A translation memory is a database of all your previously translated content.
It enables you to recycle content that you’ve paid to translate so that you only pay to translate the same string once. Over time, this can reduce the cost of translation.
Suppose the new product you’re launching on your eCommerce site needs translation. The repeated fields that that you've already translated (such as “size” or “color” or... "colour") are automatically identified from the translation memory and are applied to the target field. Not only will this reduce your costs, but it also increases the velocity of translation because the match is applied immediately.
There is an industry standard file format for translation memory (.tmx) that can be imported to Smartling at any time so that you can re-use translations that you have alrady created.
While translation memory is commonpalce in the industry (i.e. every language service provider will offer a translation memory), there are key differences in how translation memory is managed and leveraged that impacts your bottom line.
Learn how Eurail.com saved 70% in costs by leveraging translation memory. Read Eurail.com's story.
Each time a new string is translated, the source and target text are paired together and added to the translation memory. When future content containing the same set of words is submitted for translation, Smartling intuitively matches strings with previous translations instantaneously.
There are three ways a string can match with entries in your translation memory.
SmartMatch [a.k.a ICE Match!] | 100% Match | Fuzzy Match |
---|---|---|
A string that perfectly matches a string in the translation memory, including any tags, placeholders, etc. | When the text of a string matches the text of a string in the translation memory. It's considered a 100% match even if there are discrepancies with tags, placeholders, key values, etc. (not to be confused with a SmartMatch!) | Word matches that are less than 100% identical between the translation memory and source content |
SmartMatch compares new strings against existing translations in your leverage configuration to automatically apply translations to strings you've translated before. | Project managers can configure workflows so that 100% matches skip steps | The value represents the percentage to which they match (typically between 50-99%) |
No matter the match, translators and reviewers can review the translation memory database in the CAT tool, and choose to edit them instead of starting from scratch.
Product Manager Tobias Kahn walks you through the most important economic tool for translation. Watch out in-depth webinar for Translation Memory
Learn how Hootsuite reduced its annual translation expense by 33% since becoming a Smartling customer. Read Hootsuite's story.
You paid for it, so you own it. You can export your translation memory on demand 24/7/365 from the dashboard. All strings attached. No red tape.
OpenTable’s localization team has a lot say about Smartling’s SmartMatch technology: "The SmartMatch technology enables us to re-use or adapt translations we've already had done, where they exist – a huge time- and money-saver.”
Read more reviews Check out our reviews on G2 Crowd!
Want to learn more about how Smartling’s Translation Memory could directly impact your bottom line? Let’s chat!