E.g., 11/22/2019
E.g., 11/22/2019

Becoming a Successful Translation Manager

By: Lonie Goldsberry (PDI Ninth House)


27 November 2009

I began working at PDI Ninth House (formerly Personnel Decisions International) in December, 1999 and began running the translation process for our "360 development products" soon after. "360" is a method of surveying where a participant rates their own behavior and requests those around him/her to rate their behavior as well (i.e., boss, peers, directs reports and others). When I took over translation management, our process was archaic; at that time, the process was almost 100% paper driven.

This means that reviews were handwritten and sent express via courier services amongst countries, to me, my vendor, etc. I found that several of the on-going projects had been “on-going” for quite some time. Project plans were non‑existent and we never met our verbal timelines. I remember going to cross-functional meetings and repeatedly stating that my projects were in the hands of the reviewer. After about six months into the job, my colleagues informed me that, while I always spoke at our meetings, I never really seemed to say anything. From that point, things could only get better...

At the time, I knew nothing about how to manage the translation process; my love of the French language is what incited me to seek out this particular responsibility. What I did know was that I had to make this process completely electronic and try to bring the status of projects to the next level– whatever that was. The vendor I had inherited completely refused to deliver any translations electronically. In my many discussions with them, I felt like they were the customer and that I needed to serve their process. Ten years later, this discussion seems ludicrous.

Fortunately, a client assignment came in – one of my first start-to-finish projects. The client requested that we use a vendor that they had worked with in the past. The timing couldn’t have been better. And after successful completion of this initial project, I quickly fired the original vendor.

With the help of my new vendor, we quickly moved to an all electronic process. I was able to create project plans and deliver projects on time.

Eventually, the processes I used became very “standard” in the sense that there were basic steps that needed to be followed with each and every project. Identifying these basic steps was crucial in understanding that each step required a minimum amount of time (MAT). MAT represents how quickly a step can be completed in the best possible scenario with the best possible circumstances. Identifying these basic steps also helped to determined the “usual amount of time” (UAT) necessary for certain steps of our process. A good example of this is how long it takes for a reviewer to review a certain amount of text. I quickly discovered UAT by first trying to set timelines that seemed “reasonable” to me and finding out if reviewers generally meet that timeline, exceed that timeline or were late. Understanding MAT and UAT allowed me to create not only standardized processes but standard timelines as well. It also allowed me to come up with reliable, realistic timelines for more complex projects because I now understand the UAT and MAT of each piece of the process and can assemble them for an overall project plan.  

Standardizing processes and timelines and creating documentation to communicate those standards is also an efficient way working with colleagues. This helps to train them to have certain expectations that do not require one to elaborate on the “whys” or “how comes” with each and every project.

As I mentioned, I learned quickly that there were very basic steps that needed to happen in the process. This basic process looked something like this:

Translation>Review>Load>Verify the load.

I soon learned that this was the translation process in its infancy. There were quite a few problems and the most glaring was the fact that reviewers have a tendency to introduce unintended errors – including the most senior consultant reviewers. While their reviews are important not only for quality but also to reflect terminology in PDI NH’s industry, they are not professional linguists. Therefore, things like small typos and verb and gender mismatches are often introduced in these reviews. Correcting these sorts of mistakes too far into the process has real external cost consequences and other trickle-down effects for us.

On the other hand, one can find that there are greater cost implications with having more vendor involvement in the process. Decisions need to be balanced with potential outcomes. For example, for our standard content, a review implementation is always required. The vendor then has the opportunity to update the glossary and translation memory when they receive the reviewed translation. For future projects, we are better able to leverage previous translations and the vendor should be able to produce translation that is more in line with PDI NH’s needs. This demands less time of the consultant reviewer, and realizes cost savings for us.

For a client project, a review implementation may be necessary, but it depends on the content being translated and how much risk is involved. If the project is very content heavy, for example, a training workshop, then a review implementation is mandatory. The client is spending dollars on printing the materials and they have to be correct. For example, I would not require such an implementation on a “360 survey,” where we have only translated some short texts on behaviors or competency descriptions. For protection, I would be careful to provide this caveat to my client reviewers, indicating that a professional translator will not look at their changes and therefore, they are responsible for ensuring that all changes are accurate. For something more risky, such as an assessment product or service, we need not only a review implementation by the vendor after a client review; we also need to have a consultant review. The consultant review is used to protect PDI NH against any liability claim and to ensure there is rating reliability with this content. In this more rigorous process, the steps look like this:

Translation>Client review>Consultant review>Vendor implementation>Load>QA verification.

The costs around translation have now more than doubled.

PDI NH is conscientious about delivering high quality to our clients, and our consultants sometimes mistakenly believe that we should deliver high quality without the client’s input. If we are delivering a standard product, then this is true. If we are customizing content for our client, then we need to insist on client involvement. The client has customized the content, partially, for branding purposes. It is important to allow them to have the same input on the translations in the same manner they did when developing the source language (in our case, generally English). As Translation Manager, it is my job to provide advice on best practices. After almost ten years, I have confidence in the best practices that I have learned, Even if colleagues question the process, I found that taking the time to help them understand allows them to buy-in and support the process as well. Becoming a successful translation manager means never having to fix translations in a reactive manner for a client rather; instead, it means being proactive from the beginning and leaving a quality product behind with a happy client.

The bottom line is to create a basic process and then add steps (rigorous steps!) where needed for higher stake content.

In the second part of this series (in the Q1 2010 GALAxy), we’ll discuss:

  • Quality issues – learning – partnering with vendors
  • Supporting internal reviewers
  • Organizational buy-in for your translation process (on a functional level)

Lonie Goldsberry is the Global Translations Manager for PDI Ninth House, which for over 40 years has specialized in Talent Management Consulting using pioneering principles in industrial organizational psychology and cutting-edge research. PDI Ninth House does business globally in over 18 locations outside the US. Lonie Goldsberry has a master’s degree in Educational Psychology from the University of Iowa and a bachelor's degree with a triple major in Psychology, French and Literature. Lonie has been managing translations at PDI Ninth House for nearly ten years. Lonie has successfully created and implemented rigorous translation standards within the organization. PDI Ninth House’s Intellectual Property, due to service requirements, requires a high level of validity and reliability results. Lonie has been able to identify and implement the processes necessary to meet these requirements.

randomness