Test translations are one of the most disputed matters in the translation industry. Clients use them when selecting a translation agency and the latter uses them to hire translators. Some people think tests are a necessary evil; others believe they are compulsory; while some question their usefulness altogether. Let’s consider this phenomenon from different points of view.

Translator: why should I do this test for free?

Hardly a day passes without this question being discussed in professional web forums, at conferences, on social media, or just talking with colleagues. This has to be the second most widely debated topic, after rates.

The question is a legitimate one, as at least one hour should be spent typing up a test translation from a prospective client, even though its successful completion does not guarantee a receipt of orders — especially if there are dozens of bidders. What’s the point of spending your time on passing tests, if there’s only a small chance of receiving an order from a translation agency? Wouldn’t it be fairer to ask the client to pay for this one hour of work, if the test quality meets the requirements? To illustrate a translator’s sense of injustice, they often provide examples of plumbers and tilers who will not do their work for free.

Speaking of justice, let’s consider the matter comprehensively and try to see why the analogy “plumber — translator” is inappropriate.

Suppose that a translation agency needs two translators. It publishes a test to select the most qualified applicants. The test translations are completed by 25 translators; out of which 22 are bad and 3 are good. As a result, one good translator will either take their place “on the bench”, or will be “overboard”, but he or she is not paid for the good test translation; may receive no work, and feels deceived. In addition, 6 or 7 translators may also feel deceived, as although they did not pass the test, they believe their translations are good enough, and may demand an explanation and/or compensation.

If the principle of justice is to be followed, then good work really should be paid for. However, this principle works in both directions. Following that logic, if a translation falls below the required standard, then the translator should reimburse the translation agency both for the cost of checking their work and for the time spent investigating disputable situations where the original test evaluator is ultimately correct. The courts work similarly: “The defeated” party not only pays the amount of the claim but also reimburses the legal costs of the “winning” party.

Are translators ready to pass the tests based on the principle of mutual financial responsibility? Will the translator venture to offer the following conditions themselves: “If my test translation meets such and such criteria, you pay me, and if not, I will pay you”? In 15 years of this author’s experience, this has never been the case.

Suppose how such a principle could be implemented in practice. For instance, a translation agency indicates the following cooperation conditions: “We will pay for a translation test that is completed to a high standard. However, we require that you pay for its quality assurance if the translation fails to meet the minimum quality requirements. Disputes will be reviewed by an independent third party, an arbiter, whose services will be paid for by the party that was found to be incorrect.” Even though such conditions are logical and fair (each party is financially responsible for its incompetence, if it should be detected), it is doubtful whether translators will accept them. This is not only a matter of the absence of mechanisms, but rather an unwillingness to follow this principle. Mechanisms can be created if the parties are ready for such “rules of the game”.

The current situation with tests does not imply payment for the effort of passing. On the contrary, translators who can prove their professional suitability, often pay for their right to pass the test. Sometimes this is a considerable amount of money. For instance, the price of passing the well-known test TOEFL in Ukraine is about US$180 (in other countries this amount varies from $160 to $250). The rescheduling of tests, reviews, etc., are considered to be additional services and are paid for separately. The price of taking the GMAT test is US$250. Many translation associations, such as the ATA (American Translators Association), have a paid certification where you have to pass tests.

A translator spends time to pass tests, so in theory, they could demand the organizers of the tests pay for this time. However, people simply take this for granted; canceling the set price is not discussed. The explanation for this is very simple: first, a translator agrees to the testing conditions themselves, and second; they understand that this money pays for the work of the examiners.

Tests by translation agencies are in essence the same certification, only a corporative one. Translation agencies also pay for the work of their examiners. The conditions are announced from the beginning: from the very start, a translator knows that nothing is paid for passing a test, and its successful completion does not guarantee that he will receive a job. A translator has either to accept it and take the risk, or just refuse to take the test.

A statement saying that situations with tests are “unique”, and that “other specialists do not work for free” is slyness or a misunderstanding of the situation. If you decide, for instance, to install blinds on your windows, a measurer will come to you who will spend his time but does not have a guarantee to get the order. Sellers of various products give samples of their products but do not request any guarantee that these products will be purchased. If you try a new coffee brand at the supermarket, nobody will make you purchase it.

