Ever wondered what it’s like to manage translation quality in an organization on a daily basis and get paid for it? ContentQuo is launching a series of audio interviews with experienced Quality Managers who will talk about their trade and answers our difficult questions about language, about quality, and about translation industry at large. Click Follow to learn when new episodes arrive!
“Bring the human out of the process and the data! People involved make the difference”, says Nathalie Moyano
Hello everyone, my name is Kirill Soloviev. I’m the co-founder of ContentQuo and also the host of this podcast where we talk about translation quality, how to manage that, how helpful it is for the corporations and enterprises and companies, and also about people behind this fascinating line of business.
Today I’m really glad to welcome our first guest on this podcast. I know she was working in quality management in a “small” IT company called Symantec — in fact, over 10 years if I’m not mistaken. Recently, she’s actually founded her own consulting business MOYTILINGUAL. I’m sure Nathalie herself will tell us everything about that.
Nathalie — could you please give our listeners a short introduction of yourself and your own background in the language industry? And a couple of words about what you do today at MOYTILINGUAL as well?
Sure. First of all, thanks Kirill for inviting me. I’m glad to hear I’m the first guest so let’s hope I’m the first one of many to follow. And I’m very interested to hear from other quality managers.
My name is Nathalie Moyano González and I run a business called MOYTILINGUAL as you mentioned. I’m a translator and interpreter by trade. I started from studying Applied European Languages — and I have a master’s in legal & economic translations. I started my career in localization in the 20th century — that was in 1999 (I know it’s scary to say that!)
I’m based in Dublin and I’ve been here for the last 15 years. I moved to Ireland with an offer from Symantec as you mentioned, and I joined the Online Consumer Support team back then as a French translator. We were doing all of the work in house and I was specialized in virus write-ups and KB articles and that type of highly sensitive content for immediate consumption.
I’ve never left Ireland, I guess the island trapped me and it’s home for me. I love languages and communication and I’m of French and Spanish nationality: born and raised in France but my parents are Spanish. I guess that makes me trilingual — and quadrilingual if you think of dance as language!
In my spare time, I’m a flamenco aficionado: I dance and I teach. This has really played an important role in my development as a professional and as a person, especially when it comes to training and education.
So I guess that’s me in a nutshell — everything except my birth year :)
Yeah, we can skip that part! It was not on my list of questions but if people want to know, I would be able to personally connect you with anybody with that sort of interest :)
It’s quite an amazing background so thanks for sharing that with me. You’ve really covered lots and lots of ground — and not just the language industry, which is what everybody listening to this (I guess) is in, but also many other things like dancing! I’m not a dancing type at all, so I’m always fascinated by people who have the energy in them, and the rhythm, and the movements to actually do that kind of stuff.
But going back a bit to to the business side of things. First of all, I understand that you have left Symantec and your corporate career a short time ago and started your own business. Can you tell us a little bit about that transition?
How did that happen? How did your background in Symantec — or outside of it, for that matter — actually lead you to that decision? What’s MOYTILINGUAL all about?
It’s funny: I think that actually today marks the one year anniversary of my resignation to date! It was my own decision to leave, and Symantec was a great company to work for. I’ve really learned lots. So it’s something that I’ve thought long and hard about.
But as with everything, models change, and without a Language Services team, with my role being sort of merged into something different that was not quite going into quality management, I decided to leave (and I won’t get into the details of this). And so I decided to start my own business. It’s really a one woman show — MOYTILINGUAL was set up this year, in January 2018.
I really took a couple of months to breathe after nearly 15 years working in Symantec and then thought about whether I wanted to go back into full employment. But today, you can do so much with the internet — and that way, the businesses are blooming, both startups and small companies. So I thought I’d just set up my own business and enjoy a better quality of life, I guess.
But I’ll continue doing the job I love and that’s really helping companies and individuals solve their process and quality issues, and, more importantly, build stronger relationships between functions. This is all for a successful and mutually beneficial localisation experience.
