How is translation getting cheaper?

It seems just about every bulk translation agency is claiming to have found a great new way to make translation cheaper, be it through the exploitation of the under-qualified and the under-skilled or through some new-fangled technological monstrosity the latest Zuckerberg-wannabe would like to inflict on the minions. Many of these ‘technology pioneers’ send me emails requesting I ‘sign up’ for the latest crowdsourcing ‘innovation’ (they all look pretty similar to me) or agree to post a review, guest post or advertisement for their latest MT development on my site. They never receive a positive response.

The ‘quality levels’ farce

The ‘bulk’ translation market is incessant in its claims of being able to be offer translation of varying degrees of quality for less. One recent trend is the offering of various ‘quality levels’, something professional translators cannot and will not do. For us there is only one quality level: professional, publication-ready quality. We’re not really able to untrain our minds to just go with the first thing that hits us and skimp on research, just like you can’t untrain an expensive, high-quality mechanic from carrying out a careful repair. We can’t suddenly make ourselves less skilled and therefore cheaper and faster. We have to be ready to stand by the quality of our translation as individuals and have no other translators to hide behind in the event of a customer complaint, so we aim to make every job a representation of the best we are able to offer.

The real tell in this translation service at different quality levels is that one can often pay incremental amounts to have an increasing number of under-charging, usually under-qualified and under-skilled, translators edit the translation. This is like saying if you hire one twelve-year-old boy to write a press release, then get this press release proofread by his fourteen-year-old sister and finally signed off by his fifteen-year-old cousin, the result will be of increasing quality and at its highest levels, high quality. True, proofreading and editing by another translator are services commonly offered by boutique agencies and freelancers, too, but this is to provide top-notch quality, not an attempt to provide something vaguely passable by outsourcing to the under-charging, under-qualified or under-skilled in an attempt to skirt the costs of real translation.

The EN 15308 farce

Another common myth is that EN 15308 certification (or any certification, for that matter) guarantees quality, and that it can come cheap.

One agency I had not yet cooperated with recently requested that I fill in an updated version of their application form and tick a box to confirm that I would complete translations according to the so-called “Vier-Augen-Prinzip”, or “two-man rule”, meaning that all translations I produce for them would be proofread by a second party at my end.

This standard was put together under consultation with a number of large translation agencies who desired a means of communicating quality to their clients. The idea presented to the client is that this will ensure greater quality because the second set of eyes will spot anything overlooked by the first translator. In theory, this should be a second similarly qualified translator with relevant knowledge of the field, because the second translator should be sufficiently skilled so as to be able to retranslate the given document in a worst-case scenario. You can imagine my surprise when I asked for more details, only to be ‘assured’ that “the certification only requires a second person to proofread the translation” and “This person doesn’t necessarily have to be a translator herself”, since this goes against all my interpretations of quality assurance and likely the letter of the certification they hope to obtain.

Even more shockingly, when I explained I would then need to raise my stated word rates to cover the cost of the proofreader, including a rate for 100% identical segments that have not been locked (i.e. specifically marked as not for translation), I was reminded that this person “doesn’t necessarily have to be a translator herself”, and “Therefore you (I) don’t need to worry about the word prices.”

So this agency expected me to tick a box to skirt them of their responsibility in ensuring translations were appropriately checked in line with their desired EN 15308 certification. Further, they expected me to find someone who is at least bilingual in German and English, native in English, who has time to spare to read my translations for free. Just how many out-of-work, bilingual, German-speaking English native speakers with time on their hands do you know? An absolute farce and a true damnation of the supposed value of the EN 15308 certification standard, should this agency obtain it. I pity any genuine boutique agency that put real effort into implementing the principles behind this standard.

