Demonstrating the value of editingby Geoff Hart, Associate Fellow, STC Montreal
Like all other technical communicators, we editors must sometimes struggle to prove our worth to employers. We know our value, and the more clueful of our authors understand, but sometimes it takes a bit more work to convince senior managers that we serve a useful purpose. Managers generally require specific examples, usually supported by hard numbers. In this article, I've provided a few random facts and figures that I've accumulated over the years that you can share with management. My case for demonstrating our worth won't be truly compelling because of the scattershot nature of my examples. Nonetheless, they are compelling and may inspire you to begin collecting your own examples.
The examples I propose to offer for your edification fall into five broad categories:
- defense against embarrassment
- elimination of reprinting costs
- defense against lawsuits
- ensuring clear communication
- reduced translation costs
Without further ado, let's see some of the ways in which editors have (or could have) saved their employers considerable grief.
Defense against embarrassmentIn a previous job, our desktop publisher was handed an important strategic plan to lay out and print, and wisely thought to ask the authors whether the manuscript had been edited. It had not been. The story was familiar: By then, the plan was so late (and I was so busy at the time) that the authors thought it would be better to eliminate the editing stage and simply rush the document into print. Forewarned, I was able to convince them I would drop everything else and stay late if necessary to complete the edit on time, and mercifully, they agreed. This proved to be fortunate indeed: in addition to the usual grammatical and other infelicities, the authors had somehow managed to get the name of a key group within our organization wrong! Since the report would have been distributed to thousands of influential partners and cooperators, you can imagine the embarrassment this error would have caused that group.
Such relatively minor errors may be forgiven by readers, but sometimes the repercussions are far more severe. Several years ago, a colleague reported a simple typo that she noted "could have had international repercussions": the word assess became asses due to a dropped S. Even more seriously, Randy Cassingham (publisher of the wonderful This is True newsletter) reported egregious errors in Microsoft's thesaurus for Word in 1996 that forced the company to formally apologize to customers in Mexico and Spain: apparently the Spanish version of the thesaurus offered such suggestions for Indian as savage and man-eaterâ€”a serious offense given the number of Mexicans with Indian heritage. The English version was not immune to this kind of problem and contained offensive errors of its own, offering pervert as a synonym for lesbian and civilized as a synonym for western (http://www.thisistrue.com/corporate_wordplay_ii_5029.html).
Another colleague reported the time her managers chose to publish an entire symposium proceedings as "camera-ready copy" (i.e., the authors submitted the final page layout to the printer without any editing or proofreading by the publisher) without her involvement. Like me, she was perceived as being too busy to spare the time to do the work, so the manager responsible for the conference went around her. One author, who had spent his entire career butting his head against bureaucratic walls and fighting various political battles with professional colleagues, decided that this was the perfect opportunity to get even for more than 30 years of perceived humiliations, and published a jeremiad against his employer and colleagues. Since nobody edited or even proofread the collected documents, the offensive article was not spotted in time, the author's rant was published and distributed widely, and the entire proceedings had to be recalled and destroyed when a reader reported the problem.
Elimination of reprinting costsAs the previous example illustrates, embarrassment can also have a significant financial cost: the need to reprint an entire print run of a publication. The Grand Haven Tribune reports (http://www.grandhaventribune.com/paid/305915756085324.bsp) a recent example of the infamous pub(l)ic typo. In this case, the offending document was the Nov. 7 election ballot for Michigan's Ottawa County, and the missing L cost the county more than US$50,000: they were forced to reprint 170,000 ballots at a cost of roughly US$0.30 per ballot. Although "about five or six people proofread it," clearly none of them was a professional editor. Ironically, the Tribune article stated that the cost to the county was US$40,000, which is clearly incorrect if the stated number of ballots and printing cost are both correct; miscellaneous setup and other costs may account for the missing US$11,000, but my 20 years of experience working with scientific authors suggest this really is a simple arithmetic error. This illustrates that even editors are human and subject to errorâ€”but as outsiders, we are less familiar with the original writing and are thus less likely to simply accept the author's assumptions and let such errors pass.
Similarly, several proofreading errors cost the Chicago Transit Authority (CTA) US$75,000 and cost at least one employee's job. As CBS Chicago reports (http://cbs2chicago.com/topstories/local_story_230061241.html), CTA incorrectly printed the Belmont station name as Bemont (what is it about the letter L that makes it so easy to miss?), omitted the Fullerton stop, and provided the wrong telephone number (for a private individual's cell phone) at the bottom of the signs. These errors required the reprinting of 3,000 signs for CTA's trains. In addition, the person responsible for proofreading the signs was fired.
Defense against lawsuitsIt's interesting to speculate how much a good editor can save a newspaper or television or radio station annually in libel charges, thereby also reducing fees for libel insurance. Similarly, one wonders how many editors have saved the literary careers of their authors, or have saved publishers from messy lawsuits. Undoubtedly, a legal scholar could provide a great many examples of cases where foolhardy cost reductions (i.e., eliminating one or more editing stages) caused precisely such problems. This would make for a fascinating case study for STC. (Am I hinting too broadly?)
When I asked colleagues for specific examples of this kind of editorial benefit, several colleagues provided good ones, including "several examples of poor wording in environmental reports that left the author/company open to lawsuits"; as I was told these were confidential, I cannot present them here. One colleague who contributed her example in the copyediting-l discussion group (http://www.copyediting-l.info), noted how she came across the phrase "Eat shit and git!" in a set of review comments and author's responses for a contractor's report that was about to be submitted to the US Environmental Protection Agency. The phrase had somehow made it through the previous stages in the review and revision process, and because the finished report was due by the end of the day, there was a serious risk that the phrase would have made it into print without the editor's oversight.
