From — November 1st, 2018
An interview with the email team from Withings about localization, segmentation, and personalization to foster engagement
Sending a single email that doesn’t suck is challenging enough. But emailing customers in multiple languages?! While also achieving stellar engagement?! That takes some serious email chops.
The Withings email team (formerly Nokia) runs their operations from two continents — making the most of simple tools and the universal human desire to get healthier. They offered up some interesting insights on managing a global operation, encouraging customer engagement, and building excellent email flows.
[Mikael & Raphael]: On our side, it’s mostly in French, that’s our native language. Then we usually have a call with Susie where we explain our initial ideas. Then we try to write solid copy both in French and English, and then translate it into other languages like German, Spanish, Italian.
[Mikael & Raphael]: It’s not exactly the same, actually. What we do is think about something, but it’s not finalized, and then we create a brief in English. Sometimes we draft the content, but mainly we just structure the content, and we try to give directions. In the end, we have a lot of different people between each level, and at least for the French content, it’s never translated from English.
[Susie]: Not at all, and that’s kind of the fun of it though. They always initially discuss it in French, create an English outline, and we go from there. My French is terrible, (I should never be allowed to write one word in French), but my French is good enough to see what they’re doing. And now that I’ve been here so long, for the first time, I’m stealing ideas from the French.
But some things just don’t work cross-culturally. We had an example of an email we sent out, and there was a misunderstanding when we were talking about the World Cup and we said “white nights” [the term for days when the sun barely sets]. Someone in France on our internal discussion board said, “nuit blanche” means sleepless night, so they thought the English was a mistake. The term “white nights” as we know it doesn’t really happen in French — so yes, there are certain things that are just never going to exactly translate. So being savvy about all languages is really helpful, even if I’m not fluent in all — which I am not.
Usually five. English, French, German, Italian, and Spanish. Sometimes Japanese and Chinese. But our B2C website, our eCommerce website, doesn’t support Japanese or Chinese.
With the CRM newsletter, we do think through what content works translated. We do translate sometimes, but it’s not like all of our blog content is simply translated and replicated. We have specific stories, for example, for the French and German market. So, if we do a thematic newsletter on sleep, say, it won’t just mean translating these articles across the board, there will also be localized content that gets folded in.
I would say it’s hard work and planning on the CRM side, but I would be lying if I didn’t say there’s just a lot of racing to the finish line to get it right.
Sometimes it’s strange to work across the time change, because we [Mikael & Raphael] work on a brief, we share it by the end of our day, and Susie has time to work on it. Then when we come back in the morning, we have most of the work done.
In terms of tools, we tried using Microsoft360, stuff like that. But it didn’t work as well, and now we are coming back Google Docs. We have tried every tool. There is nothing better for content than Google Docs because we clearly see copy and suggest notes, and we leave each other notes day and night. There is nothing that works that well.
It’s so important to have simple tools — it lets us have a clear brief and be on both sides of the Atlantic and work on the same document, at the same time, sometimes during the night… It’s so helpful.
You have to go straight to the benefits for the user — when you have a solution or a device that you want to sell to a user, you have to explain right away what the benefit is for them. You can’t just turn around and say, “it’s a great product, it has a lot of cool features.” Sure, it’s great, but what is the benefit to the user? Because if the end user doesn’t see the benefits from it, they probably won’t buy your product, or buy your solution, or buy your program, or whatever you want to sell.
Using our data and guiding people to understand everything the product can do — is very hard in practice. We like to come at it from the perspective of: there’s tons of health content out there, but you can flatline on being told the same things again and again. You should eat healthy, you should exercise more — these things are true and they have a big impact on health. But the question is how to come at those topics from a point of reality in a way that’s engaging.
I [Susie] like to go to the people who are not necessarily the hard bodies, the people who have proven that just by changing their life a little bit each day, they can achieve great things. That to me is exciting, because I think in the fitness and health space, you’re going to see a lot of sweaty people with perfect abs. That’s not me, and that’s not most people I know.