Multiple Personality Disorder (a.k.a. Lingo24’s “Coach” project)

The latest in a long line of brainy ideas is “Coach”, from Lingo24, for which Lingo24’s CEO is currently trying to get some venture capital together. The idea is that this technology could be used to “break down translation jobs into smaller component parts, allowing the high-level, high-cost work to be sectioned off, leaving the bulk of the routine work for less skilled (and so less expensive) translators.” So he believes the bulk of “routine work” (is our work EVER routine?) can be done by the “less skilled (and so less expensive) translators” – which, given the rates paid by Lingo24, we can somewhat accurately suppose will be the under-skilled, under-experienced, and most importantly under-charging translators who are the sustenance of this bulk translation market. The funny part is that he expects real translators (you know, the ones with experience, degrees, professional memberships, further training, and that other thing… ah, yes, skill!) will do these siphoned-off “high-level” parts. The “premium” translators (who, nevertheless, are happy to work for Lingo24 and at their rates) will happily provide all the important terminology to create glossaries for the students and hobbyists that will handle the rest of the project.

As illustrated above, this model is doomed firstly because high-quality translators, or “premium” translators as Arno likes to call them, are impossible to find for the rates Lingo24 pays translators. So it hereby fails on a most basic level: an assumed resource is unavailable for the assumed cost.

Now we come to the next flaw: the fact that the assumed improvement in quality by using higher quality translators for the hard parts (only) breaks every known principle of consistency. I will try to explain some of the reasons in brief below, but experts are welcome to elaborate in the comments.

Firstly, it should be remembered that all translators translate differently. There is no such thing as a standard or “routine” translation where everything is identical – even highly standardised documents like marriage certificates may contain deviations or include additional information that requires the eye of an expert. The very best of translators have wildly different styles and it is nearly impossible to ensure this terminology and style is consistent. A minor example: The German word ‘Vertrag’ can be both ‘agreement’ and ‘contract’ in English, and there are some cases where the former is better, some cases where the latter is better, and some cases where both are okay provided they are used consistently. A non-expert, like the person tasked to decide which term should be used throughout, won’t know this. The only individuals in a position to guarantee consistent and correct terminology and style are specialist translators of a standard that puts them out of the price range of any of these bargain basement agencies in the first place, as explained above. Further, I doubt Lingo24, like their bulk translation contemporaries, would ever risk sending translations for review by translators of such a high standard: firstly because their proofreading rates will be considered “phenomenal”, and secondly because an expert will simply spot many more errors that a fellow layman would overlook, thereby further increasing proofreading, review and editing costs. We have therefore established that the assumed outcome, an increase in quality, is fundamentally unobtainable and even actively hindered by the methods proposed.

The take-home message for Lingo24’s would-be investors? Lingo24 may promise quality translation at low prices, but this does not mean it is possible. Speak to multiple translation industry experts, i.e. working translators and academics, before you decide to invest in a world very alien to those peering in from Silicon Valley.

Who says translation is getting cheaper?

Christian Arno, for one:

And thebigword, for another:

I put my hands up: rates DO look like they are going down. That is, if you are going by the dubious offers on certain ‘translation marketplaces’ and believe the likes of Christian Arno’s comments on the industry problems and thebigword’s sob story of missing funds for their sales team’s latest cross-Atlantic sales run (most eloquently explained by Kevin Lossner). You may also think it if you went by the offers sent to multiple translators at once by desperate agencies incessant in their search for that which is the ‘pot of gold at the end of the rainbow’ for project managers at translation agencies catering to the bulk translation market with “competitive” rates: the highly skilled translator with low rates. Note that they will never find that pot of gold, but will keep searching for a better ‘resource’ to exploit as they compete on price with similar ‘competitive’ (cheap) translation companies. As long as there are wannabe or desperate translators to undercut their existing suppliers, translation rates at this end of the market can continue to be dug deeper and deeper into the cellar (until eventually the flood comes rushing in).

