| Michael |
I’m Michael Stevens. |
| Renato |
I’m Renato Beninatto. |
| Michael |
And today on Globally Speaking, we’re in a really comfortable place recording. |
| Renato |
We seldom have a chance to be in a real studio. This time, we’re in a beautiful location here at HubSpot in Cambridge. We discuss all kinds of topics related to localizing content into multiple languages for a product that is very sensitive to the type of content that you publish because it’s content marketing. |
| Michael |
They’re so sensitive they even call out what content to delete. For instance, the beginning of this podcast probably is not going to make it into the German cut. Hopefully you’ll understand why we’re talking about this, but we’ll let our guest introduce himself. |
| Chris |
I would introduce myself as Chris Englund. I’m the manager of the localization team here at HubSpot. I’ve been in this role for about the last three years, but I’m actually a HubSpot veteran, not a localization industry veteran, and I’ve been around the company here at HubSpot for around eight years. When I moved over to localization, I think the thing that was most exciting to me was the opportunity of taking this pretty magical startup that’s aiming to revolutionize the way that small and medium-sized businesses do their marketing and bring that to bear in all of the markets and languages in which we do business. |
| Michael |
The trend of people going into the localization team is an interesting one. For years and years, it was always they worked their way through the localization team into other areas of the business. What drew you into this part of HubSpot’s business? |
| Chris |
Well, that’s easy. I used to work with HubSpot customers, and the things that they would always tell us about using our software is that we don’t sound like other enterprise software. We sound friendly, approachable. They really like the tone, you know? When you generate a lead or close a sale, you’re congratulated as if by a person. And that makes our users feel really good and accomplished. I was always hyper-aware that just our English language was pretty key to making a positive user experience, and so how do we bring that same advantage to our German, Japanese, French, Spanish, Portuguese users? We can’t just give them a translation. We want to give them the same customer experience and feeling. And that’s what got me excited about localization. |
| Michael |
Some of our listeners may not know what HubSpot does. So, what’s HubSpot in a nutshell? |
| Chris |
We’re a software company based in Cambridge, Massachusetts, with offices now all over the globe. And our software is marketing, sales and now services software for primarily small and medium-sized businesses. |
| Renato |
It’s one of the most used tools in the language services industry. I’ve been a HubSpot client for a long, long time, and several LSPs use their sales management system. But the marketing automation part is really the core of the technology. Marketing is very, very local, emotional, language-intensive, and it makes it quite complex to localize. I imagine there is a big element of original content writing in different languages, of transcreation and recreation, more than just translation. |
| Chris |
Yeah, absolutely. What makes us sound super-unique and approachable in the English language is also deeply rooted in North American US culture as well. So that plays really well because it’s part of our culture. But when we bring this same blog post or this same ebook or content offer into the French language, we have to be respectful not only of the language preferences, but also what’s important to that culture. And as a result, we find that our in-house language team is responsible not only for reviewing and QAing articles or content that’s been produced by an LSP, but also responsible for adapting that content to the cultural realities of the, say, French buyer persona, the German buyer persona. |
| Michael |
Do you have an example of a piece of content that sticks out to you in particular? |
| Chris |
I can give you a pretty funny story of a time I joined a call between our German localization specialist and the team of German linguists whom he was trying to give a bit of a training session to on how to adapt a HubSpot blogpost. If you’ve ever read a HubSpot blogpost, they tend to have a bit of an informal tone to them. They walk you through the author’s state of mind and where they were at the time and what was happening for them when a particular conclusion or something dawned on them. Dierk’s first recommendation to the German linguist was, “When you translate a HubSpot blogpost into German, just delete the first paragraph.” |
| Michael |
That’s an unusual translation task: delete the first paragraph. |
| Chris |
Let’s lock this one from translation. The Germans aren’t gonna care about whether you were… |
| Renato |
In a good mood or if you just had three coffees… |
| Chris |
For Americans, we like this. It personalizes it, you know? It makes us seem like it was written by a person. |
| Michael |
Oh, it creates a world, it creates a universe. |
| Chris |
Not so much in every culture. |
| Renato |
That’s fascinating, that’s a good story. HubSpot is a data company. One of the areas that is a big challenge for any localization program is identifying whether adding languages or investing in a certain market is a good decision or not. Because you’re a data-driven company, are you able to have a clear ROI or return on investment analysis on the languages that you translate into? And at what level do you track that, if you do? |