I’ve seen a lot of struggle and my core competency with MOYTILINGUAL is really cross-functional processes, data management, and Root Cause Analysis. This is where I spent the last few years of my career developing.
And I do provide as well multilingual and linguistic services, such as interpretation and translation and voice recording in my three languages. But the core is around process and data and relationships.
So I like to think that MOYTILINGUAL is about more than just translated words. I like this industry and I really want to bring the human out of the process and the data. I’m a big fan of process and tools and data. But whatever you say, people involved make the difference. And if they understand the goal and if they have a common goal, we humans are capable of anything. So I guess that’s what MOYTILINGUAL is about.
Wow, that’s quite amazing Nathalie! This actually brings me back to my own corporate career when I was a Director of Localization for a software company, and people asked me: “what’s the most difficult part of your job Kirill?”. And I always used to say: it’s about the people!
At that level, after spending five years in my last company, it became really clear that processes and tools — and even words and language — they all exist only to be something we talk about, that we interact with as human beings.
And ultimately, localization is all about people working together. Probably that’s the single most important component of quality! What’s your shot at that? What’s quality to you, Nathalie? How do you view it? How can you define it?
Wow, defining translation quality — that’s a big one! It’s really the result of a combination of things, right? We’re all familiar with the basic criteria, and we go from fluency to terminology, how correct the language is, how correct the translation is, whether you’re really rendering that message from your source language. You want your quality output to be fit for purpose, it has to be suitable for the audience you intended it for.
Quality is a hard one to define. And I think that’s where the struggle comes in — when you try to explain it to people. People would say: “Well, is it subjective?”, or “Is it objective?”, but it’s not that simple. Quality is both subjective and objective. It really depends on your perspective.
Grammar and syntax, we all agree they’re not subjective: if something is incorrect, it’s incorrect. I can be very particular with correct grammar, and syntax, and spelling, but a turn of phrase, wording — all these can be subjective.
This is where the key is, I think: when you explain quality to people, it’s about knowing the parameters of the translation, of the review experiences. What conditions do you translate a particular set of words for? Did you get a lot of reuse?
E.g. you translated two words out of a two page document, and then suddenly, some older content that may be there brings the quality down. It’s all about knowing everything, all the parameters.
You can’t just define quality as “this is good” or “this is bad”! And that’s a lot of the work I did, especially around root cause analysis. I wanted the fights between somebody saying “This is good” or “This is bad” to be over. It’s all about bringing the different perspectives and understanding the struggles on each end, and coming to a common understanding. So yeah, it’s a big question.
It’s a big question. But that really brings us nicely into into my next question. You started to talk about the work that you’ve actually been doing. Can you go back in time, to a short while ago, perhaps, and tell us how would a typical day in your life as a quality manager look like? What did you actually have to do during the day?
Oh, well. A lot! I guess I was a Quality Manager but also a Process Manager. So this was all intertwined: I was really the link between all the functions who were involved in the localization of software. From the process side, I would design and optimize the cross-functional processes working with the different functions, but also the vendors.
Then I would build and monitor our quality metrics and develop actions for continuous improvement. And I would also train and support the teams and the vendors. So all of this was all nicely tied into the same package. It was quite an effort, I guess.
And, bear in mind that a lot of this would have been done manually, especially around building quality metrics. This is something I did by observing and by listening to issues. So for me, quality was never about something simple. It was always vendor specific or more importantly, content specific: depending on certain conditions of a particular content that was translated.
The core of my work would have been around building and monitoring quality metrics, and working with the teams to improve them, to understand what was wrong, and how we can improve it. The process side was more like small tweaks: finding the little cracks in the process, the pain points where process was too tedious and cumbersome, and then trying to improve it.
So metrics and actions, and then a lot of educating. All of this came in waves but the quality was always at the core. Then I would be pulled into process optimization and into a lot of training: I did a lot of that, to implement best practices and to make sure people would follow the materials.