Ensuring clear communicationThough we're often tempted (or even requested) to confine ourselves to copyediting, good editors understand that despite the importance of punctuation, sometimes the meaning is too important for punctuation alone to carry the full burden. As NPR recently reported in the story "Million-Dollar Comma May Aid Canadian Company," (http://www.npr.org/templates/story/story.php?storyId=6383383&ft=1&f=1004), sometimes the subtleties of punctuation are simply too subtle. A recent Canadian contract dispute centered on the problems posed by a single misplaced comma that may save one of the signatories an estimated Can$2 million. In this case, Rogers Communications signed a contract with Bell Aliant to use Bell's telephone poles, and when Bell Aliant sought to end the deal, they noticed the following clause: "This agreement shall be effective from the date it is made and shall continue in force for a period of five (5) years from the date it is made, and thereafter for successive five (5) year terms, unless and until terminated by one year prior notice in writing by either party." As the government regulator that ruled on this dispute noted, the second comma makes the "and thereafter" phrase parenthetical rather than a crucial part of the concluding phrase. The result is that the contract is valid for 5 years unless Bell gives 1 year's notice. Presenting this as two separate clauses would have eliminated the confusion.
Editors can also contribute to the peer review process by polishing manuscripts to eliminate the usual range of punctuation and grammar errors, thereby improving the manuscript's ability to communicate with the reviewer. This kind of work is appreciated. Over the years, I've received a great many thank-you letters from journal editors for helping them make the review process go more smoothly; indeed, I now earn a comfortable living helping authors for whom English is a second language publish their research in English journals. One of my proudest moments was when one journal actually named me "associate editor" to reward my efforts on their behalf!
When I worked for the Canadian federal government, I found that manuscripts I edited before submission to a journal were almost never returned for a failure to follow the journal's formatting guidelinesâ€”and when it did happen, it was usually my fault for not doing a better job of policing my authors and reviewing the corrected manuscript before submission. Back in the pre-Internet days, such rejections cost us the price of one courier delivery plus a week or more of lost time. Now that we use e-mail, a rejection doesn't cost us any money and we can resubmit the manuscript within hours rather than losing a day or two while the courier package makes its way to the journal. Nonetheless, why not do the job right the first time? But we also rarely received significant copyediting comments from journal reviewers. This is important because it meant that the journal reviewers were focusing almost exclusively on the scientific content of the manuscript rather than on points of grammar and style. The implications are clear: my authors were communicating highly complex material so clearly that reviewers could concentrate on the technical and scientific content, not the writing. Unfortunately, I never collected any statistics on the results of the reviews: the difference between the reviews of edited and unedited manuscripts was so dramatic I never felt motivated to do so.
More importantly, the quality of the reviews improved. It seems logical that if reviewers aren't focusing on correcting the writing style and don't have to use all their concentration to understand what the author is saying, they can devote more of their finite supply of mental resources to critiquing the science. Similarly, there's considerable anecdotal evidence from the technical writing community that providing well-edited drafts for review by subject-matter experts helps those experts catch a greater number of substantive errors. Unfortunately, I'm not aware of any studies that specifically collected statistics on how editing shifts the review process from copyediting to substantive review. If you're aware of any such studies, please send me detailsâ€”and if not, perhaps this represents an opportunity for STC's academic members to do the research.
Reduced translation costsOne colleague in Toronto told me about the results of "a confidential internal study," in which one of his larger clients demonstrated how careful editing of their publications was saving tens of millions of dollars in reduced translation costs. Because I was unable to contact the company to obtain specific details, treat this as purely anecdotal evidence. However, you can see the possible benefits using a little simple math. Here's a personal example.
In my own work, I do French-to-English technical translations of documents sufficiently unique that translation memory tools don't help much: many phrases are reused, but so many differ between manuscripts that we haven't been justified in implementing full-scale machine-assisted translation. In many cases, I was asked to edit the manuscripts before they went for translation, and authors or their supervisors would ask me to cut the length of a manuscript by 25 to 50% to reduce printing and mailing costs. For a typical manuscript of about 2,000 words, that means a reduction of 500 to 1,000 words. At a rate of Can$0.20 per word, the going rate for these translations in our area, this word reduction amounts to Can$100 to Can$200 in savings per manuscript. Given that the editing work necessary to reduce text by this much typically took me a maximum of 1 to 2 hr (for 25 and 50% reductions, respectively) for documents of this length, the cost of the work amounted to Can$65 to Can$130 at my standard hourly rate. The result? A savings of Can$35 to Can$70 per manuscriptâ€”and often more. That's not enough to cover the cost of the editing, but it does greatly reduce the net cost. The reduction in printing costs from a 50% reduction in the manuscript's length are likely to be far more significant; for a large print run, they would more than cover the cost of editing.
You can easily extrapolate these figures to other situations. For example, in areas such as software documentation, where there is considerable repetition within any given text, translation memories let translators automatically reuse standard translations for repeated words and phrases, thereby eliminating the need to retranslate this text for each new document. As a result of this automation, translation costs are much lowerâ€”typically nearer to US$0.01 per word. Let's use that figure to simplify the calculations. Using the same length reductions I reported in my personal example, the savings per manuscript (500 to 1,000 words per 2,000-word manuscript) would amount to savings of US$5 to US$10. It's easy to modify this figure using your actual translation cost per word and the actual length of the document; for example, the savings for a 100,000-word software manual at a translation cost of US$0.05 per word would equal US$1,250 to US$2,500.
It's possible to come up with similar estimates for cost savings for printing costs, and if your work is published in a forum (most commonly, peer-reviewed journals) that charges authors a fixed fee per page of text, the savings can be substantial at a cost of US$100 per page (not an unusual cost).