| Chris |
It’s by the grace of the return on investment that we continue to have a localization team. In fact, demonstrating ROI on localization is incredibly core to what everyone on my team does. So, an example that I recently used in an internal presentation was a content offer that we launched back in December 2017 in conjunction with the French marketing team. With their collaboration, we localized a single content offer that must have cost us around $1,500. To take this ebook that had been previously written in English—you know, a sunk cost from an investment standpoint; it already existed; we had already invested in it. We had this localized into the French language. Using HubSpot, our own tools, we’re able to see what draws in visitors, what converts visitors into leads and since we’ve got sales data as well in the same system, we can also see which campaigns drive us actual revenue. And, from that investment of around $1,500, I believe this campaign has already contributed about $144,000 in annual recurring revenue. It’s a no brainer, really, to localize certain content. On the other hand, there is content that we have localized before and we look at things like video view counts or the number of weekly active users and we determine in fact, we won’t continue investing in that area as well. The data also leads us to make better investment decisions for our localization planning, thinking toward 2019. |
| Michael |
We’ve said before that companies rarely discontinue a language that they localize in. That doesn’t mean that they won’t change what content, how much or what specific pieces they’re localizing in. And you guys use data to make those types of decisions. |
| Chris |
Yeah, I’ve seen personally in other companies two different approaches to how you decide what to localize. There’s either this top-down approach where corporate decides “we’re going to localize all this minimum set of content.” But then what happens is in certain regions, it’s just inapplicable or it doesn’t translate well, and so it doesn’t get used. Conversely, there’s the other model of having the regions decide everything, but then it’s the Wild West and we don’t have a consistent experience that we can offer to users of all languages. HubSpot’s taken a hybrid approach. What we do for a great deal of our marketing localization budget is to gather the project parameters and all the needs from across the business, whether you work in-market or here in Cambridge. But then, to partner directly with regional marketing leaders to have them help us make the final call on which projects to green-light and which ones to keep in our back pocket. And we’re hoping this way, we’re able to create a customized approach to each market that still takes into account the minimum viable user experience we’d like across all regions. |
| Renato |
So, do you categorize content as global, regional, local, or is there a hierarchy of content that you create, or it’s case by case? |
| Chris |
There’s certain content that we must localize. It’s the price of admission to that market. Things like legal documents, for instance, are no brainers; we have to in order to remain compliant in certain markets. Then we prioritize our content according to its impact on one axis, on customer value, and then on the other axis, enterprise value. We can plot each project on each of those axes and that allows the project to fall into one of four quadrants. If it’s high on customer value and high on enterprise value—let’s say, an ebook that teaches a customer to do a new thing, but it also generates us leads—that’s a no-brainer. We should definitely localize that; that’s in quadrant one. |
| Renato |
You have this inbound event which is a marketing school, essentially. It’s a great event. |
| Chris |
Just two weeks ago we had maybe 24,000 attendees from all over the world. |
| Renato |
This is a major event and it’s growing because the company is growing. How does this growth affect localization? And do you know, for example, what is the contribution of the international business to the overall business of the company? |
| Chris |
Off the top of my head I don’t know the numbers. |
| Renato |
Okay. |
| Chris |
Something that’s really cool is that our CEO, Brian Halligan, actually called out localization as a driver of our international growth on two separate earnings calls, which obviously made us feel pretty awesome. |
| Michael |
How many languages do you all support now as far as your team or the company as a whole? How many offices globally do you have? What are some of those stats? |
| Chris |
We currently have five languages: German, French, Spanish, Japanese and Portuguese that we support on the localization team. We enable some other languages that are not used as core languages, but those are our five core languages. In 2017, we localized about 5.2 million words. In 2018, we’re on track to localize around 7.5 million words. So, there’s been some tremendous growth on the localization team as well. HubSpot has a European headquarters in Dublin. It’s been our home in Europe, now, for over five years, and we’ve recently opened an office this year in Berlin and we continue to invest in our offices in Tokyo and have opened an office in Bogotá this year. In addition, HubSpot’s APAC headquarters is in Singapore and we’ve got an office as well in Sydney. |
| Renato |
With this growth, are you hiring people locally, centrally? How are you organized from a localization perspective? |
| Chris |