And I would also get a lot of feedback from people. For me, it is essential to understand where people struggle, so I can help them make it better. I was the go-to person, and it sometimes felt like you’re torn into different directions. But I like it because I really like working with people!
Right! I think it’s maybe one of the hardest, but also probably one of the most rewarding things one can imagine.
I’ve started my own career working with technology mostly. Then from technology, I gradually shifted to processes, and ended up — probably the same as pretty much everyone who is ascending the seniority ladder in this industry — managing people in very different ways.
I can totally understand what you’re talking about. About people, though: is there any particular group of people in your role that you’ve been spending the most of your time with? Who were your peers? Who did you actually get to work with the most and interact with the most? Who would take up most of your time?
Ispent enough time with everybody — I never wanted to neglect a particular group. Funny enough, I would spend a lot of time with vendors, even though they were external. To me, vendors in the current industry, where everything is outsourced, hold the key to solutions. So I really nurtured that relationship and made sure that they could trust me to want to help them.
Apart from vendors who were doing the core of the linguistic work, Engineering and QA were the next groups that I would spent a lot of time with and unlock those little process issues to fix. Sometimes, you would find engineers and linguists working perfectly together. But you would also get linguists getting very frustrated with engineers who are asking them for shortening strings, for example. And engineers would get frustrated with translators, too!
In extreme cases, an engineer would be googling or using Google Translate to shorten a string and then asking why the linguist was not doing it this way and provided something long instead! My work was really being between those people and making sure that they get along. These would be my core contacts.
And then, because I was the Quality Manager (and still am!), functional managers would be next. I worked on the ground, in the field, with the people doing the work — but then I wanted to bring that to the next level. For functional managers & executives, I would present an image of what we were doing.
All of the metrics that we built would be used for setting targets for our vendors, or pointing out issues to engineers which could improve the way we did things (because we had too many bugs being logged, for example).
I also presented a more high-level view for executives, to help them really understand what is quality, to get away from that noise where this quality is bad or this customer complains, and really get to the bottom of the issue. Doing all that dialogue at different levels, with different with different groups of people — this is what it was all about.
Iwas just going to ask you, but you already started answering that. What I wanted to talk about is the value that quality management and people doing the job — the quality managers — deliver to a business.
How can people actually demonstrate that value? How can they measure it?Not just quality per se, but rather quality as a tool that companies can use to drive business objectives. How did that work in your case?
Nathalie Moyano 17:25
The value that we bring, ultimately, is happy customers! But not just happy customers — engaged vendors too. Because if the vendors produce quality, they’re going to want to continue producing quality, and you’re helping them produce that quality.
You remove the roadblocks, and then you’re going to get other types of value, like reduced number of bugs, less queries raised, less noise, less and less time spent on doing things that are not productive.
So how you demonstrate that? For me, it was through metrics. Dashboard of quality per content type, or per product range. I adapted them to the people who wanted to know what quality was about. I used to keep a huge dashboard of quality metrics, but not just scores — also down to the level of WHAT were the issues raised.
This is where the vendors really helped me to qualify the content. What types of issues? What could have caused them? I had translators and reviewers working hand in hand to provide that.
And I even had values in terms of source quality. Going back to product developers — who I worked with a little bit (not so much) — and pointing out source issues that translators, due to the nature of their work, were raising was very useful to them.
So quality management is not just for the end of line where you get your score. You also get all those details that you can feed to the different functional groups and really help them improve on their end as well. I hope I did demonstrate the value of quality management to multiple groups! It’s constant work, and it’s difficult — especially when you handle a lot of products and a lot of languages. You can get lost in the details.
But if people know you are the go-to person, they can really spell out what it is they’re looking for. And then even the vendor manager would come to me and ask for particular metrics. So all of this is down to not just the linguistic quality alone.
Indeed, lots of things for everybody to consider! But I want to dissect this a little bit into smaller parts.
You mentioned using metrics heavily, and dashboards with those metrics to monitor. The challenge of scale in quality management makes so many things harder when you suddenly start to get lots and lots of projects and lots of data to take care of.
