Are long translation test a rip off?
Thread poster: Rosa Paredes

Rosa Paredes  Identity Verified
Canada
Local time: 08:54
Member
Spanish to English
+ ...
Aug 5, 2012

Hi,
I'd like your opinions on translation tests required by some agencies/clients. IMO a couple of well-chosen sentences are all you need in order to assess a translator's skills; however, the average test is 200-300 words long (one full page or more). Just a few days ago I was approached by an agency that requested a 1000 words (free of charge) test translation which I refused to do. If you have posted some sample translations on your profile/if your credentials have been verified, shouldn't that be enough?
Thanks in advance for your feedback.


Direct link Reply with quote
 

Farzad Akmali  Identity Verified
Member
English to Farsi (Persian)
+ ...
Mostly a rip off! Aug 5, 2012

Rosa Paredes wrote:

Hi,
I'd like your opinions on translation tests required by some agencies/clients. IMO a couple of well-chosen sentences are all you need in order to assess a translator's skills; however, the average test is 200-300 words long (one full page or more). Just a few days ago I was approached by an agency that requested a 1000 words (free of charge) test translation which I refused to do. If you have posted some sample translations on your profile/if your credentials have been verified, shouldn't that be enough?
Thanks in advance for your feedback.


Hi Rosa,

In all years I have been translating, I only accepted one long translation test of three pages, although I was not sure about it and asked the client why three pages, they told that they wanted to make sure about the quality as the project is a book; I contacted some of friends 9but too late as I had delivered the translation) and they had received different pages of the same project (as a test) and it was a cheat.

I have not been contacted for long tests since then, but I prefer to check the client's Blueboard and their reputation first and then decide on accepting or rejecting those tests.

regards,

Farzad


Direct link Reply with quote
 

Marinus Vesseur  Identity Verified
Canada
Local time: 08:54
English to Dutch
+ ...
Could be a scam... Aug 5, 2012

... so do your "due diligence" and check out the potential client. I post jobs here sometimes and I find test translations a very helpful tool, better than just the information on the individual's profile page, since that can be tweaked, but 300 words is more than enough to get a good impression.

Direct link Reply with quote
 

Tony M  Identity Verified
France
Local time: 17:54
Member
French to English
+ ...
Tests of dubious value anyway Aug 5, 2012

I think tests are a very dubious way of assessing a translator's suitability, and I steadfastly refuse to do them in all but the most exceptional circumstances, for the following reasons:

Of all the tests I have done in the past, only a very small percentage have actually resulted in real work; whereas I have had many, many more jobs from customers who did not demand a test (or accepted the fact that I refused). And as many of these have become regular customers, one can only assume that they must have found my work satisfactory in the longer term.

I actually tend to think that the sort of agencies (usually) who ask for tests are probably operating with a certain mentality that likely isn't too compatible with my own way of working. hence 'asking for tests' is a negative criterion for me when selecting which agencies I wish to work with.

I do not believe that it is possible to satisfactorily judge someone's skills on the basis of a short test, unless it has been very carefully chosen (which rarely seems to be the case); what seems to me most revealing about a translator's ability is how they deal with problems, and any short test is going to be fairly random as to the problems that may or may not arise.

In addition, people often ask me for tests in a hurry — why, only last week-end, I had a rather peremptory demand for a test on Saturday afternoon, to be done for first thing Monday morning! I rather frostily replied that I don't do free tests, and even if I did, I certainly wouldn't do one as a rush job AND at a weekend without billing my usual surcharges!

Now if I take on a new project, possibly in a field that is not totally familiar to me, on the basis of a meaty translation, I will take the time to read up on the subject, do a lot of research, and certainly take the time to work and re-work my translation as many times as necessary to deliver a polished, high-quality product. It is clearly not economically viable to do all this for a free test, hence the result is unlikely to be in any way representative of the normal quality of my work; and however short-sighted some people might think it, I'll be damned if I'll give up proper paid work today in order to do a free test with a promise of "bread tomorrow".