One of the important enablers to our ability to effect change in the organization as a localization team is that we sit under HubSpot’s international operations team. That means we’re agnostic to the various teams that need localization, whether that be marketing, sales, services, support, etc. To support this kind of international growth that we referred to before, we’ve hired in four or five offices, if we include remote workers as well. My own team of project managers, for instance, is in three different locations. Our localization specialists are spread across I think three, and translators as well. |
| Renato |
What is the skill set for a person that wants to work at HubSpot? |
| Chris |
First of all, and probably most importantly, they’ve got to embody the HubSpot cultural values of heart, humility, empathy, adaptability, remarkability and transparency. Empathy and humility are big because these are linguists that have to work directly with people who have not a clue whatsoever about localization or translation. They need to be ambassadors for global growth—a lot more than simply linguists, which none of them are “simply” linguists. They’re all language people by training, so we’ve got project managers who have LSP backgrounds, we’ve got project managers with internal HubSpot backgrounds. And our linguists are all translators by trade, but they’re so much more than that: they’re technology experts, marketing experts. We’ve got folks on our team that are teaching our own engineering team how you build an internationalization framework that supports high quality localization. We look for people that have experience both in-house with a buyer or people who have experience on the LSP side. But primarily, we’re looking for an individual who’s an excellent teacher and has a great deal of empathy for understanding the business goals and needs of the people that we serve. |
| Michael |
At HubSpot, you probably talk about cost of acquisition for customers. |
| Chris |
Of course. |
| Michael |
Have you integrated the cost of localization into that? There’s the attribution issue. So, everybody is fighting; like, sales is like, “Oh that’s because we have good sales people,” and marketing is like, “This is because we have good content,” and then someone’s saying, “Well, they would have come if it was in English anyway.” But, if you have a digital customer journey and you know how much every customer costs you, it’s a much easier equation. |
| Chris |
It’s fair to say that there’s a lot of touch points that bring an individual from just learning about needs that they may have related to in-bound marketing to finally purchasing the software. It’s a fact they’ll be touching a lot of different campaigns. But I think the question that you’re asking is how do we decide, fundamentally, whether to create content natively and what is the value of that, versus what is the value of localizing content as well? We’ve run a lot of tests because we wondered the same thing ourselves. Our regional marketing teams are very hard-working individuals and they’ve got a very limited amount of time to spare. So, how can we make sure they’re spending their time in the right way? We’ve found that for certain locales, that native-language content vastly outperforms localized content. HubSpot research, for instance, is doing a lot of really original, interesting, data-based research. And that’s the sort of thing that we would want to localize, because it is really helpful information that can’t be created from scratch in the local market. We’re constantly deciding, first of all, does this content have both customer and enterprise value? Secondly, we’re also asking the question, “Is this something that we should localize or is this something that we should create in the native language to begin with?” |
| Renato |
This goes back to the ROI question. What other elements do you take into consideration? |
| Chris |
The return on the investment I mean, it can be captured primarily by customer value and enterprise value. Customers are really important to HubSpot and that’s been the case since the very beginning, but, from an economics standpoint, the reason is because we’re software as a service. So, folks pay for HubSpot as a subscription, and in fact, the monthly rate really doesn’t compare to the cost of acquisition, to your point before. And that’s why the name of the game for any business in a SaaS model is to retain customers. Well, you can retain customers the wrong way. Say, the way your cable company or cell phone company does: locking them into contracts. But the way that we prefer at HubSpot is to keep them realizing value and to continue growing themselves, to continue to learn and to grow their own marketing and sales and services programs. That means taking a look at a lot of data related to which apps, which parts of HubSpot do our customers use the most, making sure that we devote our localization time according to that. So, we might spend a ton more QA on parts of the UI or parts of the HubSpot product that are heavily used by our international customers. And that might be different by region, too. Our Japanese customers may prefer, say, the content tools, blogging, website creation, whereas perhaps our LatAm customers are more interested in sales or some of our CRM tools. And therefore, we would apply a different metric to each. So the return, it’s got to be captured in different ways. It can’t all be revenue, but it can be things like a decrease in the number of support cases by publishing knowledge base articles in that language—we can calculate what the value of that was. For the case of spending a greater amount of time reviewing the UI, we can calculate this in terms of the number of customers who use that product that then stayed on with us for yet another month. We capture the value in a variety of different ways. |
