The Sneakers Passed Inspection. Three Months Into Sales, the Soles Started Lifting
Shoe Sole Bond and Peel Strength Test Guide - Why Soles Lift After Sale, Key Bonding Steps Such as Buffing, Primer, Activation and Pressing, Peel, Flex and Aging Tests Compared, Peel Strength Criteria by Zone, How to Read Failure Modes, Supplier Process Shortcuts and Adhesive Substitution, Spec Sheet Items, Stage-by-Stage Checks and a Checklist

Hello, this is StartMade.

A Korean children's sneaker brand produced 6,000 pairs of hook-and-loop sneakers at a footwear factory in Jinjiang, Fujian. The uppers were mesh and PU synthetic leather, and the sole was an EVA midsole with a rubber outsole. However hard you bent the approval sample by hand, the sole did not budge, and the pre-shipment inspection found no problems with appearance, sizing or bonding. The shoes also passed a check in which the inspector picked a few pairs and bent the toes back by hand.

The problem appeared three months into sales. Complaints that soles were lifting at the toe and along the sides kept coming in, and the return rate climbed to 12%. Children running around playgrounds opened up the toes first, and shoes worn on rainy days came apart even faster. The brand sent the remaining stock to a test lab. Peel strength at the toe averaged 1.6N/mm, less than half that of the approval sample (about 4.2N/mm). On the separated surfaces, the adhesive sat smooth on the rubber outsole side only. It had never properly bonded to the upper in the first place. When asked, the factory admitted that while adding lines to meet the delivery date it had skipped the primer (处理剂) on the uppers and sped up the conveyor through the activation oven.

Sole bonding defects are tricky because they are almost invisible at shipment. A freshly bonded sole holds up well when you bend it by hand, and an inspector's hand check cannot tell 3N/mm from 1.5N/mm. The difference only shows up as gaping toes after thousands of steps of flexing, soaking in sweat and rain, and the heat of a summer container. This article starts with why soles come off, then covers the key steps of the bonding process, peel, flex and aging tests and where they fall short, how to read the failure mode, the shortcuts suppliers often take, and how to put bond testing into the spec sheet and the inspection checklist.

A reply of "we'll make the bond strong" or a quick bend by hand is not enough. Set minimum peel strength by zone and the test method, criteria after flexing and aging, and bonding process conditions (primer, activation temperature, press pressure and time) in numbers in the spec sheet, and test production pairs you pick yourself. That is how you prevent soles lifting after sale.

1. Why Soles Come Off Shoes That Passed Inspection

A sole bond only works when surface preparation, adhesive, heat, pressure and time all line up. Miss any one of them and the shoe can look fine when freshly bonded yet lose its grip within a few months. Most of these steps tie directly to labor cost and line speed, so a factory under delivery and price pressure starts cutting the steps nobody can see. The main causes of sole separation are:

• Skipped surface preparation: If the buffing (打磨) that roughens the upper's bonding area is shallow, or cleaning and primer are skipped, the adhesive just sits on the surface.
• Release agent and contamination: Rubber and TPR soles carry mold release agent on the surface, and if it is not wiped off the adhesive will not stick.
• Insufficient activation: If the temperature or time in the oven that warms the adhesive to a tacky state is too short, the two surfaces never properly fuse.
• Insufficient pressing: If the press (压合机) pressure is low or the press time short, curved areas such as the toe and sides in particular bond poorly.
• Adhesive problems: Using adhesive past its pot life (可使用时间), mixing in less hardener than specified, or switching to a cheaper adhesive.
• Material compatibility: PU synthetic leather coatings, mesh fabric, EVA and rubber each need their own primer and adhesive, yet some factories treat them all the same way.

Sole bonding is a different issue from stitching strength. Seam and trim pull testing is covered in Episode 142, and importing shoes and bags in general in Episode 74. This article focuses on how firmly, and for how long, the upper and sole stay together.

2. Key Steps in the Bonding Process

Knowing the order in which bonding happens makes it much easier to talk to the factory before you set test criteria. For a typical cemented (冷粘工艺) sneaker, the steps are:

The steps most often dropped are primer and activation, because skipping either leaves no trace on the finished shoe. On a factory visit, just checking whether the primer brushes are actually moving on the line and what the oven thermometer reads tells you a lot.

3. Test Types — Peel Strength, Flex and Aging Tests Compared

Just asking for "a bond test" tends to get you a factory that bends a shoe by hand and says "it's fine," or sends an old test report. Decide what you want to verify first, then choose the right test.

