Digital Proofing vs Production Press: 3 Real Differences and How to Handle Color Shifts After Approval

In December 2025, a skincare brand client came to our plant with an approved color sample to place a production order. After reviewing the first article, the client said, “The color does not match the sample.”

The “sample” in the client’s hand had been produced two months earlier by a digital proofing machine at another printing company, where it had also been approved. We use an industrial digital printing press.

We spent three days adjusting the settings and finally brought ΔE to approximately 1.8, which falls within the acceptable range under industry standards. However, the client would not accept it because their acceptance criterion was “exactly the same as the approved sample.”

This was not anyone’s fault. The color formation principles of proofing machines and production presses are fundamentally different, creating an inherent color gap that many clients do not understand. This article explains where that gap comes from, how large its impact can be, and how to preserve evidence when approving a sample.

First Difference: Color Formation Principles—Are Dots “Generated” or “Grown”?

This is the most fundamental of the three differences and also the most commonly misunderstood.

The key is the physical phenomenon of dot gain.

Industrial digital printing presses use electrophotographic imaging, as seen with Xeikon and HP Indigo. After ink is transferred to paper by electrostatic attraction, the ink continues to penetrate and spread through the paper fibers, making the actual dot area larger than the area defined during imaging. This increase varies with ink type, paper absorbency, and workshop temperature and humidity.

Proofing machines, especially inkjet proofing systems, deposit ink dots directly onto the paper. Their dot boundaries are sharper, with almost no spreading.

The conclusion is therefore: The color produced by a proofing machine represents how the design file should look under ideal conditions, while the color produced by a production press shows what will actually happen on that material. The latter is the color customers will see on the shelf and in their hands.

There is also a phenomenon that many people encounter: For the same color area, the color difference between a 10% dot and a 90% dot is more noticeable than the difference at solid 100%. This is because dot spreading has the greatest impact on midtones. It also explains why gradients and skin tones most often fail to match between proofing and production.

Second Difference: Color Gamut and Total Ink Coverage—Not Every Color Can Be Reproduced

The second difference concerns capability limits.

Industrial digital printing presses have a defined total ink coverage limit (Total Area Coverage / TAC), typically ranging from 200% to 300%. Beyond this level, ink buildup becomes muddy, drying is impaired, and blocking occurs. ICC profiles flatten this type of out-of-gamut color during rendering: when a color is impossible for the press, the profile substitutes a nearby printable color.

This creates a typical failure scenario:

A client had a brand color in a highly saturated deep blue, with four-color overprinting reaching a TAC of 280%. A digital proofing machine, particularly an inkjet system without TAC restrictions, could reproduce it, and the client approved the proof. On the production press, the profile replaced that blue with a slightly darker and slightly grayer printable color.

The client looked at the finished product and said, “It looks too gray.” The press was not cutting corners; the profile was making a trade-off.

There is another issue caused by the capability illusion of proofing machines: some inkjet proofing systems can produce colors outside the offset printing gamut because they do not need to account for dot gain or overprint drying. However, if the final product will be produced by offset printing, that color can never be reproduced by offset.

Our plant handles these situations directly: Proofs must be rendered using the production press ICC profile and printed on a production press, or at least on an industrial device operating on the same principle. If the client insists on approving a proof produced by an inkjet proofing machine, we require the contract to specify that color differences will be assessed using ΔE rather than “matching the approved sample.”

Third Difference: Durability—Impossible to See During Approval but Obvious on the Shelf

The third difference is the least visible because it cannot be detected at the moment of sample approval.

The ink layers produced by digital proofing machines and industrial presses differ in practical terms regarding rub resistance, water resistance, lightfastness, and drying. Typical differences include the following:

There is an easy trap here: Many clients test rub resistance on a proof produced by a proofing machine, receive an “failed” result, and then use that result to challenge the production run. In reality, proofing-machine samples cannot produce accurate results because they have not undergone coating, lamination, or drying.

There is another practical issue: Proofing machines and production presses do not reproduce paper white in the same way. The white produced by an industrial press combines the natural whiteness of the paper with specific paper-white corrections, while a proofing machine may produce a “cooler” white. Therefore, approving large solid areas from a proofing machine is particularly unreliable, because the eye exaggerates color deviations across large areas, especially with light colors.

Evidence to Retain During Sample Approval

Taken together, these three differences lead to a practical recommendation: Approving a sample is not approving a sheet of paper; it is approving a definition of the machine, ICC profile, and material combination.

Our plant’s current sample approval form always includes the following:

1. Proofing device—Specify the exact device, ICC profile, and profile version number;
2. Proofing material—Specify the paper or film grade, grammage, and whether lamination or coating is applied, because changing the material changes the relationship between the variables;
3. ΔE acceptance standard—Specify the acceptance criterion, such as ΔE ≤ 2.0 being considered acceptable, along with the measuring device and conditions, including the light source angle and whether the measurement includes a white background;
4. Finishing processes—Specify whether lamination is used and which varnish is applied, because these processes directly affect the final appearance;
5. Color sample records—Each party retains one copy of the original approved sample, and both parties sign and date it.

These five items may appear to be a formality, but consider the client from the beginning of this article: When they approved the sample with the previous printing company, nothing was documented. The sample was passed from hand to hand without a machine model, ΔE standard, or material record. As a result, when a color difference appeared three months later, they could only argue that it “looked different,” while we could only defend ourselves by stating that “our ΔE of 1.8 is within the industry standard.” Neither party had evidence.

We later completed a similar document with the client and approved a new sample. After the second approval, the client said, “If I had known this from the beginning, we would not have gone through all that trouble the first time.”

One-Sentence Summary

If only one sentence is needed: “A proofing machine is a design tool, while a production press is a manufacturing device.” A proofing machine provides a fast, clean, editable result. A production press provides a result capable of running continuously over the long term and remaining stable across different materials and operating environments. The gap between the two is precisely what must be approved, documented, and controlled.

As an additional note, our plant follows one internal rule: Whenever a client provides a sample produced by another company’s proofing machine and asks us to match the production run to it, we do not refuse. Instead, we record that machine model on the job ticket and state, “Reproduced using the proofing machine ICC, not the production press ICC.” Whether the color is accurate is then a technical issue, while responsibility is clearly defined. Separating these two matters makes disputes less likely to escalate.

Further Reading

The following articles belong to the same series:

• 3 Practical Applications of ICC Profiles in Digital Printing—The real reasons behind a ΔE>2 color difference between two orders on the same machine
• 3 Real Boundaries of ΔE Color Difference Control in Digital Printing—Three boundaries at ΔE≤1.5
• 3 Common “Color Calibration” Errors in Digital Printing—Mistakes encountered while technicians adjusted machines
• 3 Real Boundaries of “Screening” in Digital Printing—Screen angles, dot gain, and RIP curves
• What ΔE Color Difference Is Acceptable for Printed Products?—Acceptance thresholds at 1.5 / 2.0 / 3.0