What role does technology play in the work that quality managers already do, and how can it help quality managers do their job more efficiently and ultimately be successful?
Technology is essential. And I’m saying this as a linguist! It is not true that linguists are allergic to technology.
You know, I started back in the 20th century where we started to see MT. I was involved in fine-tuning the engine and coding dictionaries. And we saw how MT changed the way we were translating large volumes. I don’t think I know translators who would want to tackle millions of words without the help of technology. So what’s important to remember about technology is that humans are behind it! And that it’s the way we improve the way we work.
We’re all exposed to technology constantly. That last industrial revolution is going very fast, and we’re seeing how we’re adapting.
Now, this is not a product placement, but I had to say it because I did start looking into ContentQuo’s translation quality platform. For me, it’s an amazing new tool to manage quality, it’s something I’ve never seen before!
In a way, I would have liked to see it when I was still working as a quality manager full time — because I would have totally ditched my manual metrics and work to improve the tool to suit my needs back then (I can still do that but in a different context).
It’s everything that I was doing, which was categorizing issues and producing a dashboard of quality scores. ContentQuo has it all in one place, and you’re not handling multiple spreadsheets to pull things together anymore. It’s not that it’s that difficult — it’s just very time-consuming and prone to errors.
I think there’s a lot more to do with technology around localization. Every time we think we’ve reached the top, we realize there is a lot more than we can do. So it’s very exciting times! And I’m definitely looking forward to seeing more technology come into play around managing quality. It’s really great to manage quality when you have a tool that helps you.
Thank you so much for the kind word! For the record: that was ENTIRELY unsolicited! You should all try building a product some time — just to hear that kind of feedback coming in :)
I was just going to say that with the speed of the current industrial technological revolution, I’m hoping that we won’t crash into a wall like the proverbial self driving car because of our blind trust in technology. Keeping in mind the idea of people behind everything that happens is probably the best way forward.
So my next question is actually about the people. You talked so much about the language partners and vendors, specifically since you’ve worked with those external parties a lot. There is another role which is quite common in the industry, and that’s the management of vendors, or Vendor Management.
Where would you see the border line between quality managers and vendor managers, especially with your background of having worked so much with the vendors directly? What do the two titles or the two roles mean to you?
Both roles are definitely essential. And I used to work closely with the vendor manager. The two roles together would set the right strategy for vendors, but it’s definitely not the same role. I know I couldn’t have done my work without a vendor manager. And this is a role I’ve longed to see come into the team, and so it happened.
Vendor managers are more focused on… I don’t want to say “administrative”, but they are in constant talk with the vendors from a setup perspective: account management, invoicing, price setting and negotiating, and keeping the people & resources involved.
And the quality manager is not looking at that side of things. They really are looking at the content, at the output, at any quality issues. And if there was an escalation, then you would work more closely with a vendor manager to resolve that escalation — not that I’ve seen it happen, really, but that’s where the two work together.
They are very different roles for me, and I don’t think one could take on the other person’s workload. It would just be insane. And I think it would damage those relationships in a way, because you would neglect (at some point) some of the aspects of those. So: two roles, crucial and definitely working very closely together.
Thanks for that! Once again, another fine example of people teaming up with one another to deliver much better results than individually.
Thinking about that — and going back to your quality management career — what did you actually LIKE the most in your work as quality manager? What was the most satisfying, the most gratifying thing in your experience?
Working with people! I think you’ve gathered that from talking to me. I really like helping people and functions connect and work together, rather than against each other. And I’m a middle child — I’m thinking maybe this is where it comes from.
There’s so much potential in people feeding off each others’ energy and skills and competence and experience. That’s what I enjoyed the most. I was (and am) definitely not the kind of person who is shying away from people. I like to communicate and like to understand what is not going well and how can I help.
Then I bring people together. Like we said, it is all about the people. We’re using technology and gladly so — but at the end of the day, localization is all about the people involved. And if we work well together, we can do anything!