Numbers from peel strength, durability from flex and aging

A single peel strength value cannot fully explain soles coming off after sale. Even if a freshly made shoe clears the peel criterion, a poorly bonded shoe weakens quickly through flexing and moisture. Confirm the basic bond as a number with peel strength, check durability during walking with the flex test, and check strength after time has passed with peel after aging. Setting all three together is the safer approach.

What peel strength criterion to set

Rough criteria widely used in the industry are below. Actual criteria should match your product and the requirements of the market where you sell.

4. Reading the Failure Mode — Look at the Separated Surface, Not Just the Number

On a peel test report, what the separated surface looks like matters as much as the number. The same 3.0N/mm means something completely different depending on where the bond gave way.

If, as in the opening case, adhesive remains smooth on the sole side only, surface preparation on the upper went wrong. Even if such shoes test near the criterion today, they give way quickly to moisture and flexing. When you receive test records, always ask for the failure mode and photos of the separated surfaces as well.

5. Common Supplier Shortcuts

Because bond quality cannot be checked by eye, it is easy to cheat. These are the types we have actually encountered in consultations.

The most common case is a process that differs between samples and production. The sample room puts care into a single pair, but a production line has to turn out hundreds of pairs an hour. The only remedy is for your inspector to pick production pairs and test them. For change control at second-tier suppliers such as sole factories, see Episode 115.

6. Defects Caused by Weak Sole Bonding

Bond defects rarely show at shipment and appear weeks to months after sale.

• Sole lifting and gaping (开胶): A gap opens at the toe and sides and keeps widening. This is the most common reason for returns.
• Safety incidents: A gaping toe can catch and trip the wearer. This is especially sensitive for children's shoes.
• Water ingress: Water gets in through the gap, soaking the inside of the shoe and causing odor and mold.
• Returns and lower ratings: Online, once reviews saying "the sole came off after three months" pile up, sales of that model drop immediately.
• Whole-stock defects: Stock in the warehouse was made with the same process, so it is hard to pick out only the bad pairs.

Moisture and heat often make things worse. For container condensation and moisture damage, see Episode 118, and for managing returns after sale, Episode 55.

7. Items to Include in the Spec Sheet

A single line saying "strong bonding" leaves every factory with its own standard. Put the items below into a table and attach it to the spec sheet. For how to write a spec sheet, see Episode 59.

Putting process conditions in the spec may look excessive, but this is where you get the biggest effect. Testing only confirms results after the fact, while process conditions stop the cause of defects in advance. For engineering change control, see Episode 119.

8. How to Add Bond Testing to Your QC Checklist

Sample stage

When you receive approval samples, send a few pairs to a third-party lab for peel and flex testing, separately from the factory's report. Check the failure mode at the same time. Material failure is a good starting point; adhesive failure means the process must be fixed before production.

Production stage

Early in production, watch the line in person and check that primer application, oven temperature, and press pressure and time follow the process condition sheet. Get peel test records for every lot, but if the values are always the same number just above the criterion and there are no photos, start by asking whether the tests were actually done. For production schedule tracking, see Episode 102.

Pre-shipment inspection stage

The inspector picks packed shoes, checks along the bond line at the toe, heel and sides for gaps, and bends them by hand. Picking one or two pairs per size and sending them to a lab for peel testing is more reliable. Peel testing requires cutting the shoe, so set the sample quantity in advance. For pre-shipment inspection, see Episode 100.

9. Sample Request Wording for the Factory

Putting the wording below into the purchase order or spec sheet takes the bonding process and test method out of the factory's discretion, and makes it easier to assign responsibility if a sole-lifting claim arises.

"本产品鞋头、鞋跟及鞋腰部位的帮底剥离强度均不低于3.5N/mm（按ISO 17708或GB/T 3903.3测试），不接受界面破坏。按ISO 17707耐折30,000次后，粘合部位不得开胶。生产须按确认的工艺条件执行：打磨范围、处理剂种类及干燥时间、活化温度及时间、压合压力及时间、静置时间，不得省略任何工序。胶水品牌型号及大底供应商以确认样为准，如需更换须事先书面通知并重新测试。每批每个尺码至少抽取1双、合计不少于5双做剥离测试，并提交注明型号、尺码、测试部位、破坏模式及剥离面照片的测试记录和设备校准证书。测试样品由我方验货员从大货中随机抽取。如有不合格，须加抽复测；复测仍不合格的批次暂停出货。"
(Upper-to-sole peel strength at the toe, heel and sides of this product must each be at least 3.5N/mm (tested per ISO 17708 or GB/T 3903.3), and adhesive failure is not accepted. After 30,000 flexes per ISO 17707, the bond must not open. Production must follow the approved process conditions (buffing area, primer type and drying time, activation temperature and time, press pressure and time, curing time), and no step may be skipped. The adhesive brand and product and the sole supplier follow the approval sample; any change requires prior written notice and retesting. For every lot, peel test at least one pair per size, at least five pairs in total, and submit test records showing model, size, test zone, failure mode and photos of the separated surfaces, together with the equipment calibration certificate. Test samples are drawn at random from production by our inspector. If any result fails, retest with a larger sample; any lot that fails the retest will be held from shipment.)

