In my last post (Infographic Clean-up as Link Building Outreach) I mentioned that I’d be experimenting with various approaches to acquiring infographic links from websites that publish in languages other than English. This post is an update on my success to date and how you might replicate it to squeeze even more juice from those discarded infographics.
This process has (so far) gathered us and our clients additional international links from China, Japan, Ukraine, Russia, Spain, Portugal, France, Germany, Italy, Poland, Romania, Brazil and Mexico. This isn’t all that surprising considering infographics naturally gain placement all over the world, but all too often we dismiss these potential links as beyond our reach.
Previously, I said regarding non-English placements, “Whether you think they are worth pursuing is up to you, as the return on investment for your time won’t be as obviously beneficial”, which (ignoring the value we assign to incoming TLD diversity) was probably wrong. Just as a warning, I had mentally filed the method I’ll show you under ‘won’t work but test anyway’. It’s pretty brutal and you’ve probably already thought of it. Yes. http://translate.google.com/. There are plenty of reasons why this sounds like a terrible idea, and some of my decisions may puzzle you. Bear with me.
- Before we begin, you’ll need to find the sites with your content that you would appreciate a link from.
- If the site is in English, ignore the TLD and hosting and clean-up as normal.
- If the site is in Russian or any other non-English language, and you can find contact details (having Chrome translate as you go can assist), you’re in business.
- We aren’t just going to send them a translated email, but we are going to send them something fairly template-ish.
- As ever, the people who reply to your outreach will almost certainly give you or your client attribution, and a subset of those who don’t reply will attribute also. Keep those records!
This format is, for good reason, a little more restrictive than vanilla outreach. Here’s a rather ugly example (transcribed below) that can be made pretty and tailored to your needs:
Sorry for the English. I am glad you liked my client’s work “Vertical image with text and numbers”. I would be grateful if you could attribute them for producing it:
It was originally placed on http://originalplacement.com/infographics/originalinfographicdonotsteal/
Thank you very much; I look forward to hearing from you.
Past the obvious “include a translation”, I’ve applied a few tweaks to what you might consider an ordinary approach. I’ll break these down and offer a little commentary on their importance.
The URL of content as the subject line:
This may be cryptic but it will, all things being equal, significantly improve your conversion rate. It’s non-aggressive and catches their interest. It also helps them find what you’re talking about (which I repeat in the body of the email for their ease).
An apology for the mother tongue:
I do this right at the beginning. It acknowledges that I know how absurd and non-ideal this method of contact is. It’s polite, too.
The language used is pretty simple:
This is probably the most important aspect. You can’t assume that Google Translate is of the same level of quality across languages (it isn’t). You can avoid compounding the obscurity this entails by introducing neither complexity nor indeterminacy.
The aim is to avoid making Google Translate, or your readers, work more than necessary. In the same vein, you should aim to keep it short. Ideally, you want some of the translation above the fold if at all possible.
Inclusion of original placement:
This step is optional, and whether it merits inclusion varies with placement. Mostly, this step acts as proof that you or your ought to be attributed, which is helpful in cases where they have already attributed someone else.
The English is before the non-English:
This is counter-intuitive, but it somehow works much better than the other way round. My hypothesis, and the reason I recommend translation after the initial English, is because the translations will always be inaccurate and come across disjointed, pretty much like the seo spam mail webmasters receive daily. Given that my email account has “seo” in it, this could throw one manual spam flag too many.
One benefit of this tweak is that it can pique interest. It’s certainly different. My sample size here would be too small to attribute significance to, but for what anecdotal evidence is worth, including a translation (however poor) helps. My guess would be it comes across as less rude, showing you’ve made some effort to accommodate them and that you don’t have a full on Team America mentality. As for method, that’s pretty much it.
Some issues may crop up whilst you’re out there. One I’ve mentioned before is ‘fan-translations’, watermarked versions that attribute someone else as the creator. I don’t tend to worry too much when contacting the creators of these, as they tend to feel a little embarrassed by the contact and will be willing to help out so long as you aren’t too aggressive. One way that I’ve found that skirts the around issue is suppressing your outrage reflex (mine is now dead) and offering the original version to host side by side with theirs. This has the added benefit of reducing the visual/layout ‘staleness’ of the page.
If you’re more serious about this than me, you could hire or bribe fluent speakers to translate your emails for you. I haven’t done this, because I’ve had email chains six emails long, and I don’t really want to bring in an intermediary for every stage, but you may be more flexible (or come up with an elegant solution).
Lots of webmasters speak English. In addition, all of the responses I’ve received have been in English. It might be reasoned that the translated part of the email I send isn’t load bearing. Anecdotal testing would suggest it is, and common sense suggests it shows you’re at least somewhat considerate, and probably aren’t spamming. However, a little under half of these responses have been in, from what I can tell, Google translated English. My gut feel is that I wouldn’t have been as successful with this subset had I gone without the appended translation.
I’m hesitant to play around with this too approach too much, since it works so well for me, but it’s worth noting that I have the mentality of not always wholeheartedly testing things I incorrectly believe hold too much opportunity cost. The lion’s share of benefit, however, comes from the contacts you gain from your outreach.
There are definitely possible solutions less hideous than mine, but I’m yet to try one as easy to implement or with as high a success rate. Any suggestions?