What ΔE Difference in Printed Color Is Acceptable? The Hard Factory Constraints Behind Three Standards for Judgment

In November 2025, a printing plant in Zhejiang that produces export cosmetic gift boxes received a return shipment from a client. The client was a major beauty brand, and its acceptance criteria specified “color difference ΔE≤2.0.” The plant’s pre-shipment inspection found ΔE=1.8, which met the requirement. However, when the client brought a colorimeter to the site for testing, the result was ΔE=2.3, which failed the requirement. For the same batch of boxes, the two ΔE readings differed by 0.5. The client refused to accept the shipment, resulting in a loss of 180,000 for the printing plant.

The boss came to me and asked: "Why does the colorimeter at my factory give different readings from the colorimeter in his hand?"

I said: “It’s not that the instruments are different; it’s that you two are using different criteria for judgment.”

1. The question “How large a ΔE is acceptable?” has no one-line answer in the Chinese-speaking community

When an outsider asks, “What ΔE difference in color between printed pieces qualifies as acceptable?”, most printing plant owners are unable to answer right away. This is not because they lack expertise, but because “acceptable” applies to 3 completely different contexts.

Context 1: National Standards System. China’s national standards GB/T 7705 “Decorative Printed Products” and GB/T 9851 “Quality Requirements and Inspection Methods for Printed Products” specify only the concepts of “color difference within the same batch” and “color difference between samples of the same color”; they do not establish a unified ΔE threshold for conformity. The specific values depend on the terms of the contract or the printshop’s internal standards.

Context 2:ISO international standards. ISO 12647-2, Process Control for Four-Color Offset Printing, provides clear reference ranges for ΔE: color variation within the same batch ΔE≤2.0 (against the reference proof), and color variation between samples of the same color ΔE≤3.0. This is the standard most frequently cited by major European and American brands and in export orders.

Scenario 3: Brand-Defined Specifications. Major brands in the cosmetics, electronics, and premium food industries often include requirements such as “color difference ΔE≤1.5” or even ΔE≤1.0 in their internal acceptance manuals. These are not national standards, but the brand’s process red lines. If the printing plant does not see this requirement before accepting the order, it is highly likely to have its products rejected during mass production.

Each of the 3 contexts corresponds to a different set of "passing thresholds"—ΔE=2.0 is the baseline, ΔE=1.5 is an unwritten industry rule, and ΔE=1.0 is the threshold for premium brands. If the printer does not clarify in advance which standard the customer is using, acceptance will inevitably lead to disputes.

2. ΔE fluctuations of 1.0 are normal even within the same printing plant but across different batches

Let me share some real data. In September 2025, I helped a printing plant that produces tea gift boxes conduct its monthly quality review. The customer was Dayi Tea, and the contract specified ΔE≤2.0. Using the same file, the same ink formulation, and the same press operator, the printing plant produced 5 consecutive batches and randomly sampled 10 items from each batch for color-difference testing:

Batch 1: Average ΔE=1.4, maximum 1.9.

Batch 2: Average ΔE=1.7, maximum 2.2 (1 sample failed).

Batch 3: average ΔE=1.2, maximum 1.5.

Batch 4: average ΔE=1.9, maximum 2.4 (2 samples failed).

Batch 5: average ΔE=1.5, maximum 2.0.

5 batches: 3 batches passed and 2 batches failed. Customer complaint rate: 40%.

The pilot captain asked me: "Same machine, same batch of ink, why are the fluctuations so large?"

After breaking down the color-difference data for 5 batches, I found that the issue was not “large color differences” but “uneven color differences within the same batch”—with an intra-batch range of 0.5 and an inter-batch range of 0.7. The real problem was not that “the machine cannot perform well,” but that the machine’s condition is unstable every time it is started: ink agitation time varies, fountain solution concentration drifts, blanket tension changes, and workshop temperature and humidity are not controlled.

This case illustrates one thing: ΔE is not a measure of "printer capability," but rather an indicator of "process stability." Even with the same printer and the same equipment, it is entirely possible for boxes produced in the morning to have ΔE=1.2 and boxes produced in the afternoon to have ΔE=2.4.

3. The print shop’s self-check ΔE differs from the customer’s manual measurement ΔE by 0.5. Where is the problem?

Returning to the case of the Zhejiang printing plant at the beginning. Two colorimeters measured the same batch of boxes with a difference of 0.5. This was not instrument error, but a difference in measurement conditions.

The first difference: light source. The printing plant uses a D65 standard light source for internal inspection (color temperature 6500K, simulating daylight), while the customer uses a D50 light source (color temperature 5000K, simulating store-window lighting). For the same color, ΔE=1.8 under D65 and ΔE=2.3 under D50. Change the light source, and the values change.

The second difference is the background. When the printing plant measures color differences, the samples are placed flat on a white backing; when the customer conducts acceptance inspection, the boxes are in their finished form, and curved surfaces, folded corners, and shadows can all affect the reflectance readings of the colorimeter. Whether the boxes are flat-printed or supplied as finished cartons, a ΔE difference of 0.3-0.5 is normal.

The third difference: different measurement aperture sizes. Printers typically use an 8mm measurement aperture for their own inspections (to measure large solid areas), while customers use a 4mm measurement aperture for acceptance inspections (to measure fine details). The measurement area differs by a factor of 4, and the ΔE of small color areas is always higher than that of large color areas.

The solution is not to “buy a better color difference meter,” but to clearly specify the measurement conditions in the contract. The standard wording is: “Color difference shall be measured in accordance with ISO 12647-2, using a D50 illuminant, a 4mm measuring aperture, and a white backing, with the sample taken from the flat printed area of the finished carton.” Once this sentence is included in the contract, 99% of color difference disputes can be avoided.