10. Three Real Cases

These are anonymized versions of cases we often see in consultations. Amounts and figures are examples.

Case 1 — Children's hook-and-loop sneakers

This is the case from the opening. The brand sent test results on returned shoes and remaining stock to the factory and held it responsible. With no peel strength criterion or process conditions in the contract, the factory insisted it had "made them exactly like the sample." The brand eventually pinned down the lots made during the line expansion and recovered part of the cost, but returns handling, scrapped stock and halted sales still left a loss of about 41 million won. From the next order, it put zone-by-zone peel strength and a process condition sheet into the spec and decided to inspect the line in person early in production.

Case 2 — Women's sandals with TPR soles

A company importing summer sandals from a footwear factory in Dongguan, Guangdong, received returns a month into sales for soles that had come off entirely. The detached soles were glossy, with almost no adhesive on them. It turned out the sole factory had changed, and the new one used a different release agent, while the footwear factory kept giving the soles only a quick clean as before. After adding sole cleaning and a dedicated primer step and requiring retesting on any sole supplier change, the problem did not recur.

Case 3 — Men's casual shoes

A company buying casual shoes from a shoe factory in Wenzhou, Zhejiang, felt reassured when peel strength right after shipment came in at a comfortable 4.0N/mm. But stock that had gone through summer ocean freight and more than six months of warehouse storage started showing shoes opening at the sides. Aging tests on the same lot showed peel strength dropping to 1.9N/mm. The cause was the factory switching to a less heat-resistant adhesive to cut cost. The company then added a peel-after-aging criterion and the adhesive maker and product name to the spec. A comfortable peel value on a freshly made shoe is no guarantee it will last six months.

11. Common Mistakes

• Not setting bond strength in numbers, as in "make the bond strong"
• Judging bond quality from the inspector's hand peel check alone
• Looking only at the peel strength number and not the failure mode
• Looking only at values for freshly made shoes and ignoring strength after flexing and aging
• Testing the approval sample once and never checking production
• Not knowing the adhesive or sole factory had changed

12. Sole Bond Control Checklist

Specification stage

• Set minimum peel strength by zone and the test method
• Stated the failure mode criterion (adhesive failure not accepted)
• Set criteria after flexing and aging
• Attached a process condition sheet covering primer, activation and pressing
• Set per-lot sample quantity and what happens on failure

Sample and production stage

• Had the approval sample tested separately by a third-party lab
• Checked on the line early in production that process conditions are followed
• Received per-lot peel test records with failure mode and photos
• Confirmed the adhesive maker, sole supplier and notice requirements for changes

Pre-shipment inspection stage

• Checked the test equipment calibration certificate
• The inspector picked production pairs and checked bond lines and gaps
• Ran peel tests on samples by size

Wrapping Up - Check With Numbers and Separated Surfaces, Not a Bend by Hand

At shipment, every sole looks firmly attached. Differences in bond quality show up as gaping toes after thousands of steps, sweat, rain and a summer container. Writing zone-by-zone peel strength and failure mode criteria, criteria after flexing and aging, and process conditions into the spec, and picking production pairs to check yourself, does not cost much. Compare a few hundred won per pair in process cost with a 12% return rate, and the answer is clear.

There is also something you can do right now. Ask the factory currently producing for you to send its latest peel strength test records, including failure mode and photos of the separated surfaces. If there are no records or only a single number, put together a sole bond control sheet and send it before your next order.

StartMade supports you on the ground with spec sheets covering zone-by-zone peel strength and process conditions, third-party testing of approval samples, process checks on the production line, and bond checks and peel testing before shipment. If you have already received sole-lifting claims, we can start with testing the remaining stock and finding the cause; if you are preparing a new shoe, we can start by setting the bond criteria with you.