| Renato |
Is there a demand inside the organization for this type of information? Is it a decision-support kind of thing? |
| Chris |
If you look hard enough at the data, you will find a chart that shows that you did a good job. You just have to look long enough at the numbers and find the numbers that support your conclusion. We’re such a data-driven organization culture here that the language in which I’m speaking with you about how we calculate localization ROI is exactly the same factors that leaders on our services team, on our sales team, are also using. A month ago, I was presenting on the upcoming 2019 budget ask to our national HubSpot executive leadership meeting, and the points that I was making were all based on return on investment; figures and calculations that would be used in any other area of the business to calculate appropriate investment. So, we’re all speaking a common language when we’re talking about customer retention, customer value, enterprise value, cost of customer acquisition as you mentioned before, and when I say that localization contributes to bringing down the cost of customer acquisition, we can compare that same investment to another investment vehicle, maybe entirely outside of localization, that could also bring the costs of customer acquisition down. I have to feel responsible for saying that localization can do it better, and if we can’t, then that means we have to legitimately go back to the drawing board and come up with something that, say, could compete. |
| Renato |
Are you planning to add new languages? And what is the decision criteria for that? |
| Chris |
That’s just the time of the year that we’re in right now, which is planning for next year, and we typically choose languages on an annual basis. This year we didn’t add any new languages. We wanted to go deeper and get higher quality in the languages that we already do support. As you know, localization often starts out within one part of the company. In our case, that started out in two little parts of the company: our product and on our marketing team. But now we’re trying to color in some of the blank spaces: our services, our support team, our knowledge bases. That’s been our main focus for this year, and I think that’ll be our main focus next year as well. We’d like to be able to say because we’re a company that prides ourselves on customer value, that we’re providing the same value to you no matter what your language is. We usually add languages and decide on which languages we need to support on the basis of the number of users that we’ve got within that country already. HubSpot has customers in well over 100 countries, so we’ve certainly got users whose native language is not supported right now. We can look at that information, we have analytics from our website and we have deep analytics from our own product. We also look at where people get stuck, too. We may have a lot of customers in a country where people who live in that country already have a high fluency with English, and so we might choose to defer that investment. Or we might face an opportunity where we’ve got just a few customers and we’re kind of in this early adopter phase in one country, but we know if we just would invest in the language, it would push us over the top. When you buy HubSpot, you have unlimited access to a support team that you can call at any time for help. You have access to a knowledge base, you have access to account managers, a consultant, etc., etc. For us to add a language actually means also scaling up every single one of these functions. And that’s why my boss, for instance, is on the international operations team as a VP of international operations and strategy. Nataly is making sure that we’re not just simply spinning up the UI in the Dutch language, but rather, we’re thinking across the entire business about how we would support the Dutch language in everything we do. |
| Renato |
Dutch is one of those countries where you have a high threshold for English. |
| Chris |
Luckily we’ve dodged a bullet on that one, right? |
| Renato |
Maybe it’s not so high in Italy or Hungary or other markets like that. |
| Chris |
Yeah. |
| Michael |
Anything else you want to cover? |
| Chris |
I would just underline and hammer home the fact that I think it’s really incumbent upon localization teams like ours to continually tie their investments and the decisions that they’re making to the returns that they provide. It’s not going to always be 1,000% return on investment, but I think that we should be willing to hold a high bar for our localization investments and be willing to say it may not be worth continuing to localize this piece of content in this way in future years. If I’m investing a ton of money in marketing localization, that really ought to produce some sales, and not just be done for show or to say we’ve checked the box and we’ve got just as much in German as we have in English. |
| Michael |
If you had to give advice, where to start, to another localization manager or director, with raising that bar, where would you have somebody begin? |
| Chris |
Every type of localization project they’re doing should be tied to a success metric that they’ll eventually be able to measure and that other people in the company agree is an important success metric. |
| Renato |
Chris, thank you so much. |
| Chris |
Thank you. |