A test, in its essence, is kind of a job interview. You spend your time on any job interview, yet it occurs to nobody to make your potential future employer pay, although a job applicant can spend several hours on getting to the office and speaking to the HR manager. You should think of a test translation more as a marketing activity and promotion of your services than actual work.

Another important point is who contacted whom: a translator who contacted a translation agency, or vice versa. If a translation agency finds a translator and is interested in working with them, a translator has a strong negotiating position: they can insist on cooperation without a test. If they file their application with the translation agency themselves, he or she is on par with their fellow competitors and the translation agency has no reason to make an exception for them.

The following reasoning is also popular: “Why should I spend time on a test, if I can accomplish a real order during this time?” The counterargument to this is: indeed, why should you take the test if you already have orders? Why do you need new clients when you are fully loaded by the existing ones? If you pass the test, you will not find time for the new orders anyway. If you do not pass the test, again you will lose time. In such situations, it is logical not to pay attention to new job offers and save time both for yourself and your clients.

In conclusion of this section, we would like to add that a translation agency understands translators’ emotions very well, as they find themselves in the same position quite often: they regularly complete test translations for their clients; engaging their best experts in this activity and subsequently do not receive any orders. Nevertheless, they continue taking the tests as the same principles apply to both them and their translators, only on another level.

Therefore, a test should not be perceived as a usual order: it is not a job but a step to getting the job. It is like putting out feelers — no result is guaranteed, but you will get no result at all if you do not do it.

Translation agency: how to assess a translator’s competence differently?

Why do translation agencies prefer to select translators using test translations? They could select the best-looking CVs, be guided by references, or look through the translators’ portfolios. Why do they so seldom use alternative methods?

Let’s start with CVs: first of all, they show the ability of an applicant to “sell themselves”. Unfortunately, marketing skills are not always an indicator of professional competence. Fine words and beautiful looks very often hide a rather mediocre translator. And vice versa, a modest or unattractive curriculum vitae can mask a gifted person. That’s why a method of selecting translators by CV often does not produce the desired results.

At first glance, references look like a more reliable source of information. Nevertheless, they also do not yield 100% accurate results. Firstly, there is no guarantee that the person who gave these references and the person who received them have a strictly professional relationship: a flattering reference maybe just a friendly courtesy. Second, there is no confidence that the level of quality required from the translator by the referee was similar to the level required for work at (or with) the translation agency. It’s quite probable that the previous employer or client was just incapable of assessing the translator’s work professionally. That’s why references should come from those whose requirements are not lower than that of the translation agency. Ideally, they should come from respectable competitor companies. However, translation agencies rarely provide such references, as they have no intention of giving their valuable human resources to their rivals. As a result, the amount of valuable references is small.

As for translation portfolios provided by applicants, they also give rise to doubts and concerns. Doubts — because the translation could be made by anybody, or simply downloaded from the web. Concern — because a translator could upload their translation to the web without obtaining the client’s permission, violating confidentiality principles. Such violations may cause serious difficulties for the translation agency that gave the work.

Therefore, if a translation agency does not know a translator, tests are the most reliable way of hiring. This situation will last until some authoritative organization emerges that has tests for translators and issues a certificate after their successful completion. Everybody wins: qualified translators will not spend time on passing tests at every translation agency, while translation agencies will drastically reduce both the costs of hiring and their risk from working with new translators.

It is also worth noting that such organizations exist in some countries, for instance, the ATA in the United States. Unfortunately though, in most regions such certification institutions are either absent or are simply not trusted by market participants.

Client: how do I know that you will complete the order to the same high standard as in the test?

It is a risk for clients that a translation agency will engage its best translators to complete a test translation, while it will use less qualified translators in any further work. Is there any sense in selecting a translation agency using tests?

Some translation agencies use such tricks. That’s why a well-made test translation does not guarantee that subsequent orders will be done with the same quality, particularly if several translators are to be engaged. However, tests at least allow the filtering of those suppliers of translation services who do not have professional human resources, from even making a first impression.

In other words, tests do not guarantee quality but rather reduce risks. One of the options for a further reduction of risk is to give a translation agency a small, “real” order and then check the quality before giving them a larger order.

Such methods are used by translation agencies in their relations with freelance translators: first, they are tested, then they are given a small order, and only afterward do they start providing regular work. If the work is of good quality, the translator is practically always provided with work.

Test or a translation for free: how to detect a fraud?

Inexperienced translators tend to have “test phobia”, i.e. they believe that translation agencies give them real orders disguised as tests, thus earning money on naive beginners and not paying them a penny.

This is, of course, nonsense. Professional translation agencies never stoop to such manipulations, as it does not make sense for the following reasons:

• Economically unreasonable. Even if they divide say a 10-page order among ten trusted translators, it will be a very difficult and costly task to combine their translations (made with good quality) into one coherent text. That’s why, under conditions when one translator is not able to complete a translation because of a deadline, the project is always divided among the minimal number of translators, to make the editor’s life easier. To give this 10-page order to ten applicants for the position of translator, whose competence is unknown, means complicating things even more: not only will all the parts be translated differently, but at least three-quarters will be translated very poorly. Add together the costs for project coordination (managers also need to be paid), the posting of “fake” job announcements, and its related headache, and you will see that it is cheaper to give the text to a trusted translator. Consequently, it is not profitable to disguise real orders as tests, not to mention the ethical aspect of this. It is much more profitable to earn money on long-term cooperation with translators than to earn one time by fooling them. In reality, there are few good translators. Therefore, if a translation agency does not pay for the completed translation, it commits corporative suicide.
• Reputation. If a translation agency has worked in the market for a long time and has a good reputation with its clients and translators, it will not risk losing it for earning peanuts on fake tests: sooner or later this will result in losing clients and making financial losses. If this fraud is found out, a translator may punish this translation agency, by disseminating on the web its dishonest practices, and it will be much more difficult for this company to hire new translators.

But where does this translator’s phobia come from? It is very simple: casual persons and non-professional translation agencies that have not worked in the market for very long sometimes use such a scheme. They do not have a brand or a reputation; they are not going to build a respectable company that will work in the market for many years to come. They think of using the scheme only to “earn and run away”.

How can you differentiate such fraudulent and short-lived firms from serious translation agencies? There is always a risk of coming across them, but making several simple checks can considerably reduce it:

1. Look for the official website of a translation agency and assess its quality. It is doubtful that a translation agency that spends good money on high-quality design will be penny-wise and pound-foolish. If there is no website, a question arises as to whether such a company exists at all.
2. Find out more about the company. For instance, how many years has it worked in the market: the longer it has worked, the less risk there is. Look for feedback about this company on the web, in particular try to find comments on ProZ.com and other professional portals. Google its telephone number, email address, and sender’s name. Try to find its profile on LinkedIn.
3. Pay attention to the mailbox the company used to send you emails. A serious company always has its corporate email domain. If you receive emails from a free Gmail account or yahoo.com, it should make you think.
4. The volume of a test should not be too big. One page (250 to 300 words) is enough to assess the quality of translation. If the test is more than 500 words, refuse without doubt, or ask to reduce it.

If you find some suspicious information about a translation agency, ask the test sender directly, or do not engage with this translation agency at all.

A skill of no small importance is to “filter negative feedback” posted about translation agencies on the web. For instance, if a translation agency is a large one and 1000 translators work with it, surely some do not like something as statistics work this way. If out of one thousand translators 980 cooperate with a translation agency successfully and 20 express their indignation, it means that those 20 people’s claims are not well founded, as 980 translators do not notice that they are fooled. Anonymous comments should particularly alert you. When a person does not say his or her name, they do not take any responsibility for their words. It is quite possible that they invented their claims. That’s why you should be sure to weigh the company size and the number of negative comments before making your conclusions.

Conclusions…

As you can see, the subject of translation tests is a complicated and multifaceted one. Mechanisms of their use are not always well elaborated. Sometimes questions arise regarding their usefulness and ethical aspects. However, at this point, the translation industry has not invented effective alternative mechanisms for assessing a translator’s qualification. Therefore, all parties have to accept them as given and treat their use with understanding. An alternative way is to create such mechanisms, but such an option requires joint will and organizational efforts from all the professional market players.