In addition the health aspect, one of the biggest challenges in connected devices is how you get people to continue using your product after a couple of months, couple of years — how to keep them engaged. We’ve worked hard to have very powerful engagement, content, and life cycles. The big one is to make sure the user knows how to use this product. After a couple of weeks, we teach them how to activate some very cool features we know will help them use this product in their own way.
The team was very reactive on social with heart rate during the World Cup final — opting in 7,000 users in France for continuous heart rate during the World Cup, seeing the actual heart rate rise with every goal that France took all over the country. This initiative was very interesting because it was a way to engage our users.
And that’s a way to say to people, hey, you own a cool product! It’s a funny way to use it today, because there’s the final World Cup final, but also it’s a way for us to remind them they have this cool feature. Maybe they’re not using it all the time, but that could be the right occasion to discover it and to see what the benefits are.
If we decide to use the app, it’s not the best channel for deliverability. Not as strong as email. We have less impact, generally with push notifications. But it worked well for the World Cup social campaign since we were targeting mostly active users and people who are using our watch everyday. So it made sense to target them through that channel.
I think the big advantage we have is connected devices. So we really have the ability to personalize, to do content with users, for them. With the data we can leverage through the app and other places, we can create what I would say is our most impactful campaigns, which is the weekly report. It’s very simply made, but just this email is seen by 600,000 users a week.
It has the biggest open rates, and they love to see the data, the trends. It’s something that is an add-on to the app, but it shows that the email is a very profitable channel to leverage. The weekly report is very efficient. People love it. It’s a very powerful tool, and because it’s by email, it has also enabled us to re-engage users within the app.
I think thousands. There are variables like your current weight, the trends, etc. that it displays. But then we also have an introduction paragraph that speaks about your data, and this part is the most complex because it’s all integrated in the email. Based on your data, we randomly show you one piece of copy that’s written several ways to describe the same thing. So, it creates paragraphs that explain to you what your week looked like based on your data. We have an internal CRM tool to send these emails, and we developed an internal tool to make the thousands of customizations possible.
We try always to be positive. If you have gained weight and you wanted to lose weight, we aren’t just saying, “last week was a bad week.” Instead, we try to encourage people and say “okay, don’t worry. If it’s below one kilogram, it’s not significant.” Or, “forget about it, and get active this week.” For people who succeed in reaching their goals after that, we just celebrate it.
We also listen to social from a content perspective. When Susie first came on, a lot of people were saying they thought the app was being too hard on them, bumming them out… I [Susie] had my brother in Los Angeles saying, “I think the app called me fat!” And I’m like, “the app did not call you fat. You are fat, but the app didn’t call you fat!”
So we also tweak content based on user feedback. Nowadays, everyone is so used to personalized emails that you have to go that extra mile to make it special and wake people up. I think listening on social and listening to your users from a content perspective is super important.
One of our most successful initiatives was the Healthy New Year back in January. We identified that in January especially, a lot of people want to lose weight, but they also want to keep it off. An idea we came up with was to have a 10-week New Year life cycle. So, for 10 weeks, we sent people daily tips and all the metrics you can improve when you want to lose weight. You have to take care of nutrition, you have to take care of sleep, you have to be more active.
The emails people received were a mix between the weekly reports and some weekly missions (like 10,000 steps). Every week there was one theme with two or three broad articles correlated to the theme. The idea was to make people discover all the possibilities — like walking with a partner or stepping on a scale every day or two to make sure you tracked your progress. In the app we also featured the content in the timeline. So they could read the email, or app users have the same experiences with the same content.
Basically, during the 10 weeks, you would receive daily tips, good content articles, up-to-date tracking, some suggestions to track progress, etc. We also had a community challenge — 10,000 steps a week for all our subscribers. And we put it on social, too, with a Facebook group. People would share their screenshot with 10,000 steps, and we really loved that. The best part was when we received feedback from people after this 10 week program. They told us “yes, I managed to lose some weight, and now I’m trying to lose it for good!”