In a few exceptional circumstances, I have done short tests; in this case, I have included within the document a statement to the effect that this test is not truly representative of my work; in addition, instead of researching tricky terms, I have been very upfront, leaving comments about terms I was unsure of, and explaining, for example, that I think it would be a or b, but would need to do more extensive research to check. On a number of occasions, delighted customers have come back to me saying that this way of working gave them a much better insight into the way I work, and that they were able to tell from the intelligence and pertinence of my questions how much I knew about the field — and of course, I got the job!

I am particularly mistrustful of those tests by agencies who say they are 'bidding for a contract', and hence both they and the end customer will be selecting on the basis of my test. I have never, ever had any work, or even feedback, from such people. Now, it could just be I'm a lousy translator... but I hope not, and my other customers seem happy enough! So perhaps they are just a lousy agency, not able to put together a convincing bid, or taking too big a margin to make it economially viable. Also, people often ask for free tests before asking my rates; sometimes, they say "Your test was fine, we have selected you for this job... but your rates are too high; will you work for [30% – 50% less than my rate]?" Now, before even telling them whether or not I will do their test, I first tell them my rates, and if they throw up their hands in horror, the matter stops there...

One of the 'exceptional circumstances' I mentioned above is when I am approached by a new direct customer; in this case, they may not be familiar with commissioning translations, and without the safety of an agency acting as intermediary, I can understand their being wary. The advantage of such customers is of course that they are falling over themselves to give you as much help and information as possible, and there is almost always a technical contact who will gladly answer my questions.

I think that 200 to 300 words is acceptable, if I have at least agreed to doing a test at all; the other day, I was asked for 1,000 words (as you were), but when I refused, they said "Well, please just select 2 separate extracts totalling approximately 300 words"; that seemed honest to me, since they clearly weren't just trying to scam a free translation out of me.

But I had my doubts about another agency, who sent me about 350 words to test, only, it was an entire document (sales leaflet) — they said they were trying to 'hook' a new customer, and I suspect they intended to do it on the back of a first 'free' translation (but not mine, honey!)

I have also been involved with translation tests from the other side, i.e. being the one who had to assess other people's work. In this case, the text was relatively short, and consisted of two fairly carefully chosen extracts (in Italian, a language I have only a very hazy idea about). I have to say, I was horrified by the results, from around 20 translators, all of whom claimed to be native speakers of English. In about ½ the cases, the quality of the translated EN text clearly proved they were lying; in other cases, the obviously native EN speakers seemed to have a pretty poor grasp of Italian — like if even I can spot their mistakes, it must be bad! Amazing how many people missed basic points, writing texts that were not even logical, and in a couple of cases, with residual spelling errors that even Word's spell-checker would have corrected. All in all, it was a very revealing exercise, and only about 3 out of the 20 produced translations that were even of an 'acceptable' quality — there wasn't one I could honestly describe as 'excellent', or even 'good'. So perhaps after all, in that particular case, the translation test was really vital.


Direct link Reply with quote
 

Marinus Vesseur  Identity Verified
Canada
Local time: 08:54
English to Dutch
+ ...
Not a good way to asses a translator's qualifications... Aug 5, 2012

... but it can be a good way to check their suitability for the job at hand, their professional attitude, whether they value speed over quality and so many other aspects, which you sorta said yourself in your extensive comment.

If I am looking for someone with experience, say, in the field of photography, their test translation will quickly reveal whether they have that experience. Of course, if you don't have that experience, it will show as well, so you'd be disqualifiying yourself by admitting it, which is more or less the point. The vast majority of colleagues declare themselves experts in too many fields - I see no other way BUT test translations if you are looking for someone with specific expertise.

Tony M wrote:

I think tests are a very dubious way of assessing a translator's suitability...

