Jul 22, 2010

The end of agencies?

I had an interesting conversation with an agency owner tonight. He described how representatives of various agencies in a certain non-European country have been making the rounds here lately with larger agencies and companies offering technical manual translations with two rounds of proofreading for about half of what I typically charge for translation alone and about a third of what this agency typically charges. The frustration in his voice was clear as he talked about one project after another with 20,000 page volumes that get sucked up by these rajahs of translation who have armies of highly skilled translators working night and day at their keyboards for about 80 euros per month. "Really?" I asked. "Where do they get the skills to do this?" "Over here," he said. Now I'll admit that I haven't spent much time on German college campuses lately, and I only occasionally visit German software and engineering companies in person these days, but when I have had occasion to do so I haven't noticed a lot of people from this magical land of Fearsome Cheap Translators acquiring the skills here that would be necessary to make sense of the average crappy source text that I see from a German patent lawyer, a Swabian automotive engineer or a Frisian wood-burning stove specialist. Perhaps my old eyes can't penetrate their clever disguises of body paint in European skin tones and those ever-so-clever Kaiser Wilhelm false moustaches. Perhaps I am just in deep denial; a spectre must indeed be haunting the halls of agencies as one after another talks of contracts and opportunities lost on a scale that I with my modest earnings can scarcely imagine.

Perhaps, I thought, it's time to go to law school. There is surely a great future for litigators when those Boeing repair manuals translated by moonlighting elephant trainers for one cent per word show their worth and tons of metal rain from the sky.

In the past few years I have heard from many cherished agency customers that for their customers price is all that matters and they must play the "emerging country outsourcing game" to survive. I'm sure that's why I haven't seen any translations for the Maybach line from Daimler in a while. Some guy in Laos is surely doing those German marketing texts better and faster.

Despite all I've heard and the well-grounded commercial reasoning to explain this "inevitable" trend, the words "live by the sword, die by the sword" keep coming to mind. Substitute price for sword. Feed the beast with Trados-inspired discount scales that bear no real relationship to real effort in most cases and it only gets hungrier. Fail to communicate what real value you may add and don't be surprised if it isn't appreciated. Deal with people who are unable to appreciate the qualities of languages they haven't mastered and at the same time fail to build relationships of trust, real partnerships, and you might as well be casting pearls you-know-where.

I asked if this Fabled Land also had a stable of superior translators for Czech. Apparently not. Yet. I suppose the Lithuanians are safe for a while too, though they should watch those Cambodian tourists carefully. They might be on a mission of linguistic espionage.

I think one of my friends from this country will be in town in the next week or two, and I'll have to ask him if I've missed something. If there is in fact a gold rush in his country I hope he gets a few good pans full at least.

But these apocalyptic tidings reminded me of another prediction I heard some months ago regarding the End of Agencies, and I think that there is indeed something to this particular forecast, at least for the long term. It would go nicely with trends I have observed with larger companies and insourcing of translation project management functions again.

With the growth of online networks and databases for translators, editors, occasionally project managers, it is becoming easier for companies of all sizes to make direct contact with service providers for the language and skill combinations they need. Portals like ProZ or Translators' Café have serious disadvantages, of course, because it is often hard to determine the reliability of those advertising services until one has tried them, and sometimes there simply isn't enough time available to do a thorough evaluation. However, the growth of quality-evaluated networks promises to change that perhaps. ProZ has made some fitful efforts in this direction, but at each stage the usually reasonable efforts of the site have been met with howls of protest from the trees. More promising, perhaps, for companies looking for reliable service are evaluated networks intended primarily for agencies, such as the Quality Translation Network from LSP.net. I've seen this system from the inside for several months now as I've worked with the company on documentation, localization and administration issues. I've used the evaluation system for service providers myself a few times and see how the ranking system works. And it occurred to me that an "end customer" with a regular need for translations might just as well join such a network with an inexpensive SaaS account and directly access quality-evaluated data on all the roles needed and build a team dynamically for specific projects. Take provider X for one role, provider Y for another and specific freelance editors or translators for other roles. Feedback from satisfied or dissatisfied customers on a project manager's performance in complex multi-lingual projects might avoid many expensive mistakes.

Far fetched? Not really. When it comes to top quality, often the agencies are competing for the same small pool of service providers. I don't get sent the same specialist text by more than one agency every week, but it's happened a few times over the years, and more than once I've heard from someone who is asked by agency Y to edit his translation done for agency X. And that in common language pairs. It really is a small world.

Give the premise of a small pool of truly top individual service providers in a free market, a reliable quality evaluation system is a potential gold mine for quality- and cost-conscious corporate translation buyers. Will such systems arise and evolve to the point that the translation agency as we know it will be as dead as the dodo in our lifetimes?


  1. Hi Kevin, I'm only replying to the first half of the post. In this "flat" world we translators have to compete with our counterparts from "cheaper" countries. Compared with European translators, Asian translators are just more "cost-effective".

    As an English-to-Traditional Chinese translator I have to compete with translators from Mainland China. They use Simplified Chinese, yes, but it's easy to convert it to Traditional Chinese using a function in Word. In Taiwan we see more and more Traditional Chinese documents containing Simplified Chinese word usage. This is often viewed as "invasion" among translators in Taiwan. But there seems no way to stop it because those from China are cheaper, and some clients care more about the price than the quality.

  2. Hi Kevin,

    Regarding your second point, the end of agencies, I believe there will rather be a transformation of those into online language testing centers, or they may include this feature as well into their repertoire of services...

  3. Hi Kevin,
    You made some very good points in your post. It's true that these days, it often times seems to only be about price when it comes to the translation of certain subjects in specific language combinations. Because the industry is so fragmented and there are new translation agencies popping up all the time, it can make the translation process more difficult for everyone.

  4. "I don't get sent the same specialist text by more than one agency every week, but it's happened a few times over the years, and more than once I've heard from someone who is asked by agency Y to edit his translation done for agency X."

    Yes, this does happen sometimes. Once, an agency canceled a test translation after we could not agree on a rate, and later another agency sent me this same test translation (completed by someone else) for evaluation.

    Thank you for your extensive post,


Notice to spammers: your locations are being traced and fed to the recreational target list for my new line of chemical weapon drones :-)