Does your website offer translated localized versions of content targeting different countries around the world? If so, you need to be aware of a few warning flags that may mean that your localized content isn't performing as well.
That’s what we're going to be discussing today with an international SEO consultant who prefers one-to-one discussions to hanging out at conferences. She's the co-founder and co-host of the SEO Nerd Switzerland community, and founder and host of the podcast and job board WorkinSEO. A warm welcome to the In Search SEO podcast, Isaline Muelhauser.
In this episode, Isaline will share four content localization red flags, including:
Unusually low KPIs for a translated language
Receiving traffic from another country that's not in your target market
Translating keyword research
When customer support receives an unusual amount of requests in a translated language
Isaline: Hi, David, thanks for having me.
D: It's great to have you on. You can find Isaline over at pilea.ch.
So Isaline, why is content localization such a challenge?
I: Mostly because people take a lot of time to create content and optimize the number one language, which I usually call the original language. And then they think that because they spent so much time optimizing that part of the website, it's going to be easy for the rest. But they forget that you can't fix that with hreflang tags and you're going to have to dig into the content.
D: Absolutely. And you can, of course, just translate. I'm reminded of a few stories from Michael Bonfils, who's really good at international SEO as well when he talks about that you need to really get someone locally in there who completely understands the idioms as well as the language itself. The informal phraseology that is likely to be used in another language as opposed to direct translations.
I: I 100% agree.
D: Well, today, you're sharing four warning flags to be aware of about content localization. Starting off with number one, KPIs for translated language are significantly lower.
1. Unusually Low KPIs for a Translated Language
I: This is one of the main reasons why people come to me. Most of the time, they don't say it that way. They say they need keyword research. But the money is the trigger. They come to me because they’re not making as much money as they expected when they launched a new language. And for the ones who come to me without having a clear view of the KPI of the translated language, I'm asking them to really have separate dashboards. For me, the first thing you should have is separate dashboards in Google Analytics for the different languages. And not only Google Analytics, but also Search Console. So Data Studio dashboards or whatever you're using, you should separate the languages. Also, in case the languages have several countries and you're targeting one country. So French, for France, also separate the country and filter the data.
Usually, this is the first flag where you see that your KPIs are significantly lower. And when I say significantly lower, also keep in mind that there is a proportion of the population. For example, in Switzerland, the French is 20% of the people. So when I say significantly lower, do not look at an absolute number, but filter the data and look at smart numbers, think about how big your target is. Because obviously if it's smaller than the original language, obviously it's going to be lower, but how significant is it?
D: Are you also talking about significantly lower from a conversion rates perspective? So if you just translate your site into another language, you should actually expect lower conversion rates?
I: Well, no. That's the thing. Very often people focus on translating the contents of the website, and they think about the articles and the contents on the page, but they forgot about the architecture contents. For instance, in English, you have lots of verbs and nouns that are interactive, and which are the same as search or look. And in French, or in German, these would be two different words. So if a human is not looking at the translation, you might end up having on your button the noun which is not interactive. Very often the button doesn't have this intention of the user doing an action because someone translated the contents, but did not translate the architecture of the button and the links. What I call the interaction language is missing.
You need to check that everything that is typically SEO works well, like in the Search Console, you have the right keywords indexed for the right pages, and so on. But also see if people actually converting, because that's the other thing. Sometimes you can have visibility, but you struggle to have the actual conversion. So you need to think about not only the content that is going to drive traffic, but also the content that is not driving traffic per se, but is helping to convert to get money in your company. And very often this content is forgotten.
So the buttons and everything that's interactive on a website are important. But also everything about the conditions of an activity, especially on eCommerce websites, the conditions and FAQ and these kinds of things, which do not obstruct traffic, they are not well translated, and people lack trust. If you read an FAQ and you don't understand what are the conditions, you are not converting. That's two things. Looking at the KPI and also not forgetting everything that's not properly SEO, i.e., not attracting traffic, but nevertheless very important for the business.
D: Okay, so a blend of UX here as well. So if I have an English-speaking website, and I want a presence in the French-speaking part of Switzerland, what would be the best way to go about doing it in a reasonably cost-efficient way, but also trying to ensure that the website actually converts once it's translated and localized? Are there specialist firms to go to? Is it sufficient to use an outsource website like Upwork to find someone locally in that country to check things over?
I: Definitely. You have four levels of work. You have the actual translation that can be done by robots or by a human. You can optimize the translation with an SEO by adding keywords and stuff. Then you have the copywriting step where you have an actual human localizing the language. Obviously, having a human copywriter doing the work is going to be expensive. The first thing you have to do is organize all of your URLs and content in order of priority. And you need a plan because you're probably not going to be able to localize everything at once. What you need is a plan over 12-20 months to know what you're doing, and not randomly localizing some parts of the website. The first thing you should do if you have an eCommerce site is to check what is the most important category for that market. If you have an eCommerce, not all products are going to have the same potential for return on investment for that market in terms of the competition you have in the country. So check what's the most important projects and start with the highest investment for this project. And you can do some lower investments, either with machine or translator for the other projects, and plan to do the localization for the other products in several steps. But that means that you can have a really good view of your expected return on investments and really manage the expectations of the stakeholders and telling them this is what you invested for this product, and that's what you can expect. And this is what you invested for these products so don't expect too much about this.
