Posted by: patenttranslator | July 21, 2017

The LSP Model and the Anti-LSP Model

Despite two decades of ferocious attacks by the translation industry on our profession, there are many of us who do not subscribe to the corporate mass-production model laid out for us by the translation industry. We refuse to work for it and instead are doing our own thing, independently of and differently from LSPs (also known as Language Service Providers, or Lame Service Providers).

I am going to call the non-LSP model the anti-LSP model and practitioners of this model anti-LSPers in my post today.

I don’t think it’s even debatable that most of us anti-LSPers are much better off, financially, emotionally and otherwise, than translators who obediently serve the LSP model and long ago gave up on the dream that is at the heart of every small business: the dream of becoming an independent, professional specialist who is free to pick his own customers and to determine rates and conditions under which he is willing to work, instead of capitulating to the translation industry and just doing as he’s told.

The financial aspects of what we do for a living, although they are not everything, are kind of important in this world too, wouldn’t you agree? As Woody Allen put it, “Money is better than poverty, if only for financial reasons.”

Based on what I hear, the anti-LSPers among us are doing quite well. Not all of the time of course, but more than just some of the time.

I have friends who translate various languages and subjects, who live in different countries and who almost never (or never) work for corporate translation agencies, or for the LSP model.

When I hear from translators who are working for direct clients, I mostly hear them complain in person or on social media that they are too busy and find it difficult to keep up with the demand. I certainly have this problem myself as I am currently forcing myself to work seven days a week and hardly have time for anything else.

When translators who work for the corporate translation industry’s LSP model are complaining, sometimes I do hear them complaining about too much work too, but mostly they are commiserating with each other about low rates, unreasonable demands on their time, and translation agencies (LSPs) disappearing without payment, only to appear later under a different name.

From a historical perspective, the LSP model is quite recent. It was created as a result of technology and globalization about two decades ago, while the translator-direct client system has been around for millennia.

What I am today calling in my silly post the anti-LSP model in fact has a very long tradition behind it.  St. Jerome for example, the patron saint of our profession when it was still considered a real profession, (before the LSP models started calling us translators “vendors”), worked as a freelance translator 16 centuries ago for approximately 30 years on his translation of most of the Bible from several languages into Latin for only one direct client, namely the Pope.

As there was no internet in St. Jerome’s times (not even fax machines!), he had to deliver his translations to the Pope in person, although it is possible that he had friends who were helping him with timely delivery, unless he saw the walking as a welcome break from his sedentary and solitary profession, which is likely, I think.

Did St. Jerome work for a medieval intermediary, an equivalent of a modern corporate LSP? I don’t know the history of the Middle Ages that well, but I sincerely doubt it. Most translators probably worked directly for their clients in the dark and backward Middle Ages when money was not yet speech and corporations were not yet persons … when in fact there were no corporations yet.

And we still have anti-LSPs in our enlightened modern times, although the LSP model is also very common, and some translators may even think that it’s the only game in town.

So how does the modern LSP model work, and how is it different from historical and current anti-LSP models? The LSP model is a system that is quite efficient at what it does – namely creating large numbers of translated words in a relatively short period of time based on a strict division of labor.

This is how the LSP model works:

  1. Owners of the LSP system are almost always thoroughly, proudly and defiantly monolingual people (especially in English speaking countries), who specialize in the fine science of buying low and selling high, or at least in buying as low as possible and selling a little higher. The great majority of the LSP owners actually don’t understand anything about “translation” per se and they therefore logically see very little difference between the division of labor at a McDonald’s franchise, for example, and at the new type of translation agency called an LSP – let’s not forget that large LSPs are also franchised now, just like fast food restaurants.
  1. Employees or freelancers working for the owners of the LSP model, usually called project managers, who sometimes know something about foreign languages, but usually very little and almost always absolutely nothing about the languages they are handling. In their defense, this is really not their fault because they are asked by their monolingual employers to handle any language, something that not even St. Jerome himself could do. These workers could also be described in the pyramid of workers supporting the LSP model as managers or overseers of the actual workers, who are called:
  1. Translators, namely the people who, although they are in fact generating all of the profit for the owners of the LSP model, and thus also paying the mostly meager salaries of managers or overseers employed in the LSP model, are at the very bottom of the system and typically have absolutely no say about anything.

The image of plantation slaves working on cotton fields somehow always comes to mind when I think about the role of translators in the LSP system.

Although a system with this kind of division of labor and separation and segregation of work and workers makes sense up to a point in the environment of a fast food restaurant, the system clearly does not work in the context of translation business.

The LSP systems works quite well when it comes to generating large amounts of texts with many words in a short period of time.

But the Achilles heel of the system is that it is designed in such a way that the people running the systems (LSP owners and managers employed by them), can’t tell whether the large amounts of texts with the many words generated by the worker bees called translators are translated accurately, reasonably well, or mostly incorrectly.

The system cannot work as well as a fast food franchise because unlike flipping of burgers at a fast food chain, translation represents a highly knowledge-intensive labor that cannot be divided into a few simple, interchangeable steps (such as making fries and flipping burgers), and then distributed efficiently among several or many far-flung translators toiling in respective countries on different continents.

