate the factory variable, compare the same style across factories, or compare each factory's deviation from its own historical baseline over time. This is also where checking for vanity sizing drift matters — a factory's numbers can look consistent internally while still drifting away from your intended size chart over several seasons.
A sample factory comparison scorecard
Here is a simplified way to structure a quarterly comparison across two hypothetical factories producing the same women's woven top style:
| Metric | Factory A | Factory B |
|---|---|---|
| Chest measurement deviation (avg, vs spec) | +0.3 cm | +1.1 cm |
| "Too small" return rate | 4.1% | 7.8% |
| "Too large" return rate | 3.2% | 2.9% |
| Exchange-for-different-size rate | 5.5% | 9.6% |
| Units produced this quarter | 12,400 | 9,800 |
In this example, Factory B is producing a garment that runs measurably larger against spec, and its combined size-driven exchange rate is nearly double Factory A's. That is a concrete, dollar-quantifiable case to bring back to the supplier — not a vague complaint about "quality," but a specific, measurable deviation tied to specific cost.
Quantifying the cost of inconsistency
Once you have a scorecard like this, translate it into cost per unit: return shipping, restocking labor, discount-to-resell on returned inventory, and the lifetime-value hit from a customer who had a bad first experience. Retailers already know that inconsistent sizing is one of the top drivers of apparel returns and lost repeat purchases — research on retail returns consistently shows fit and sizing issues as the leading reason shoppers send items back, according to the National Retail Federation. When you can show a supplier that their variance is costing you a specific percentage of margin on a specific style, negotiations change tone quickly.
Holding factories accountable with data, not opinion
Bringing generalized fit complaints to a factory rarely produces change, because "customers say it runs small" is not actionable for a production team. Bringing a scorecard with measurement deviations, return-rate deltas, and cost impact is actionable — it gives the factory something to trace back to their own cutting and grading process.
Set fit-consistency SLAs
Consider writing fit-consistency thresholds into supplier agreements: maximum allowable deviation from spec per measurement point, maximum size-related return rate per style, and a review cadence (quarterly is reasonable for most apparel categories). Tie repeat violations to volume allocation — factories that consistently hit their fit targets should get more of your production, not just the ones with the lowest unit cost.
Catch drift before it reaches customers
The scorecard approach works retroactively, using return data that already happened. To reduce fit-driven returns before they occur, pair supplier scorecards with pre-production sample review and in-line measurement audits. Combining both approaches is central to a broader strategy to reduce fit errors before they ever reach production, rather than discovering them through customer complaints months later.
Turning fit consistency into a competitive advantage
Brands that actively manage supplier fit variance build a reputation for size reliability, which shoppers increasingly cite as a purchase-decision factor. Sizing tools that use accurate, verified body and garment measurements — rather than static size charts alone — become far more effective once the underlying garments themselves are consistent, since the recommendation engine is no longer trying to compensate for supply-chain noise it cannot see. In effect, factory consistency and smart size recommendation are two halves of the same fit-accuracy strategy: one controls variance at the source, the other manages the variance that remains.
Frequently Asked Questions
How often should I audit fit consistency across factories?
A quarterly review is a reasonable cadence for most apparel brands, since it gives enough return-data volume per style to be statistically meaningful without letting problems compound for too long. High-volume core styles that run through multiple factories can justify a monthly check, especially in the first two seasons after adding a new supplier.
What is a normal amount of fit variance between factories?
There is no universal number, since tolerance depends on garment category, fabric, and construction method, but most quality teams treat deviations under half a centimeter on core measurement points as acceptable and anything above one centimeter as worth investigating. The more useful benchmark is each factory's deviation from its own historical baseline, since a sudden shift often signals a process or staffing change worth a direct conversation.
Can better size recommendation tools compensate for factory inconsistency instead of fixing it at the source?
They can reduce the customer-facing symptom, since a sizing engine tuned to actual garment measurements can route shoppers around a batch that runs large or small. However, this treats the symptom rather than the cause, and it cannot fully substitute for fixing the underlying supplier variance, since inconsistent inventory still creates operational costs in returns, exchanges, and restocking regardless of how well shoppers are guided at checkout.
