Can we have a word?

The advantages are clear. Professional terminology management contributes to translation quality and increases the productivity of translators and everyone involved in the translation process – this makes it a factor in cost-effectiveness. Well-maintained terminology resources save time by minimising ambiguity. What’s more, they ensure that translators can more effectively use translation databases. They form the basis for terminological consistency, a key marker of a quality translation. The translators themselves save time, since they don’t have to research as much or ask the customer as many questions. So goes the theory.

With this article, we’d like to shine a light on our translation company’s practice-oriented terminology work. As it happens, terminology data is generally not ready for immediate use when we receive it.

Take, for example, our customer XY: a public-sector organisation that has its own translation division but also commissions external service providers. It maintains terminology sources in a variety of formats, chiefly Word, Excel and PDF files, which external translators can download from an online portal. There are currently more than 30 source files to take into account. Consulting all these individual files for information is understandably quite a hassle.

This is where I come in. You may remember our last newsletter, in which my responsibilities at exact! were presented as part of our “A look at the role of …” series. We’ll now take a closer look at the practical side of my work.

Ideally, terminology data from individual files should be compiled into a single, server-based Multiterm termbase. Our project managers can then include this termbase in all translation projects for the customer. Thanks to the automatic terminology recognition function in Studio, the translators then have direct access to all terminology information in real time when working on a translation.

The first step when creating a database like this is to review the sources and decide whether to simply import the data directly or manually enter it in the termbase. We usually create an initial inventory by importing Excel files, then subsequently update the termbase manually to avoid duplicate entries.

When preparing the source data, it is important to adhere to a few basic specifications so that entries are as precise as possible and can be used effectively during translation.

Based on our own practical experience, we have developed several rules to follow when entering terminology from customer source files:

Based on these rules, I now check the customer’s files and decide how to best enter the terms.

The following image shows how terms are typically listed in the customer files:

In accordance with our rules, the abbreviation is entered as its own term in the termbase to ensure term autonomy.

Another specification we consider important is the status (meaning: allowed/forbidden). This field is a simple but effective tool for communicating customer specifications like the struck-through text. The corresponding entry in Multiterm looks like this:

The burning question many people will ask at this point: Is that really cost-effective?

It is true that preparing terminology data provided by customers takes time. However, experience has shown that this investment is always worthwhile, since translation quality – and the satisfaction of everyone involved – increases whenever terminology is streamlined. Manually searching for terms in lots of individual files ultimately increases the chance of errors. This is why we create termbases even for just a few customer terms in order to ensure that translators adhere to customer wishes (including those expressed in customer feedback).

At exact!, we sometimes also record terminology in the available termbases while translating if we determine that key terms of the relevant subject are involved. Our resources are therefore continually growing, and our customers receive translations in line with their terminological specifications.