This sounds really, really promising Nathalie! Perhaps you could think of a specific story or anecdote from your work with people around quality? Something funny, or something that you really remember after all those years — something that you could share with our listeners?
Like I said, I did a lot of training — especially for engineering and QA.When you’re localizing UI, you always hit that wall of UI “real estate”. And I would get regular reports of vendors who were asked to abbreviate strings, as opposed to shortening them if they couldn’t just shorten them.
So I used a lot of those examples. That’s where me working with vendors so closely was great, because I got so much feedback and so much information from them.
I remember a session I did in the US a while ago. I was doing the part on Do’s and Don’ts of shortening and explaining the difference between shortening and abbreviating. And I like to use reverse psychology, so I asked them how do they feel about shortening a string? Let’s imagine something like “install assistant” — bear that string in mind.
There’s no option to wrap the string around, and translators cannot shorten it — so they have to abbreviate it. So I asked the US engineers how would they feel if (in English) I would ask them to abbreviate “install assistant” because they cannot shorten it. You know, clip the last six characters… I’ll leave you with that. You can write down “install assistant” and remove the last six characters. How would they feel?
That was a funny way to explain the difference between abbreviating and shortening between languages and why it mattered so much that we tried to adapt the UI because adapting language is not that easy. I got a few laughs. I’m not hearing you laugh yet?
I just did the exercise in my mind. (laughs)
What single thing would you say is the HARDEST when you’re trying to manage localization quality for a large organization? And please don’t say it’s the people, okay? (laughs)
Whether the people are inside or outside of this, I’m just trying to be a bit more specific. Which single problem did you struggle with the most? What was the most problematic thing for you to solve in your work as a quality manager?
Aside from the people? As you rightly said, they are the greatest and the hardest thing.
In localisation, probably the hardest thing is still not being involved enough in the product development process. We’re still too detached: products get developed in English normally, and we’re just dumped a lot of stuff to translate. And we never get to say, or hardly ever get to say much.
Back in my time as a Quality Manager, I would have seen a lot of work done on terminology. You can really get to talk to the writers and the developers and set that, so that terminology would not suddenly fluctuate during the product life cycle — because it has such an impact, and it costs a lot of money as well. So I think that’s probably one of the hardest things.
If you want real quality output, you have to start early. It’s not just a reactive process. For me, it was the the hardest thing, and that’s why I started collecting a lot of feedback on source quality — because it impacted so much the translation quality at the end of the day.
So if I have to limit it to one thing, that’s the one thing that comes to mind!
So pushing quality upstream is basically what you’re talking about?
Yeah, the usual complaint, right? (laughs)
Iwas just going to say that TAUS has these annual Quality Evaluation summits, and the European version of those is happening in Dublin. For the past couple of years, we’ve been talking there about how to do that, how to push quality upstream.
Many companies, especially large ones, realize they need to do it. But not that many are actually able to do that consistently. Doing this on an operational level is hard, and probably at least in part because of the organizational structure. I think this is a worthy challenge for everybody to attack!
I am now building my own startup & having a very, very small team where we virtually have no organizational structure. I see that these things can be very different. And I think it’s that small team experience and being able to pierce those org boundaries is what makes people so successful when they’re managing localization — quality or otherwise. Thanks for that!
Now, another scenario that I was always pondering. It might be a very practical question that you might have experience with. What do you think could happen — what’s the worst that could happen if an organization suddenly STOPS investing into translation quality management and decides to abandon its efforts? What might go wrong?
Well… I think first there would be smoke, and then probably flames. (laughs)
I cannot emphasize enough how crucial it is to manage quality! And that’s across functions and levels. I truly think that quality is everybody’s responsibility. But it has to be managed by a strong core, so that it can stand.
I do not have examples, and I certainly cannot comment on what’s happening [at Symantec] after me. But it is not a good idea. What I’m seeing a lot in certain spheres is the de-specialization of language quality, of language industry. And I’m not talking about [inside] the industry — here we are so strong in it.
