Fred Brooks’ book, “The Mythical Man Month” (buy it now at Amazon), so famous within the software engineering world that it makes your eyes sting, is applicable to almost every field of human endeavour.

I read the book over 25 years ago so you might think that the fundamentals of software engineering haven’t changed all that much. Well, I’m not sure about that, but I do think that our fundamental approach to projects (whatever project in whatever field) hasn’t changed.

We received a call on Tuesday: 150,000 words had to be translated from Italian into English by Friday morning.  To be precise, the call was on Tuesday afternoon and delivery was required Friday morning first thing.

72 hours (not even) to translate 150,000 words is roughly 2,000 words an hour (most translators do 2,000 words a day and don’t tend to work 24 hours a day). And that’s not counting any revision or quality checks. In order to hit the delivery, a veritable troupe of translators was created and set to work. [Not sure if there is a collective noun for translators but although the alliteration is ok, maybe a "symposium of translators" would be better...thoughts on a postcard to...]

Even with the best will in the world, a rush job is a rush job. Quality kinda goes out the window. And with so many translators involved, consistency kinda goes out the window. The only positive, I suppose, is that the price kinda goes out the window too – the client will pay whatever it takes.

If our translation were a piece of software it would be so horribly bug-ridden that it would be unusable; in fact it probably wouldn’t even run. But bugs in software are easy to see. “Bugs” in a translation are quite different – that’s why we do proof-reading and quality checks. That’s why simply applying more resources, as Mr Brooks says, isn’t the answer.

Whichever way I look at this job, I cannot see how we don’t end up snookered. If we had turned the job down, the client would simply have found another translation agency to say “yes”. And we might have lost the client forever. It’s certainly a risk. But in doing the job, despite every caveat, the client might still complain about the translation quality and might decide to choose another agency anyway.

But what really keeps me awake at night is the client. It must have taken weeks if not months to produce the original document. 150,000 words is a pretty chunky print job. If I had been the author of that document, I wonder how I’d feel if my company valued the translation at three days’ worth.

Now before you start replying with “yes but” -  sometimes emergencies happen, someone can make a mistake – let’s think about the consequences. What if the document is a company prospectus and is being used to seek investments for millions of pounds. How much would you invest, knowing that the 200-page English document you’re reading was translated in three days by 30 different translators?

Fortunately, this wasn’t this case, but it makes you wonder, doesn’t it…exactly where will that translation go? and exactly how will it be used?

We recently conducted a mini-survey of some clients. Here are two results which I thought were very interesting.

Many large companies consolidate their translation needs in a centralised department. Through this consolidation and rationalisation, the company seeks to reduce costs. Every translation request goes through this single department which also manages the external relationships with Language Service Providers.

These centralised departments essentially become an internal supplier of translation services to their internal customers (overseas offices and departments).

We asked them which factor was most important to them.

46% rated the cost of the translation as the most important factor. If you add “On-time delivery”, then 60% of their focus is on project management: cost and time.

Centralised translation service: internal supplier

Then we asked their internal customers – those overseas offices which actually make use of the translation – what they consider the most important factor.

Translation consumer

The project management aspects are not nearly so important (totalling only 31%). Far more important was that the translation was accurate and well written (69%).

When creating a centralised department, therefore, a company explicity announces the tension between money and quality. Implicitly, it is pitting the internal supplier against the internal consumer who have demands which appear to be in conflict.

The demands of the consumer meet the constraints of the supplier

This tension is important and it would be naive to think that it didn’t exist. But it can only benefit the company if the system is truly in tension. That is, if both “money” and “quality” are exerting pressure.

If the centralised department succeeds in reducing costs to the point where the translations are poor quality, has anyone really won? Remember that a centralised department is also a monopoly and if this were to exist in the free market, there would be a regulator. So the company needs to provide the internal consumer with a method to send feedback to the internal supplier. And this feedback needs to form a real and important part of the overall system. It can’t be simply a “nice to have”.

A feedback mechanism, where data is collated, organised and reported, can highlight problems, opportunities and successes. A centralised department can use this data to reduce costs, improve quality, increase consistency and shorten project duration.

Only through a robust quality feedback loop can you keep the entire system under tension and only then will you see both lower translation costs and better quality. You can have your cake and eat it: you can improve project management whilst improving quality.

We know it can be done because we’ve provided consulting to companies on exactly how to do it: how to install a translation quality feedback mechanism that works for both your internal translation consumers and your centralised department.

If you’d like to know how to organise a centralised translation department with effective quality feedback, email us at info@intrawelt.co.uk and ask for our free white paper.

Very interesting couple of days today. Obviously I can’t mention names or give away too many details but it’s been an eye-opener. Maybe I’m just a little naive – you decide.

A very eminent person contacted us. Not a fabulously wealthy person but one whose name reaches places other names don’t. If I were to tell you who, you’d say, “what, HIM?” And I’d reply, “Yes, HIM”. With a rather self-satisfied air because I had actually come into contact with him.

Anyway, you get the picture. Someone eminent.

He needed a translation done. No urgency but he was going to publish it and, since it would carry HIS name, the quality had to be top, top notch.

We gave our best quote and emphasised that our translator would be one of the best we have, an expert in this field. Our reviewer, similarly, would be another of our top linguists in this field. And we’d do a final, double, double-double quality check in-house to make sure it was super top, top notch quality.

We didn’t get the order.

Why not? We called HIM back to ask.

Despite our emphasis on quality, which was simply an echo of the customer’s needs, we were asking more than the competition. The client, this eminent man, told us he went with a company that was asking 7p per word. The translation was from German to English. The company that won the order allegedly will do a translation, review and quality check. And presumably they’ll make a little profit too. All for 7p per word.

If someone tries to sell you a new BMW for a couple of grand, the alarm bells start going, don’t they? What’s wrong with it? Is it nicked? Whatever it is, you know something’s not quite right.

If Del-boy Trotter tries to sell you a “genuine” Rolex from his rather tatty suitcase “down the market”, you know something’s not right, right?

So when a translation agency promises a high-quality medical translation from German to English, with a professional review by another German-English linguist, followed by an in-house double check, all for 7p per word…

Well, you get the picture.

An experienced, qualified, professional German-English translator – a freelancer, without translation agency overheads and without a reviewer / quality check – will cost a minimum of 6p per word. Good reviewers could ask one third of that (sometimes more) – another 2p. That’s 8p per word without the agency costs let alone any profit for the agency.

Something just doesn’t seem right.

I like peanut butter. Do you know how many different brands and types of peanut butter there are? (I stopped counting at 74). But I like just peanuts in my peanut butter – no added sugar nor salt – and I feel better buying the organically grown stuff too. Know how big my choice is now? A lot smaller.

When it comes to translation services, you’ve got another massive choice. And if you don’t speak German how can you judge the translation? How can you “filter out” translation agencies or freelance translators?

It’s not easy but consider an average UK-based freelance professional translator (sources/assumptions are listed at the end). A careful and thorough experienced professional can translate about 2,000 words a day or 10,000 words a week. If the translator works 48 weeks a year (well, they must take a holiday sometime!), they translate 480,000 words a year.

The entry-level translator salary is £22,000 a year: 4.5p per word.
Experienced translators can earn £35,000 a year: 7p per word.
And senior translators could reach £50,000 a year: 10p per word.

But just as every author has their work reviewed, so every translation should be checked. “Proof-reading” really means “copy-editing and proof-reading” because we compare the translated text with the original and check the translation. Rates tend to range from a third to a half of the translation cost (although many charge by the hour).

So the next time you need a translation done, consider a senior translator will charge 10p word, then the reviewer will charge 3p-5p per word.

And if you have a large project, or many language combinations, factor in the Project Management costs too.

If you’ve just Googled “English German translation” and agreed 3p a word, are you buying peanut butter made from just organic peanuts or are you buying peanut butter made with added dextrose, palm oil, sugar, salt, colouring, flavouring and preservative?

Which would you rather put into your system?