Localization firm SDL recently announced that Best Western has signed up for SDL’s “Knowledge-based Translation System” to better manage its global Web sites. According to the press release, the “implementation will involve translation of more than four million words on four thousand web sites plus daily updates into four languages: French, Spanish, Italian and German.”
Best Western wants to increase the amount of localized content on each hotel Web site, but doing so manually is simply too expensive. SDL’s solution involves using a mix of translation memory, machine translation and workfow software to keep translation costs and turnaround time low.
Both companies estimate that Best Western will save $2 million in translation costs within the first year of operation. Clearly, machine translation (MT) is playing a key role in the savings. But is machine translation really ready for prime time?
SDL thinks so, and I agree. The key to success, however, is workflow. You can’t use MT as a last step or an only step and you need to carefully control the type of content you expect MT to handle. But if you do it right, MT can help you expand the amount of localized content without blowing the budget. The main reason I believe MT will succeed is that companies today simply have much more content than budget, and this won’t chane anytime soon. MT can help companies get more localization bang for their buck, and this will translate into a real competitive advantage. However, the road to successful MT is long and many companies will fail along the way. It will be very interesting to see how Best Western does in the months ahead.
PS: Here is the current Best Western global gateway.
I like the globe icon, but this gateway will need to be improved to more effectively direct traffic to the localized sites. It needs to be moved to the top of the page and the pull-down menu needs rethinking. I recommend a splash gateway for starters. The global gateway may seem trivial, but it can make a big difference when it comes to traffic – the metric most often used to measure the success of local Web sites.