TopRank is fortunate to have an excellent team of consultants, many of whom contribute here at Online Marketing Blog. It is therefore, a rare occasion that we invite guest posts. The globalization of search and to follow, search marketing, brings a tremendous need for quality information and best practices. Search Engine Optimization for North American English can be a challenge on it’s own, but what about SEO for 5 languages and 12 countries? Or 30?
That’s why I’ve invited Andy Atkins-Krüger, CEO of UK based WebCertain, a specialist agency in multilingual search marketing to provide a list of the most common challenges companies face when embarking on international SEO efforts. Avoiding the big mistakes can save companies time, money and embarrassment.
After more than a decade of working in international SEO, it would be true to say that many of the same issues present themselves time and again. Lee kindly invited me to describe the commonest of these to help marketers take positive avoidance steps, so here they are.
1. Translating keywords is by far the most dangerous trap of all in international SEO rather than the technical hosting issues or the cultural risks – not appreciating that ‘keywords’ cannot be translated is rule number one. If you’re not a linguist, this can be a difficult concept to appreciate but the fact is that ‘keywords’ are convenience words – not really normal words – created by people to help them search and then responded to by search marketers.
So for instance, let’s take ‘car insurance’ by way of example. The correct translation of this into French would be ‘assurance voiture’ where ‘car’ equals ‘voiture’ and ‘assurance’ equals ‘insurance’ which does see a small number of searches. However, most search volume is at ‘assurance auto’ where ‘auto’ is an abbreviated form of ‘automobile’. French searchers and speakers have simply adopted this phrase out of convenience. The translation simply goes to the wrong place. This happens in all languages including English.
The solution to this is in fact, very simple. You simply recreate the keywords in the target language exactly the same way you would do in English. What that means is using a native-speaker of the target language – who is also trained in search marketing – researches them from scratch. That’s why we employ some 45 nationalities within WebCertain!
2. Not giving consideration as to how you’ll manage content when multiple languages are involved is a particular blind spot to many causing some nasty budget surprises. Worse, many organizations (the larger are more guilty) will invest considerable sums of money in producing truly FABULOUS English content – and then hand it over to the localization team with little budget and no thought for its SEO value.
A better strategy is to build your English content with localization or translation in mind. In other words, the copywriter’s brief should be to create the content without in-jokes or cultural references that a translator will simply not be able to translate. And bear in mind that fresh copywriting in each new language will be significantly more expensive than using translation – although a good option is also to mix fresh copywriting on particular local subjects which warrant it and using localization for the rest. The ideal would be to work with an international search marketing company which can localize and optimize at one and the same time.
3. Believing that an associate or, worse the CEO’s nephew, has studied French and therefore would be able to make a good stab at the language is not going to fly. You need someone who learned the language at their mother’s knee ideally growing up in the country in order to have the degree of intuitive understanding that will be required.
4. Taking an agency’s international claims at face value is an understandable mistake. One agency who claims a vast team of people who can work in 40 languages intrigued me so I undertook a detailed credit check. They happened to be based in a country where it’s a requirement to declare the number of employees within accounts and they employed just 3 people – which makes roughly 13 languages each. And if you see organizations which offer more than 40 languages then they’re almost certainly sub-contracting to translators (which you definitely don’t want) because there are only 42 languages including several versions of English which are regularly targeted within international search.
5. Choosing new target countries based on existing analytics is a good idea to support your export initiative, but is not the best way to decide in which languages to roll out your new web site additions. For instance, if you sell Supergizmos to the Dutch on your existing web site then what a good idea to try and expand those sales by checking on what the Dutch are looking for. You may find that they actually search for ‘supergizmos’ in English because that’s how they most easily expect to find them. Perhaps some additional support via paid search targeting the Netherlands would be a good idea. But localizing your web site into Dutch would target the same people who are already buying and may not increase their propensity to buy.
Meanwhile, some keyword research might reveal that the Italians (who fanatically buy supergizmos) are not using your web site at all – so an Italian language web site would incrementally add to your sales in the way that adding Dutch would not. Your analytics are never going to tell you this.
6. Finding excuses to run with a dot com – and not using local domains is very common. It would be true to say that I have made it something of a personal mission to promote the local domains – especially in the US – with some success. A number of proponents of the dot com have changed their minds after looking more seriously at the problem. Local domains are better for SEO because they give the best geographic information to the search engines AND users prefer them AND people you want to link to your site also want to link to local sites.
If the dot com decision is a policy decision and outside your control – fine we can find some workarounds. But take note of that word ‘workaround’ – that’s not what we ideally want right?
7. Not getting local links or hosting is a major handicap for many international sites – these do make a difference – though aren’t quite so critical if you have local domains.
8. Launching new countries e.g. Ireland, without thinking of the impact on the old site has seriously hampered some organizations success. Duplication on international sites remains a major issue – particularly for the world languages such as English, Spanish or French. It is very common to find that a site which has recently fallen in terms of performance in the UK, had just had a duplicate copy of the site provided to Ireland or Australia and because it wasn’t on anyone’s radar – no one realized what the consequences would be. There are many different ways to solve this issue – including sacrificing smaller sites to protect the larger ones (do you really need to rank well or Angola or is Portugal more the target?).
9. Responding to cultural differences is key – but this is only really good marketing. I get a little tired of all the stories about the ‘Pinto’ in Brazil being mixed up with part of the male body. In fact there are a great many famous Brazilians with that name as a surname, so I asked one of my Brazilian colleagues how come and he said “They have a very hard time at school! These make great stories – but the truth about culture is that countries reveal it in what they search for. Good keyword research can be used not just to improve the performance of your site generally but to understand how your potential customers are thinking and which products might be the best ones to target selling to them via that web site.
10. Lack of research is the nub of the problem. Few people have time to undertake really thorough research to most effectively power up their global export or marketing programs. The best trick is to find an effective international search marketing agency as they will have all the tools you need.