4. ΔE≤1.5: Can Small and Medium-Sized Printing Plants Achieve This Standard?

In the supplier acceptance standards of major cosmetics brands (Estée Lauder, Lancôme, and Chanel), a color difference of ΔE≤1.5 is a common red line. For small and medium-sized printing plants, this figure is essentially a threshold for determining whether they can accept orders.

I spoke with the owner of a printing plant in Suzhou that specializes in cosmetic boxes. The plant has 2 Heidelberg CD 102 machines that have been in operation for 15 years, and their production processes are very stable. The contract provided by the client specifies ΔE≤1.5. The owner told me: "We have been producing cosmetic gift boxes for 8 years, and roughly 70% of the batches genuinely achieve ΔE≤1.5."

A 70% pass rate means that 30% of batches will be rejected. The customer allows a 5% concession acceptance rate (that is, 5 out of 100 batches may exceed the acceptance threshold). In other words, the printing plant must survive within the constraints of 70% “fully qualified” + 25% “slightly exceeding the threshold but acceptable by concession” + 5% “rejected.”

To maintain this 95%, the printing plant needs to do 3 things rather than “buy more expensive equipment”:

First, ink batch management. Different batches of the same ink may show slight variations in hue. The printing plant should establish ink batch records. Each time new ink arrives, print a test sample first to check the color difference, and use it on the machine only if ΔE≤0.5. The ink storage room should maintain constant temperature and humidity and be protected from direct sunlight.

Second, the press operator’s self-check rhythm. After startup, sample 1 sheet every 500 sheets to measure color variation, with the data recorded in real time. If the color variation begins to drift, stop the machine immediately to check the ink-water balance rather than waiting until a full batch is completed. A qualified workshop supervisor will review the color variation data 3 times per hour instead of waiting until a batch is finished and then reworking it.

Third,customer sample management. The printer must keep the “standard sample” provided by the customer sealed and stored as received; it must not be left in the workshop exposed to wind and sunlight. Each quarter, compare the new sample with the old one to confirm that the sample itself has not faded. This requirement is often overlooked—many ΔE disputes are actually caused by “the customer’s own sample having faded,” meaning the printer is being blamed unfairly.

5. Acceptable Color Difference Is Not the Finish Line, but the Brand Owner's Entry Threshold

Here is a potentially disputable point: ΔE compliance does not mean the printed product is qualified; it is merely the minimum threshold for the brand owner’s acceptance inspection.

For truly premium brands, the acceptance checklist includes not only color difference, but also 4 hard criteria:

First, registration accuracy. The positioning accuracy of text, lines, and pattern edges is generally required to be within ±0.1mm to ±0.2mm. When a 4-color offset press is used, even if one color is misregistered by 0.15mm, a visible “color fringe” will appear along the edges of the text, and no amount of further color correction can remedy it.

Second, varnish uniformity. A variance of ±2μm in varnish coating thickness will result in visibly uneven gloss. During customer inspection, the box is tilted under strong lighting to examine the reflections. If the varnish appears “blotchy,” it will be rejected outright.

Third, peel strength. The adhesion of film lamination or UV coating is usually tested by applying 3M tape and then peeling it off to see whether the coating detaches. The peel strength requirements for cosmetic boxes and electronic product boxes are ≥98% (that is, no more than 2 detachments out of 100 peel tests).

Fourth, abrasion resistance. Packaging boxes are repeatedly rubbed against other surfaces during transportation, storage, and shelving, so the ink and varnish must not discolor. The standard practice is to perform 50-100 dry rubs using a rub tester, with the color difference not exceeding ΔE=1.0.

These 4 items, plus color deviation, constitute the brand’s true “passing threshold.” If the printing factory cuts corners on one or two of them, the customer will still return the goods, even if the color deviation ΔE=1.0 is perfectly within standard.

6. 3 Standard Procedures for Printers to Address Color Discrepancy Disputes

Finally, three practical actions to take when the printing plant encounters a dispute over color differences.

Action 1:Clearly specify the measurement conditions in the contract. The light source, aperture, background, and sample form (flat print or finished box) must be documented in the contract or technical appendix. During customer acceptance, measurements are taken according to the contract conditions, and the printer acknowledges the results; if the contract conditions are not followed, the printer has the right to refuse to sign off.

Action 2:Retain color-difference data for each batch. The printer’s self-inspection color-difference data must be archived for 3 years. If a customer raises a complaint, the printer can present the self-inspection data to prove that ΔE was within the acceptable range when the goods left the factory, and that any color-difference changes caused by transportation or storage are not the printer’s responsibility. This provision provides legal self-protection.

Action 3:Proactively review customer samples. Each quarter, retest the customer-provided “standard sample” to confirm that the sample itself has not faded. Sample storage procedures should also be documented in the SOP: store away from light, at a constant temperature (20-25°C) and constant humidity (40-60% RH), in a sealed container. Implementing this measure allows 90% of “the sample faded on its own” disputes to be identified in advance.

If you are preparing a new color-difference acceptance contract or are stuck in a color-difference dispute over a shipment, send us the box dimensions, printing process, ink brand, and current measured ΔE value. Within 30 minutes, we can help determine whether the issue lies in the printing process, the contract, or the measurement conditions.

Further Reading

What ΔE Color Difference Is Acceptable for Printed Products?

3 Practical Boundaries of ΔE Control in Digital Printing

The Real Differences Between Pantone, CMYK, and Digital RGB Color Gamuts

3 Color Difference Control Methods for Packaging Printing