Although this feature is precisely what the translation industry is very proud of and likes to boast about on LSP websites, I think that Pope Damasus II would have clearly recognized even 16 centuries ago that this would be an unrealistic way to go about to get long translations done when quality is paramount, as was the case with the first translation of Bible scriptures into Latin.

Why else would the Pope give St. Jerome 30 years to finish one job, when he could probably have found five St. Jeromes in and around Rome, parceled the job out to them and then have had the Bible translation done in six years, especially since what is now called the Bible was written in several different languages?

Another reason (in addition to the need for speed to keep customers happy) why the LSP system does not work very well is that translators are completely powerless in a system in which all decisions are made by people at the top, i.e. owners and managers who are not translators and who often know very little or nothing about translation.

Whether the creators of the LSP model realize it or not, the model they created, with its emphasis on maximum speed and the lowest possible cost of a production unit called a word unfailingly keeps generating lower and lower quality final products, frequently resulting in what is referred to with the technical term “total crap” or by the acronym “FUBAR”.

The final product created by this system does consist of production units called “words”, but the final product that is actually desired by the customer is not the same thing as the maximum number of words that can be created in record time at the lowest cost of the production unit.

Based on the LSP model, a larger, possibly somewhat urgent project – although who knows whether the project really is urgent, or whether the urgency is the result of a somewhat capricious decision of a mid-level manager in a widget-making factory – is typically organized by an LSP coordinator in the following fashion:

An LSP model project manager who needs to have several long documents translated on a very short deadline, (which is often completely unrealistic), dips into a database of translators conveniently available for example on the Proz or the ATA website and fires off an impersonal email to a dozen or more translators, such as the following email that I received a few days ago, as I am listed in the ATA translator database.

“Dear Translators

 Would you please confirm how many words you could take on from the attached?

Word count: see file overview

Deadline: 10 am, 2 days from now.

Thank you!

 Natasha P., Just Another LSP, United Kingdom.”

Natasha included for my convenience several files for translation in four attachments in PDF and Excel files. I took a quick look at the files and saw that there were many thousands of words to be translated per each file. I should add that I received the email one day before a major holiday in the United States.

I wonder whether Natasha realized that 10 AM in the UK is 4 AM on the East Coast and  1 AM on the West Coast of the American continent, which means that each of the translators that she “reached out to” (that’s how project managers in the LSP model like to put it because it sounds so dramatic!) in the United States would in fact only have about one day to translate thousands upon thousands of words during a major holiday when most people naturally want to be with their families.

Maybe not, because why would she set such an impossible deadline for delivery? Or maybe she simply not give a damn, which seems to be more likely.

But this is the way the LSP model works. Little details like this are completely irrelevant in the grand scheme of the LSP model. And if Natasha sent the same email to Japanese translators (this was a translation of medical documents from Japanese to English) who live in Europe, the United States, as well as in Japan, Canada and Australia too, how could poor Natasha possibly know which country has a holiday on any given day?

Did Natasha get the job done, I wonder? She probably did. There are so many translating worker bees in this world and they all need work, so they’ll just grab as big a piece of a job as they think they can handle and these worker bees in the end will get the job done, by hook or crook.

As I said already, the LSP model is very good at generating large numbers of words in a record period of time.

I used to be happy grabbing as many thousands of words as I could handle when times were slow while working for the LSP model myself for many years. There were so many bills to pay, I felt I had no choice.

Fortunately, eventually I did work my way up to being a practitioner of the anti-LSP model, and although this model also has its disadvantages, I am much, much happier as an anti-LSPer, and not only because I make quite a bit more money than the worker bees trapped in the LSP model.

I believe that now is a very good time for other translators to enter the translation market as anti-LSPers simply because the LSP model creates so much crap that direct clients are actively looking for alternatives to the FUBAR products that are so typical of the LSP model.

Although I did not realize it or give a name to it then, the anti-LSP model was the concept on which I instinctively based my small translation enterprise by trying to identify my own direct clients, which in my case are mostly patent law firms, starting about 20 years ago.

The fact that after so many years this little anti-LSPer is still here and busier than ever, while completely bypassing the LSP model, indicates to me that the anti-LSP model is based on a sound concept. I also think that, unlike the LSP model which aims to maximize profit at all cost by generating as many words in the shortest possible time, the anti-LSP model is applicable not just to patent translation, but to just about any field of specialized knowledge where expert translations of highly experienced translators are sought after, frequently required and reimbursed in accordance with their perceived value.

It is so much more fun working for the anti-LSP model, which is to say working for yourself. You still get to translate, and you can get much better rates than what you can hope to receive from the modern LSP model.

Are you happy in the LSP model?

If not, may I suggest that you concentrate on creating your own version of your own anti-LSP model that will work better for you?

Are your direct clients going to find you and are you going to find them?

I certainly hope so.

Good luck to you.


  1. Great post, thank you. So true in so many ways.


  2. So, you would definitely still recommend newcomers to enter the profession?


  3. Yes, if it is the right kind of newcomer.

    But only if he has a good, realistic plan to become independent of the LSP model.

    Otherwise, probably not.

    Liked by 1 person

  4. Some very interesting points there – and thanks so much for the Stromei video. I love all his stuff.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: