Are you one of those people that give their all when they create content — only to realize that what you have been working on just does not transfer to other languages that easily? Does it always take too long for your copy to be published in all languages you need? And are you wondering what to do to solve that? Then this article is for you.
How to solve this?
The answer is actually quite simple: do not wait until your content is ‘done’ before you start translating, but do a “shift left” and include your translators in the process of creating the content. Give them the same brief as you do the copywriter and ask them to translate all versions the copywriter delivers.
The translators will get a close look at your writing process. From the changes you make they can see what really matters to you and they will understand much better what your content is meant to convey. And of course, the translation will be done much faster, almost at the same time as your original copy.
But… what about the ROI?
If things were only this easy. Often, translating earlier versions is seen as a time-consuming, expensive addition to translating after the content is done.
So how can you actually prove that this really leads to more value and faster publishing? How do you prove the ROI of this collaborative model? And how do you set this up in the first place? For this we need to take two more things into account.
The issue with translations
Translation is usually organized in a waterfall process. Induced by the current business model of translation companies (LSPs), content must first be ready and consolidated in order for a translator to start working. But then translating begins, and the trouble starts. There is no way up the waterfall, so the translators have to deal with the source content they get.
Lots of ambiguous sentences, untranslatable idioms and sentences that become too long in the translated version are issues that translators have learnt to cope with, but that reduce their productivity severely and may impact the quality of the end result — time, costs and quality of the translations are all affected.
Seeing the value of translators
Because of the current translation ecosystem, a lot of value that translators can bring is not used. Translators can be seen as the first users of your content — and hardcore users. They will notice issues that the authors have not thought of when writing or editing, simply because translators dive into the matter more thoroughly than any reader. In current translation ecosystems, there is usually no efficient infrastructure for translators to express their valuable feedback and it is often seen as ‘complicating things and disturbing’.
However, if you include translators earlier in the authoring process, and you let the translators collaborate with the author, this will not only improve the quality of your translations, but will also impact the quality of your source content.
First publish, then translate, then edit
For this to work well, the authoring process needs to be adjusted: to a more collaborative, iterative process that allows all contributors to give feedback on publishable, even though not finished versions.
Some time ago I wrote an article called “Painless authoring”. In this article I explain how a paradigm shift towards taking a more agile and collaborative approach on authoring leads to a less painful process with better outcome. In this article you will see that if you add translation to the equation, these benefits are multiplied.
The most important mind shift you need, is that you actually “publish” early versions. Publishing can be thought of in many different ways, obviously, it doesn’t necessarily have to be to a public website. And more importantly: publishing is not the end of your process. It’s the beginning.
Now you can do two things with these early publications.
- ask peers for feedback, so you don’t get caught by surprise when you think your content is all done
- ask your translators to translate AND give feedback and ideas on anything and everything: is the structure clear, is the wording accurate and clear for readers, etc.
The nice thing is that translators are usually quite good with language and content and they have seen A LOT of copy. Imagine having this whole army of linguists at your disposal that are now only asked to translate, but that can actually help you make your content better in the very early stages.
On the basis of the feedback the content gets adjusted and changes get implemented in all languages. Obviously, the time to market for these languages will be much faster and the translated copy will be better, as the translators will be engaged in the content already and have seen the process. Therefore, they will know better what is important to the copy writer. Also, if you empower your translators and see their value, they will be more content and feel more involved and committed. This is key to a good translation.
But also, what will happen is that your copy writer will be much faster. They do not have to make the content perfect before they publish, and there will be so much feedback about what MUST be changed, and not what COULD be changed. This reduces a lot of rewriting that you could basically consider waste and gives a writer so much more confidence.
Of course, if you’re really bold you already publish content for readers to give feedback on and you measure if the goal has been achieved. All this to make sure you “maximize the work not done”.
How do you know?
An important question is: how do you know if this actually works? This is where it becomes interesting. I use the Lean methodology of continuous improvement to know what to solve, and whether the solutions are working. The key to this is that you measure your process, and not just your output. In a nutshell, this is how it works.
There are a few metrics you can use to evaluate your process. One of these is velocity (aka speed). To see if your process improves, you must measure your baseline first. Having this metric is not a given, as often the processes of creating content and translating it are measured separately, if at all. So first thing you do is measure the time from starting the creation of copy until you have published in all languages. Then you check where most time is wasted, what the root cause is of and you solve these issues one by one.
The second is of course quality of your copy, or how I would rather call it, effectivity. This you can measure by making sure you set a measurable goal for your content and comparing the different languages. Again, you define the bottleneck, research the root cause of the differences between the languages, and come up with a solution.
In my experience, the root causes of the bottlenecks you run into are usually the lack of collaboration and the waterfall approach. If you apply the solution described above, it is very important that you measure the effect on these process metrics, and also define what it is worth to you to increase the velocity or effectivity.
This will tell you what the ROI of these improvements are. And will make it easier for you to decide if the increase in costs, if there are any, are leading to more revenue, because your time to market is faster or your effectivity is higher.
This experimenting method requires working in single item flow, rather than in batches. This way of working is key to process improvement. I will explain more about this in a future article. If you would like to know more already, please do not hesitate to comment or contact me.
What do you need?
Does it sound easier said than done? Just give it a shot and see for yourself.
Here are a few things you might find you need.
- An agile mindset: you are not afraid to show your preliminary work to the world
- Direct contact with your translators
- Trust in the capabilities of your translators
- A way to publish your content to the group of your choice
- Tooling so translators will understand what has been changed. Usually a translation memory is used for this, but this is not necessarily the best option.
Tell me how it went!
If you are inspired and would like to give it a go, I’m very curious to hear your results. Also, if you want to discuss this solution further, do not hesitate to connect.