A few thoughts on test translations

Test translations, whereby a potential client, often a translation company, asks a translator to complete an “audition” translation for free before beginning work with the client, are a frequent subject of controversy in the industry. Translators wonder if they should complete unpaid translations, if they should set a limit on the length of test translations, if they should offer to provide samples of their work instead, or if clients will be reluctant to use them if they do not complete test translations. Clients understandably want to hire highly-skilled translators, and one element of this is often to give the translator a test that many other translators have taken, in order to compare the new translator’s work with that of established and trusted translators. In rare cases, translators either suspect or have proof that unscrupulous clients have used “unpaid tests” as a way to get some translation work done for free, which adds to the atmosphere of distrust surrounding tests.

One of the most common questions surrounding test translations (from the translator’s point of view) is “should I take unpaid tests?” In and of itself, this question doesn’t provide enough information to ensure a reliable answer, since the answer depends on a variety of factors. How long is the unpaid test? Who is the client? Is the test a tryout for a specific project? How much does the translator need or want the work in question? All of these are important factors to consider when deciding whether the time investment in taking an unpaid test is a good one.

Interestingly, item D of the ATA Code of Professional Conduct and Business Practices for “employers or contractors of translators and/or interpreters” reads: “I will not require translators or interpreters to do unpaid work for the prospect of a paid assignment.” Although I guess that there is some room for interpretation (so to speak!) here, i.e. what constitutes “requiring” work; is a test translation “unpaid work” or something else entirely, this clause seems to take a stance against unpaid test translations, at least when they are given by translation companies that are ATA members.

For translation companies, the alternatives to unpaid tests, such as paid tests or small paid assignments, are more expensive and more risky, unless they have vetted the translator’s background and experience before administering the test. For an agency that receives many unsolicited resumes, it’s much easier to have the next step in the application process be “if you’re interested in working for us, complete this test and return it” (which is likely to weed out many candidates). Also, translation is not the only profession where unpaid testing, even if it’s not referred to as such, occurs. Within the past few years, I’ve conducted my own “unpaid tests” on a primary care doctor and a financial planner, both of whom offered a complimentary half-hour consultation before I committed to using their services.

Ideally, translation companies should consider paid translation tests as a cost of doing business, in the same sense that they see recruiting, hiring and training their in-house employees as a cost of business. Barring that, I think that it makes sense for translators to set some guidelines on when and how they take unpaid tests. In my own case, I ask the potential client to confirm two things before I take an unpaid test: 1) that they currently have or anticipate having a need for additional translators in my language combination; I do not take unpaid tests for the purpose of being added to an agency’s general pool; and 2) that my rates (and I provide a rate sheet) are within the range of rates that they pay for my language combination. If the agency cannot confirm these, I don’t take the test; other translators at a different point in their careers may have a different opinion about this. In addition, I think that it makes sense to put a limit on the number of words that you will translate as an unpaid test; I think that half an hour’s work (for me this would be 200 to 250 words) is a reasonable limit, but again this is a personal decision.

Yet another option is for the translator to offer the prospective client some samples of work that he/she has done in his/her areas of specialization. To me, this is a more reliable indicator of the quality of the translator’s work and it also allows the agency to review a larger sample of the translator’s work than it can by administering an unpaid test.

5 Responses to “A few thoughts on test translations”
  1. Jill April 16, 2008
  2. Kerilyn Sappington April 16, 2008
  3. Simac April 17, 2008
  4. Corinne McKay April 18, 2008
  5. Corinne McKay April 18, 2008

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.