Corrections: Typesetters’ Dirty Little Secret

By David Alan Rech of Scribe Inc.

David Alan Rech


We have a secret to tell you. Take a look at your typesetter’s bill. Do you notice how expensive the corrections charges are? In Scribe’s experience, the amount charged for corrections is often substantial and can represent up to 50 percent of the total bill from your typesetter. More than one of our competitors has remarked that they were sure of continued business because publishers couldn’t cure themselves of corrections. This has been kept a secret because it adds to revenue, but corrections in pages are the most expensive alterations you can make. And while making alterations in pages is common, it’s expensive and easily avoided.

This claim may seem contradictory to your viewpoint. After all, making a change in a desktop publishing program seems easy and quick. Consider, however, the number of people who must review an alteration after pages are set. First, there is the creation of the page proofs—that’s one person. The proofs then must be received, which takes one or two people, depending on whether pages are sent directly to the author or through an editor. The review process can involve up to three people if there are separate examinations by the author, editor, and proofreader. If you have simultaneous reviews (because somehow people believe this saves time), then someone must compile alterations. An editor will often assess the alterations to make sure they should be made, adding yet another player to the mix. The alterations must be entered (that’s another person) and checked (ditto). Finally, a new set of pages for the next review must be created. All in all, this process can involve up to ten individuals (and the process described does not include the need to recheck the full set of pages at every proof stage). In this light, an alteration no longer seems to be that simple.

In the past, proofreading and page alterations were necessary because errors cropped up in the process of setting type. In the computer age, typesetting errors should be rare. In our analysis, page alterations are rarely due to “printer error” (PE). In fact, PEs make up less than one-tenth of 1 percent of reported alterations in Scribe. Yes, typesetters make errors, and missing those errors can be critical. But typesetters are not the cause of the vast majority of alterations, nor is traditional proofreading the best way to catch PEs.

Alterations in pages are caused by other factors, including errors that carry over from copyediting, unnecessary editorial changes, frivolous author alterations, poor planning, “proofreading” changes (which are frequently not real errors but editorial alterations), and going to pages prior to having a solid manuscript. There are a number of easy steps you can take to reduce the number of alterations in pages and improve your books.

Change your copyediting and proofreading procedures and expectations. Certainly, it is important to provide copyeditors with house styles and general guidelines. But it does not end there. Our best clients provide explicit instructions, request samples, convey their expectations, and hold us to them. It is important that copyediting results in the highest level of manuscript integrity and that all issues are resolved during this phase.

Proofreading should not be a second copyedit. Proofreaders should be provided with explicit instructions about what to catch and what to ignore. Also, they must be informed about the upstream editorial decisions. I cannot list the number of times proofreaders undid a conscious—and correct—decision an editor made because they had a differing opinion of how to treat a manuscript (or were misinformed about said decision). This not only adds time to the project (and results in alterations) but also degrades the book.

Solidify your manuscripts before typesetting. We have mentioned this in another article, but it is important to finalize your manuscript before you typeset it. Modern typesetting, even with complex books, is a fairly quick process. Waiting to typeset until the manuscript is complete and completely edited does not disrupt schedules. There are certain procedures that require a typeset page, but most of the text-based, editorial, work should be done before books are typeset. And the time involved in major shifts in books is always longer than doing it correctly the first time.1

Be more disciplined with your authors. You must limit the level of alterations that authors perform in pages and explain the reason for this. It’s also important to make sure that authors do not consider the manipulation of typeset pages to be part of the writing process. If they need this, there are ways to create setting manuscripts or undertake other strategies to allow them to change a “book” without having to alter pages. Authors need to understand that alterations in pages should only be made if there are egregious errors. And as publishing professionals, we need to help them control their impulses. Even the most high-profile authors can be co-opted to work more efficiently.

Watch your decision-making process (plan in advance). It is crucial to plan all aspects of a publication, including the desired features, editorial approach, and design, prior to typesetting the book. Developing a reliable cast-off procedure and having all the ancillaries (images, tables, figures, appendixes, etc.) in your possession prior to setting a book is also important. We consider the reflow caused to reduce the page count or adjustments due to removing, adding, or swapping images to be unnecessary alterations. Be sure to consider all the ramifications of changing course prior to a change in pages. We once had a client change her mind regarding the opening spread of each chapter, a seemingly simple alteration. The resulting page shifts, however, caused us to have to rework the index. What seems an important aesthetic choice can become an expensive one that may not provide a good return on the investment.

Start a typeset quality control (QC) process. Typesetting errors are important to catch. A thorough, systematic typeset quality control process is more likely to catch layout errors than a traditional proofread. The QC should be based on a checklist that the person performing the task is going through to systematically check elements.2 Errors in pagination, running heads, chapter titles, and so on are more easily discovered through a good QC than through reading text—more so if you use available tools to check appropriate elements.

Stop relying on PDF files. PDF files are better than print in many ways, but they too have their limitations. At some point, however, a print book must be reviewed in PDF. An undisciplined page review and proofreading process that uses PDF files creates massive inefficiency. First, the reality is that even good proofreaders will miss things. (They are only human, after all.) PDF files have no tools to aid in proofreading or to catch errors and inconsistencies. A word or text processor is better equipped to facilitate proofing of manuscripts.

Another problem is how alterations are indicated in PDF files. Usually, the corrections are done with comments or marks directly on the PDF page. Reading, entering, and checking alterations in PDF format is slow and prone to error. Consider the typesetter who has a PDF and the typeset program (for this example, InDesign) open on his or her computer. He or she must find the error, find the page in InDesign, make the alteration, and check to make sure the correction is applied without error. All of this is done while flipping between the two programs. This is a slow process. If you cannot stop the proofreading process, it is best to provide the typesetter with a simple report (e.g., a list in a Word or text file) of what needs to be altered. This will help speed up the process and allow the typesetter to employ a quicker method that is less prone to error, such as copy and paste (which, in turn, avoids any corrections that need to be retyped, a common introduction to mistakes). It also helps to avoid the same inefficiency by the editor who usually checks page alterations prior to the typesetter entering them.

At Scribe, we have been trying a number of experiments to lower the level of alterations in pages. These include the practices enumerated earlier as well as ways to accommodate insistent authors and editors. In every case, we have found the total time on a book to go down, and the resulting quality is either better than or consistent with previous methods. Thus the goal of reducing alterations in pages has resulted in higher quality and lower overall cost. At this point, it seems necessary to abandon practices that have their roots in movable type and develop new ones grounded in the reality of today’s electronic publishing world. It’s time to expose the current process of alterations in pages for what it is: a bad habit encouraged by vendors seeking profit. With this secret now exposed, we need to find better ways to publish books that end our dependency.

1 This time accounts for altering books because of shifts, added content, and the like. It also considers correcting the errors that always result from this type of process.

2 For an example of this, feel free to visit