What makes a match? August 31, 2009
Posted by Jill (@bonnjill) in Business practices, Translation.trackback
The Medical Translation Blog has an excellent explanation of the difference between translation memories and terminology glossaries. If you haven’t seen it, be sure to check it out.
I particularly like the comment suggesting discounts for matches aren’t such a good idea. In fact, Trados used to suggest a 30/60/100 scale: 100% for anything below 85% matches (some agencies use lower percentages for matches, but in my experience anything below 85% essentially needs reworking and our full attention so translators should be paid accordingly), 60% of the full word price for 99%-85% matches and 30% of the full word price for 100% matches and repetitions. In an ideal world translators wouldn’t offer discounts for matches at all. After all, we are the ones who shelled out the money for our expensive TEnTs (translation environment tools, aka CAT tools). Why should agencies expect to be able to benefit from our business purchases? It’s not like they expect discounts because we have the right specialized dictionaries on our shelves…
I know plenty of translators who do not offer discounts – period. It’s up to each individual translator to decide whether or not it makes sense for them to offer discounts on matches. I have some clients who do not demand discounts based on Trados analyses and some who do. It makes more business sense to work for ones that do not, but I also work for agencies that do require Trados and discounts. It all depends on how busy I am when I get the request and whether they provide a TM or expect me to use mine (which is a whole other can of worms)…
BTW, I have no problems offering a discount to agencies that provide me with a fully licensed copy of their required TEnT. I don’t have to pay for it, so I have no problem passing on a discount to the agency. I have one agency that provides me with their TEnT and a year license. Once the license runs out I simply get a new license code from them. I wish more agencies did this.
Musings from an overworked translator 
TM tools make work so much more efficient (they do in my fields of specialisation, anyway) that I only need a very limited time to earn back my investment. Anything after that is profit, even if I do give discounts for matches.
An example: I’ve just finished a job with a total of almost 64,000 actual words in almost 16 hours. Using the discount percentages that I’ve agreed with this particular client, I’m being paid the equivalent of 14,000 words, which still makes 875 words per hour. The difference between the two is enough to buy every commonly used CAT tool there is… And I don’t see it as offering my client a 78% discount, but as being paid a fair amount for the work that I’ve done.
So in my opinion, discounts for matches do make sense in every possible way. The investment I need to make in hardware, software, internet etc. is reflected in my base rate, though. I keep a close eye on how much I earn per hour for each of my clients, and if that amount consistently drops below the minimum I’ve set for myself, I increase my word rate and/or look out for other, better paying clients…
That’s a good point, Ellen. I think we’ve all had jobs like that, but they are sometimes few and far between (at least in my field of specialization). However, it is very important to make sure your investments are reflected in your base rate.
I tend to look at it in terms of my hourly earnings. If I can offer a discount on TM matches and still make my average hourly income plus X, then I’ll be open to doing so.
How big that “X” is then becomes a matter of negotiation. My policy is that I still have to earn more than my average income without a TM, to compensate my expertise and investment.
The trick I think is to get a realistic idea of how much the TM is going to help you, and offer a discount (or not) accordingly. A lot of new translators in particular allow themselves to get talked into giving a huge discount, and then find themselves cursing the gods of translation late Sunday night, because the TM didn’t speed things up as much as they expected.
This is especially true when dealing with a TM you didn’t create. You should have a good idea of the quality of the TM itself before agreeing to a discount.
Those are excellent points, Ryan. I couldn’t agree with you more – especially about new translators being talked into horrendous discounts and regretting it.
Thanks for the link to our story on the Medical Translation Blog, Jill!
I’m sure that the decision around TM discounts depends on a service provider’s area of specialization. In our case, we work for the same medical device and pharmaceutical clients again and again, and we see the kinds of impact from TM leveraging that Ellen described in her comment.
In our environment, not offering clients a TM discount simply isn’t an option – otherwise we couldn’t be competitive. That means that all of our translators and editors have to accept the same (or similar) TM discounts – otherwise we would lose money hand over fist.
One interesting aspect to the TM discounts is >how much< of a discount should be factored in. For example, should "high fuzzy matches be discounted by 40, 50, 60%? Who decides and based on what? It all seems kind of arbitrary…