But I’ve heard so many times language quality being reduced to just speaking a language! I speak three, I studied translation, and I worked hard in the localization industry. But there is the whole world saying “What is this language quality about? Is it good or bad? Let’s [just] move on.”
My fear is that it would be reduced to something that can be discarded, or that anybody can do. And it’s not like that, and having tools to manage quality better is going to be definitely a great step forward! But you will still need the right people to manage those tools. And to set it up to have it used by the the right people. So it wouldn’t be pretty — I think it would really be terrible.
But I don’t think we’ll get there. When you are handling customers, for example Japanese customers or German customers, who are very particular about quality, this is great: you know the customers would be the first saying “What is going on?”
It’s one thing thinking you can save on costs and translate content of those knowledge base articles with Machine Translation without a fully maintained engine. Then you get things that you can barely understand, but you get by.
However you cannot do that with with software or with medical tools instructions. So it IS crucial. And again, I can’t emphasize it enough! In setting up my business, this is my message — and I’m not there waving a flag, but it’s really something I take to heart and that I’m passionate about. I’ll do my best to keep the war going! We need to manage quality.
Honestly, I think it’s a worthy war and especially so as organizations start to grow. When you’re small, there are so many competing things that you need to pay attention to. Being active about managing quality is something not many companies can do, really. But as you grow, as you scale up, this is something that is really important to make sure that the risk doesn’t go up.
We’ve talked quite a bit about training and about education. I totally agree that all of those factors play into the result of localization. But with scale — and that’s my personal experience — what happens is that personal expertise and know-how is constantly clashing with the circumstances of processes and the contracts and organizational boundaries.
And it’s in the interplay between the two that actually makes getting to quality at scale so hard! You put in the best ingredients, but then you can never be 100% sure of what comes out of your process — until you actively invest into controlling and managing quality! I think that’s the point here.
We’ve covered quite a bit of ground here. I’m really thankful for all the things you’ve been sharing today! To close this up on a… let’s see if that turns out to a be pessimistic or an optimistic note: What piece of advice you would give — not necessarily to the outside world but to your old self — as you’ve been starting out your quality management career many, many years ago? What would you say to a younger you?
Inormally have no regrets. I’m an eternal optimist. I think that you take the path that you take, and you learn on every journey. But now that I know so much about the industry, maybe I would put more effort into getting upstream as we mentioned.
Things like User Acceptance Testing: it was a thing of the past, and it was done in English — it was not done in other languages. Maybe strike stronger relationships with that side of the business, and get more involved in Product Dev. I did work a lot with developers, but there was never enough time. It’s something that I would have liked to focus a bit more on.
Engineers are trained by engineers, and we look at i18n and l10n — this is great, we get an agile process and we’re churning. But linguists are never really involved enough. If I could only strike strong relationships with that side of the business and work more to show what the challenges are, for linguists further down the line — maybe try and influence some change there. That would be one thing for the younger me.
Ithink that’s absolutely perfect advice, Nathalie, and I couldn’t agree more! Having started my own business and left the corporate world several years ago, I am now looking at my old career, and this was EXACTLY what I would say: “Okay, so can you talk the talk that engineers talk? Can you talk the talk that business people talk?”
That is, can you articulate the value of localization and the value of quality in terms of how it would actually help the company do better, and how it would help its customers and its users be more happy? I think that’s one skill that everybody listening to this can try and learn — and it’s not that hard!
Just get out in your spare time, on your weekend. Try doing something yourself, try building a startup — that’s what I say to everybody! That’s my advice. You’ll be amazed at how much better you’ll become at managing localization quality in particular, and localization in general. Being better at life is the ultimate benefit that we’re all after here.
Alright. I can’t think of any more questions for today. Thank you very much! This was Nathalie Moyano, business owner at MOYTILINGUAL and a long-time Localization Quality Manager at Symantec with us today. Thanks very much for listening, and let’s see all of you in the next episode of the podcast!