We really love the opportunity to do big data studies to help inspire people and say “look, even if you just walk a few more steps a day, you can lower your blood pressure. You’re going to sleep better if you get a little bit more activity. You’re going to sleep better if you lose a little bit more weight.”
Showing the correlations is really fun, and so is seeing people sharing and using it all around the world. When people are trying to walk 10,000 steps a day for a week, seeing them share their pictures was so inspiring. Here they are in Scandinavia, here they are in Germany, here they are everywhere, sharing beautiful pictures of getting outdoors, and for me, that was really inspiring.
There’s one dedicated sequence for each product. I think it’s five emails. We have different sequences, so the first one is welcome, and we have three different emails. First one is rediscover your product, what are the cool features you can activate from the app. Then you have two other emails that are really good at speaking about themes that are linked to our products. For example, if you have a scale, you are going to have an article about weight management, and you’re going to have an article about nutrition. So, things that are really correlated with your experience.
That’s kind of the onboarding sequence: you have welcome, you have engagement, you have satisfaction after that. So, two weeks after you start with the product, you will receive a satisfaction survey to see how it’s going. Are you satisfied with your product, is there anything that we could do to improve your experience? That’s a great tool for us to detect what are the features people are loving, and what are the features that people are not fully satisfied with. And then we can improve it with the product managers after that.
I think we have hundreds of automated emails. Something interesting is the more you interact with us — through the app, through our emails, social stuff — the more you receive follow-ups. If you are identified with a satisfaction survey as a promoter of our devices, we will send you emails so you can share your feedback with others. We may even contact you to create testimonials.
An important one is we tried to improve our last smartwatch, and based on the satisfaction survey, we noted that the key improvement request was about notifications. So we have been working a few months now on extending notifications (right now you only have text messages and calls). [Update: After this interview Withings launched smart notifications for all apps.]
That points up another challenge that the CRM team has, which is that you buy your device, but then the app team makes your device do a lot more things than what it did when it was in that box. For example, a big one was automatic swim tracking. We often need to communicate with customers and be like, by the way, now your watch now also does this!
Then later it was “hey, we can recognize up to 30 activities. Your watch will know if you’re playing basketball or soccer. That wasn’t a thing when we taught you about your product!” So, trying to fold that into the original sequence, trying to educate people on how their product does a lot of new tricks now, that’s fun and that’s also challenging, because then you have to go again and tell the full story. Not just tell them how it works, but maybe have supporting content with people who do these different sports, or supporting content for these upcoming features. We’re always having a look ahead. We know it’s on the horizon, and we have to be ready with that content and communication, and sometimes they develop things really quickly, and you just have to be ready to support it.
We ran a test a couple of months ago to test what was the impact of the call to action in the email design. We tested two places just under the title of the email. We have the title introducing our latest product (for example Nokia Sleep) and then the product image. Or put the product image first and then the introduction under.
We learned that if the user already knows the product, putting the call to action in the first place will work better. But if the user doesn’t know the product, it won’t perform better.
Importantly, we know we aren’t going to send only one email, so we will have multiple touchpoints, but touchpoints are linked with the activity of the user — whether or not they clicked on the first message. We really adapt the communication. So this test was to try to separate the analysis on these two types of things. We try to maximize the visibility of the message at large but also try to customize the rest of the experience.
Our messages work like a funnel — if we have a launch email for everyone, we’ll send a relaunch with a different subject line to people that usually open our emails but didn’t this time. Then for all openers, we’ll send engagement content related to that product. And then for all the clickers, we’ll send something else.
RGE co-founder. Speaker. Lame marketing guy. If you found a typo on the site, it was probably my fault.
Love email & marketing news and wanna pwn your friends in email trivia? Of course you do.