Book Printing: Color Shift Problems in the Book Proofs

Sometimes things go horribly wrong. I think there’s nothing worse than “hearing” the exasperation of a loyal client in her email, knowing that a multi-year working relationship is on the line.

I recently heard back from a client for whom I had been printing a small color swatch book for years. The swatch books pertained to make-up and clothing color choices appropriate for a woman based on her complexion.

The small color books, bound with a screw and post assembly, are essentially PMS books for the fashion industry. I have written many articles about this project, which my client reprints every few months. All 22 original print books had been produced without incident until just recently. They had been direct reprints of between one and ten copies of each of the 22 original master books. The books had been printed, laminated, round cornered, and drilled for the screw and post binding. The book printer produced the color books on an HP Indigo. The color was dead on. Reprints of the print books went like clockwork. Until they didn’t.

The Change in the Job Specs

The change in the job specifications that precipitated the problem was a small one. My client would print 100 copies of the 22 master books (various numbers of copies of each to equal 100 total color books) to fulfill orders for her clients. However, this time, in order to have more colors to bind into some of the print books, my client had created a single sheet containing an additional 30 colors. Some colors matched pages already included in the 22 master copies; many did not.

The goal was to print these at the tail end of the job, once the other books were complete, and then run them through the same finishing operations: lamination, round cornering, drilling, and such, but then to deliver them loosely packed in a carton (not bound on screw and post assemblies). This would cost an additional $200+ instead of the (almost) $3,000 price tag for producing the 15 copies of the single page as a stand-alone job. Why the difference? Because all of the makeready costs that would comprise the almost $3,000 would also cover all of the finishing work for the reprinted 100 copies of the older print books.

But there were problems with the color accuracy—for the first time in the history of the numerous direct reprints. Five of the colors in the 30 extra (master) color chips (the 30 loose chips of which 15 copies would be printed) were already included in the original 22 print books, and the proofs of these colors did not match the colors in the original books.

Fortunately, all the other colors (new ones and colors that had to match the original colors in the 22 master books) were ok.

What Caused the Problems?

Keep in mind that the clock was ticking. My client had clients who wanted books. Their shipments had to be back-ordered. My client also had a new financial backer who understandably also wanted accurate colors.

Fortunately, the sales rep at the printer had a complete set of printed and laminated copies of the original books plus a set of unlaminated proofs of the additional 30 loose color chips. So a list of the five problematic colors gave her a good starting point to resolve the color matching problem.

The sales rep had her plant manager check the HP Indigo color calibration. To be safe, he ran a second set of proofs on a higher-end HP Indigo digital press. He sent second proofs to my client’s financial backer (at my client’s request, assuming the problems had been resolved), but my client’s financial backer said the revised proofs were identical to the first set, with the same five problematic colors still off target (specifically too light). Ouch.

Where Do We Go Next?

Fortunately, since my client sees that the printer is taking this very seriously and trying to make things right, she has given us more time to correct matters. Here are some of the things we have learned and/or have considered relevant to resolving the problem:

1. 1. The first digital press had been upgraded from a prior model. Apparently, this particular HP Indigo had been altered to improve it, but the color calibration was not yet accurate. This affected primarily the blues, reds, and purples in my client’s color book. That is, the color problems were localized. They did not affect all hues.

1. 1. When the plant manager moved the job from the first HP Indigo to the larger, higher-end HP Indigo, the problem didn’t go away. Assuming the revised proof was correct (which to his eyes it was), the printer sent the second set of proofs to my client’s financial backer. My client herself didn’t see them. So I asked my client to have her financial backer cut each color swatch in half and send her a complete set of proofs (so both my client and her financial backer, who live in different cities, would each have her own complete set to facilitate communication). Why? Because two people will always see color differently (in this case three, or even more, since the printer also had a set of the same proofs).

1. 1. I also asked both the book printer and the client (and her financial backer) to look at the colors in different lighting conditions. Why? Because color will look different in sunlight, incandescent light (the traditional light bulbs with filaments), and fluorescent light. Presumably color will also look different under LED light.

1. 1. I asked everyone to cover each eye (one at a time, back and forth) and check the color. (For some people, including me, colors appear slightly different when they are seen by one eye and then the other.)

1. 1. I asked the book printer whether any of these colors might be especially problematic when reproduced with cyan, magenta, yellow, and black toners. This is because my client’s financial backer had selected PMS colors, and the HP Indigo simulates PMS colors using process color builds. Granted, extra colors can be used on the HP Indigo (up to seven), and these will expand the overall color gamut, making it possible to match more PMS colors. But not all of them.

1. 1. The book printer noted that the lamination would darken the colors (some more than others) and make them more intense. This might be perceived as a color shift by either my client or her financial backer. (Keep in mind that color has three properties: 1) hue, or the named color, like “blue”; 2) lightness/darkness, or value; and 3) intensity or purity. Laminating the color chips would affect two of these three variables.) That said, the reason this was a problem is that the proofs were not laminated, but the color pages in the original 22 master print books had been laminated. So if my client or her backer were matching the proofs of the 30 loose color chips (unlaminated proofs) to the master books (laminated pages), the colors would not look alike. In some cases the difference would be minimal, but in other cases—apparently—the color shift would be more dramatic.

1. 1. The printer also noted that if the five problematic colors were adjusted (by the prepress department) to make them accurate, this would affect all other colors on the 30-color digital press sheet (my client could wind up with five correct colors and 25 colors that were “off,” the opposite of the current situation).

1. My client told me that she still had the proofs (unlaminated) from the first printing of all 22 master books. I asked her to send these immediately to the printer. He would be able to more easily adjust the colors of the 30 new, loose color chips to match the colors in the original books because he would be matching unlaminated color pages to unlaminated color pages.

This is where we are now. We’ll see what happens.

What You Can Learn from This Fiasco

1. 1. I always say that when you buy commercial printing, you are not buying a commodity. You’re buying a process. At a time like this, it helps to have a long-term working relationship with the printer. Only a long-term partner will take the time to resolve a problem like this. Keep this in mind as you choose printers for your own work.

1. 1. Color on laminated, digitally printed pages does not look the same as color on unlaminated pages.

1. 1. If you make a color change in part of a job, this may adversely affect the color in another part of the job. This is true for offset printing as well as digital printing.

1. 1. People see color differently, depending on their gender (women see color better than men) and on many other variables, and color can look different depending on the lighting conditions and the surrounding colors. (Red paint in a closed paint can is actually black, since color is a function of light and the physical action of the cones and rods in your eyes. That’s why red cars look gray under street lights at night.)

1. 1. The time comes when “good enough” is good enough. Only you can make this call. In my case, only the client who devised the color chip product for selecting make-up and clothing can say whether the colors in the proofs are close enough to the original colors she chose for her fashion system.

1. When in doubt, start with the obvious, and start at the beginning. For instance, I asked my client to check the cyan, magenta, yellow, and black color builds in the InDesign art files to make sure the color builds in the original books and the 30 extra loose colors were identical. I may also ask the printer to make sure the printing paper for the proofs is the same as it was for the original 22 books (whiteness and brightness). I already asked whether there’s any chance that the PDF files or InDesign files for the same job could be “off” (damaged, inconsistent, etc.).

As you can see, this is not an exact science. A lot of people at the book printer have been working hard to make this right so my client will be happy. And she has been patient. We’ll see what happens.

This entry was posted on Monday, August 28th, 2017 at 1:04 pm and is filed under Digital Printing. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.