Twelve years ago, digital channels were in their infancy. Global paid search and display were used here and there, but most of the volume for traditional and digital marketing was driven by the core markets. This consisted of the US, the UK, France and Germany. Generally, operations weren’t so globally centralised such that one network could run everything individually. We started at a time when pretty much everybody used a very sporadic methodology to get their content done. When anybody needed content, they would ask around to see if someone even vaguely spoke the language they needed. One week it was German, the next it was French, and so on. There was no structured or consistent system in place.
When clients regularly began asking for multilingual content, big networks and global agencies did not have enough control over other offices, their staff and P&L to get content done internally. In addition, no one in the network was purely employed for linguistic expertise, but rather for technical and/or digital skills, which do not automatically translate to excellent writing skills in your native tongue. We realised that there wasn’t a structured process for this kind of thing in the industry at the time. You either hired translation vendors for one-off jobs or went about things with the ‘do me a favour’ policy that was asking someone you know in the office who speaks the language a bit.
We saw an opportunity to implement a quick and reliable system that would allow us to give clients what they wanted. It would allow for more consistent quality in terms of output compared to when translations came from people with varying levels of fluency in their respective languages, from the guy in IT one week and a native speaker in the office upstairs the next.
Limitations with Traditional Translation Agencies
The translation vendor solution that some networks used worked in as much as getting someone when you needed them, but the vendors didn’t necessarily understand the formats. Back then translation agencies had no idea what paid search, display, or paid social were.
Vendors would receive content like they would any other document, their focus being to give the agency back the best translation. They didn’t know about character limitations, nor did they know that Google doesn’t allow superlative. They weren’t aware that specific CTAs in your description lines increased CTR, and they didn’t know what to cut off with translations from English to languages such as German or Finnish, since their sentence structures are much longer and don’t fit word counts so easily.
All these limitations cost a lot and caused delays. Agencies would send something to translation agencies who would charge by word count which cost a lot of money, especially back then when keyword lists were so big. It was all about creating as many keywords as possible from the start, because broad match did not offer a search query report to automatically identify missed opportunities and negativize unwanted combinations. Hence, you had to focus more on phrase and exact match to maintain control as you could not fully control what broad match would generate for you. This meant millions of keywords, resulting in a high word count, thus a high per word rate cost. Unsurprisingly, this was totally inefficient – as many as 89% of keywords don’t even get an impression.
Essentially brands or their media partners ended up paying for something that got no traffic. The same applies to SEO — you don’t need that many keywords, it’s about research and quality over quantity. You should pay for an analyst’s time rather than a word count. It’s a problem which still exists today, changing the perception of localisation in terms of how you will charge for work. We decided to sort this out. We wanted to tell our clients that any market in any language is possible, even with budget constraints.
Creating a New Breed of Digital Localisation Specialists
Locaria was created to bring expert translators and digital marketers together. We started by growing our network of high-quality linguists, recruiting through a rigorous process and only accepting the best. Then, when it came to building their digital know-how, we started from scratch, training our linguists on all channels, formats, platforms, and metrics. It was a tough process, as translators back then were used to a certain way of doing things. Receiving content, translating it, sending it back and getting paid. Many translators focused on website localisation to get their word counts, but we needed them to do more. We wanted them to develop a passion for what we began calling ‘Performance Linguistics’.
When we tell a linguist that we’d like them to improve structure and test how they could increase the CTR with a different version, they know what CTR and CPC mean. A linguist’s job is really to focus on fine-tuning work for a particular purpose or objective, a specific target or KPI. That’s a difficult job. It requires an understanding of the whole range of methodologies available, how to make them work for each channel, and then the ability to model them out based on what works best. This is something we’ve learnt with our clients.
This new model of localisation for online content has driven tangible gains for all our clients. High levels of customer satisfaction create a fantastic work environment for our content analysts to innovate in, and highly structured training attracts the best linguists to our community.