D: And number two of your warning flags is if a website receives traffic from another country that's not in your intended market. How do you identify that? And what do you do about it?
2. Receiving Traffic From Another Country That's Not in Your Target Market
I: Yeah, that's another reason why people contact me especially when you have two languages spoken in two countries, and the cost of the work is very different between the countries. We have this between France and Switzerland. It might be tempting to ask for a French person to work on the content for a Swiss website. But the thing is, very often, what happens is you attract traffic for French which is not your market because your price is set according to the Swiss market. See what I mean? It means that people are going to check your website, but they're not going to buy because there are cheaper alternatives in the country. That's a big flag that you can see also in your KPI. If you filter the country, in French, and you see that when you filter out the countries that you're not interested in, while you don't have that much traffic, that's a big red flag that your content is not localized.
D: And number three, you're asking to translate keyword research. Why is that an issue?
3. Translating Keyword Research
I: People asked me if I can translate this keyword research. And my answer is if you need to translate keyword research, ask a translator and export the data from Ahref. You don't need an SEO specialist to translate keyword research. But if you want to know about the opportunities in this country, then I can do keyword research and apply it to your topics and do a URL mapping for your website. So don't ask to translate, you don't need an SEO specialist to translate to keyword research. Because an SEO specialist is usually more expensive than a translator. So decide if you want an SEO to look at the opportunity that you can later apply to your topics, blog posts, and URL. Or if you want an SEO specialist to just find the best possible alternative for this URL. Or if you want a specialist to translate a content brief. But don't ask to translate the keyword search. The risk to just translate a keyword search… when I translate my process is different. When I translate, I look for a better alternative. And I provide an alternative, even though the search volume might not be interesting. And I leave on the sides all the topics that are interesting in my country, but they are not existing on the current website. When I do keyword research, I look for all opportunities. And I can flag a topic for Switzerland, which doesn't exist on your website yet. So it would be interesting for you to create content, even though this content is not meant for your original language. So the processes and the deliverable are very different. So I advise any SEO specialists to not agree on translating keyword research. I have seen that it creates issues in the client’s expectations because they check and they're like, "Oh, there was this equal opportunity. And you didn't tell me?” And I'm like, "Yes because you asked me to translate and I translated.
D: Absolutely. I guess translating is a massive potential issue because you're not necessarily identifying every single keyword phrase that’s relevant for that local market.
I: Exactly. And that's the thing I see very often in Switzerland, is that people tend to think that content should be the same in both languages. They tend to think that you have content in German, and you should translate the same in French. But actually, you can have similar but different content in both languages. There's no reason why you can't add a paragraph for the French audience in French, that is not translated in relevance in the German content. If your audience needs that information, just provide it if it's relevant to the topic, even though it's not existing in the original language. That's totally okay.
D: Good point. It’s not just for search engines, it’s for actual people as well. They’re much more important than search engines a lot of the time. And the fourth warning flag is where the customer support receives significantly more requests in a language compared with the original language.
4. When Customer Support Receives an Unusual Amount of Requests in a Translated Language
I: This is an often forgotten thing. If you only think of your KPIs on your dashboards, and you forget about the actual work of your teams internally, especially again for e-commerce… At the end of the day, the time your employee spends answering the same questions, it's time you invest in and time you pay for. So if you can reduce the time invested to answer the people and provide better information it is going to be a spring for your company.
In Switzerland, customer support receives lots of questions about Switzerland not actually being in Europe. Is it going to be the same conditions for the delivery and the return of the European countries or not? If you answer this question all the time, it means that something is missing in the FAQ or in the conditions and this can be another flag. It's not 100% SEO, but it's something I recommend to look at. At the end of the day, I'm an SEO consultant, but I'm also a consultant here to help you make more money.
D: Should every set of FAQs also have a search box above them? So if you're seeing people typing certain queries that aren't included in the FAQ, that's probably a good indication that you should be writing about that.
I: I'm not always a fan of the search box on websites, because very often people rely on Google to do the search within a website. And they're going to type into Google keywords like delivery, FAQ, and brand name. I think this is something you can also see in the Search Console if you have lots of requests for these types of keywords.
D: Good tip. Let's finish off with Pareto Pickle. Pareto says that you can get 80% of your results from 20% of your efforts. What’s one SEO activity that you would recommend that provides incredible results for modest levels of effort?
The Pareto Pickle - Prioritizing Your URLs
I: Prioritizing the URL. Meaning, strategize which URLs are most important, then you know exactly what you're doing. This will take you maybe two hours at the beginning of the project depending on how big the website is but it will save you a lot of time throughout the whole project.
D: Is this referring to canonicalization as well and deciding which version is canonicalized?
I: I mean prioritization for choosing if it's a basic translation, optimized translation, or copywriting translation.
D: Right. And would you do that by traffic? By projected ROI? Is there any way to actually decide on which URLs to spend the effort on to get the humans translating those pages?
I: I would base that on the opportunity of the market crossed with the most important projects of the company. It should be a project the company does well, but it should also be a project the market needs. There is a good fit of what you actually offer and what the demand is.
D: Well, I've been your host David Bain. Isaline, thank you so much for being on the In Search SEO podcast.
I: Thanks a lot, David, for having me. It was a pleasure. And congrats also for all the wonderful podcasts you have because I've seen lots of friends and community members of women in tech SEO. And that made me very happy to see.
D: Superb. And may it continue. And thank you for listening.