Vulnerability of the bulk translation market

This lower price segment strikes me as very vulnerable. If you compete mostly on price, you leave yourself wide open to competitors who can beat you on price through efficiencies you are not in a position to exploit (e.g. tax ‘optimisation’, cheaper ‘resources’), as well as the ever-looming threat of slightly-less-crap machine translation. Someone will always find a way to be cheaper or beat your level of exposure to potential clients. If your business depends on their business, your business model is equally unsustainable. So to any translators still battling in this market, my advice is to up your standards, up your rates and up your quality – or be ready to jump ship (or climb out of that cellar) before the plunge.

Rates are not going down

According to the BDÜ and SFT, rates are slightly increasing, albeit not necessarily in line with an increase in living costs. For example, Kevin Lossner published a study of the three-year trend in rates detailed in the BDÜ rates survey back in 2011. Couple this with the latest survey results, and you’ll see rates are still increasing. Similar evidence is to be found in the SFT survey results, which are available free online, although they are now a touch out of date. They do nevertheless indicate that rates are largely standing still if not increasing. If going by the statistics collected by the BDÜ and SFT, rates are most certainly not going down, and if anything, they’re going up.

You’ll also often see claims by agencies that plenty of professional translators will accept very low rates – and the truth is that while some people who call themselves translators may accept such rates, these people are not likely to be professionals. Refer to the BDÜ or SFT rate surveys above, as well as the ITI/IoL rates survey to find counter-evidence – professional translators do not work for the kind of rates thebigword presents as reasonable (0.038 GBP or less than 0.05 USD per word).

The disparity between what the bulk agencies think is reasonable and what professionals are actually charging is quite fascinating, isn’t it? It was such distinctions that made me finally see the value of professional associations (I admit that it took me far too long). In a world where the services of bilingual housewives and language students are such a hot commodity, it makes sense to distinguish oneself and exchange ideas with fellow professionals, not fellow ‘translation providers’ of varying quality. This includes the exchange of ideas about what is an appropriate rate to charge our clients.

Your rates are going UP… if you want them to

How else do I say this? My rates are going up. The rates of most of my translator friends are going up, at least in line with inflation. Why? We don’t accept low rates. It really is that simple.
Next time you’re encouraged to bid low or lower your rates, remind yourself of the following:

  •  Underprice yourself and you will be undervalued. Where there is information asymmetry, the buyer cannot help but make an uninformed choice (and risk buying a ‘lemon’). If the buyer chooses you, make it for the right reasons – perceived quality and professional contact. After the first job, if satisfied with the quality, they will have no reason to go elsewhere.
  • Assuming you get that client, how long do you hope to keep this client? A long time, right? Then you must choose a rate that will keep you happy for the short term and content in the medium term. Long term, you should be able to raise your rates. If the client chose you because you were cheap in the first place, it’s unlikely they will react kindly to a rate increase, even one that is in line with inflation.
  • Every time you submit to low rates you are lining the pockets of those who seek only to undermine your very profession, and in so doing, you provide false confirmation that yes, quality does come cheap. Not only that, but such behaviour may be frowned upon by colleagues who will at best consider you a bit unwise, and at worst a desperate, poor-quality translator who has let the side down. It’s just how it is.

In this industry you are limited only by what you ask for. If you ask for a low rate, that is what you will get. If you ask for a fair rate, you have at least a chance of getting it and filtering out those who will undervalue your services.

In my experience, my highest-paying clients, both agencies and direct clients, are the most pleasant, most prompt in their payment practices (usually seven to fourteen days), and most respectful of my work. They also seem to like to refer me to their friends when they are in need of translators.

Be under no illusions

The bulk translation market served by those who offer different quality levels and various quality assurances that at the end of the day mean nothing, but all for a reasonable cost, is a completely different market. These things simply don’t coexist because the providers of high-quality translation will not work for those rates.

Quality translation will always have a cost that reflects the value provided to clients, and there will always be clients who recognise this and will gladly pay dearly for our services. It’s all about finding (and rescuing) them.