In a few exceptional circumstances, I have done short tests; in this case, I have included within the document a statement to the effect that this test is not truly representative of my work; in addition, instead of researching tricky terms, I have been very upfront, leaving comments about terms I was unsure of, and explaining, for example, that I think it would be a or b, but would need to do more extensive research to check. On a number of occasions, delighted customers have come back to me saying that this way of working gave them a much better insight into the way I work, and that they were able to tell from the intelligence and pertinence of my questions how much I knew about the field — and of course, I got the job!
..


Direct link Reply with quote
 

Angelina White  Identity Verified
United States
Local time: 08:54
Russian to English
Free Tests vs. Paid Samples Aug 6, 2012

I have done plenty of free tests, but never more than a page. Honestly, I can only recall one that actually resulted in a job. It was for an agency here in Russia, and the test was very short (about half a page or less) and pretty easy in my opinion. Since then, I have been asked for specific translation samples of my work from that same agency, but it was paid and they gave my sample along with ones from other translators to the end client to pick the sample that they liked the best. Mine was about half a page, was paid according to my regular rate, and it resulted in about a months worth of work. If I were you, I would ask the client why they require such a long test, and take it from there. If they refuse to give you a valid reason, then I would avoid working with them. Even if they do give you a valid reason, you might offer to just translate a small portion of that test for free, saying "I'm willing to translate up to 300 words of this test for free." Stand your ground - if they are a reasonable agency, they will work with you.

Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 17:54
Member (2006)
English to Afrikaans
+ ...
Several issues Aug 6, 2012

Rosa Paredes wrote:
1. IMO a couple of well-chosen sentences are all you need in order to assess a translator's skills.
2. The average test is 200-300 words long (one full page or more).
3. Just a few days ago I was approached by an agency that requested a 1000 words (free of charge) test translation which I refused to do.
4. If you have posted some sample translations on your profile/if your credentials have been verified, shouldn't that be enough?


1. A few well-chosen sentences will only help weed out the very bad translators. It will not help determine whether the translator is at least average.

2. You should decide for yourself how many words you find reasonable. Yes, the average that is considered reasonable by most translators here seem to be in the range of 200-300 words.

3. You should decide for yourself how many words you find reasonable. 1000 words seem a bit excessive to me, but it is no reason to suspect foul play.

4.1 The advantage of having all your translators translate the same test is that you can more easily compare them and determine what errors are important.

4.2 Sometimes the client uses a portion of the job itself as the test, to see if you can do it. Using portfolio translations won't be any good in such a case.

4.3 A portfolio translation is bound to be good (if the translator puts it in his portfolio) and may not represent what the translator does every day.


Direct link Reply with quote
 
There are other ways Aug 6, 2012

Hi

I was interested in this discussion because for a translations company's point of view, checking out a new translator's competence is vitally important. However, this does not necessarily have to involve a translation test piece. We have to be more flexible than that I think. Top of the list for us comes personal recommendations/references from tried and trusted translation sources (PROZ (including ID), clients, or a verifiable fellow translator). This may suffice, or we may ask for membership of a recognised professional body for freelance translators/interpreters (e.g ITI, IoL), or a list of translation projects over the past 12 months. If a translation test piece is required 200-300 words should be perfectly adequate. All WORDtrans' newly appointed translators are closely monitored until we are sure of their capabilities and reliability. WORDtrans is growing and we are also looking for new resources. If you are interested in working with us, click on this link: http://www.wordtrans.com/work-with-us/


Direct link Reply with quote
 

Tony M  Identity Verified
France
Local time: 17:54
Member
French to English
+ ...
Not necessarily... Aug 6, 2012

Samuel Murray wrote:

4.3 A portfolio translation is bound to be good (if the translator puts it in his portfolio) and may not represent what the translator does every day.


Sadly (and astonishingly!) not necessarily!

I have seen things written so badly in forums here by people claiming to be native EN speakers that I have checked their profiles; it is often immediately obvious that they most certainly are not native EN speakers, and while I am of course in no position to judge the quality of their translations EN > their native language, I think I am qualified to judge the quality of their translations in the reverse direction — at least on the level of the quality of the target text (even without reference to the accuracy of interpretation of the source text).

Already, when someone claiming to be a native EN speaker offering supposedly professional translation services into EN has a poorly-written profile page in EN, it has to rather call into question their abilities; how difficult is it to get a relatively short profile checked by a colleague or friend?

In addition, I usually look at their portfolio translations, and there it is simply flabbergasting to see the often quite appalling results, sometimes with blatant errors. Now if someone is daft enough to post a translation in their portfolio that is going to make them a laughing stock, this makes me wonder if in fact their EN is so bad that they don't even realize!

Either way, it only goes to reinforce my opinion that generally speaking, most non-native EN speakers (even if they lie about it) are not capable of offering a professional quality translation into EN. I know where to go when I (quite often) need the services of a translator into EN (or other languages of course).

[Edited at 2012-08-06 16:12 GMT]


Direct link Reply with quote
 

JH Trads  Identity Verified
United States
Local time: 10:54
Member (2007)
English to French
+ ...
Yes Aug 6, 2012

.

Direct link Reply with quote
 

564354352  Identity Verified
Denmark
Local time: 17:54
Danish to English
+ ...
Yes, always Aug 7, 2012

I fully understand that translation agencies need to verify the skills of potential new translators, but that is their problem and they should pay for this - any test translation should be paid for as they represent time/work that the agency is requesting to be done. That should mean that agencies would need to check any material sent to them by potential translators properly before they ask for test translations randomly from any number of people.

Portfolio translations, in my opinion, are very indicative of translators' skills. As Tony says, many portfolios will include rubbish translations, so that is indicative of poorly qualified translators, hence translation agencies simply need to peruse the portfolios and then they won't need to bother to contact such translators.

I have also been 'at the other end', having to assess the work of many potential translators based on translations of their own choice. Only very few have passed, simply because I have presumed that what they chose to present would be their best work, in which case I would not be willing to accept spelling or grammatical mistakes, let alone badly phrased sentences.

I think we need to shift the focus from us as translators responding to the beg and call of lazy/busy agencies (project managers) who are not willing to do their own research before contacting heaps of potential translators asking them to submit all kinds of details, including test translations, references etc. and then often not even bothering to respond to our emails once we submit to their requests - and direct focus at us as professional translators who offer a service for which we should be paid, as any other service provider would expect to be.


Direct link Reply with quote
 

Philippe Etienne  Identity Verified
Spain
Local time: 17:54
Member
English to French
How much time would you spend to secure a new returning client? Aug 8, 2012

Whatever we may think of agencies' procedures to find translators, my own metric is the likelihood of them choosing me: the more likely, the more time I would spend on a service offer.
The agency background, my own history info about them if any, translators they work with (BB), rating (BB again), their communication tone, etc. will help in assessing whether the agency will consider my profile and rates.

Any test takes time, and depending on factors above, I will choose not to reply/reply but not do a free test if odds are against me, or consent to a free 300-word piece if I have chances.

For me, one hour is the max time I should spent to reply to an agency enquiry (custom service offer, any quotation and any test). One hour without earning anything has a cost, so I have to discriminate the "good" from the "less good". I would need 3 hours to do a 1000-word test, so this doesn't even enter the field of possibilities.

Philippe


Direct link Reply with quote
 

Rosa Paredes  Identity Verified
Canada
Local time: 08:54
Member
Spanish to English
+ ...
TOPIC STARTER
@Philippe Aug 8, 2012

Thank you for your reply,
I do share your criteria to decide whether or not I do a test.

Cheers,

Rosa


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Are long translation test a rip off?

Advanced search







PerfectIt consistency checker
Faster Checking, Greater Accuracy

PerfectIt helps deliver error-free documents. It improves consistency, ensures quality and helps to enforce style guides. It’s a powerful tool for pro users, and comes with the assurance of a 30-day money back guarantee.

More info »
Wordfast Pro
Translation Memory Software for Any Platform

Exclusive discount for ProZ.com users! Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search