Email in a bilingual land: evidence that language of choice is more effective

Canada is a bilingual country. The federal government (and many national organizations) operate with stringent standards on services and communications being available in english and french.

This is done to protect people’s right to access government services in either english or french, whichever they prefer.

I get that. What I don’t get is why people doing online communication think that means we have to give everyone both languages in the same message.

For bilingually mandated organizations – governmental or non-governmental – it seems that being seen to be bilingual is actually more important than being understood in english and in french.

That’s fine for government. I suspect the current government would make its program information available in Latin if they thought they could get away with it and it would keep costs down.

But for social change organizations, whose health and success involves public engagement, sending email (for example) in bilingue, Canada’s third official language, is bad news.

See, no one speaks bilingual and they’d prefer not to read it in email either.

I’ve always thought this. But now I have proof. Some proof anyway.

Where I work, we have serious religion about producing everything in english and in french at the same time and with the same quality, etc. I wouldn’t have it any other way.

But someone interpreted that to mean that bulk emails must always contain french and english in the same message.

This means both languages in the subject header and either english column/french column or english-follows-french email body. Both these are fraught with problems:

• The second language in the subject is almost never read.
• Hansard style messages don’t render well in many email clients – Blackberries? Hello!
• Over-under messages must always put one language first.

And people who read both, or prefer to read in the language-put-second have more work to do and are less likely to open your message and less likely still to click.

As proof I offer the following.

We sent out a notice of an online survey (complete with juicy incentive) to a 35,000 or so member list. For some addresses we had a language preference. We sent them a message in the language of their choice.

For others, we did not, so we sent them a Hansard style ‘bilingual’ message. The results tell the story:

Language-of-choice emails got a 20 or 21 per cent higher open rate and a 16 per cent higher click rate than bilingual messages with the same content. They’re more likely to be forwarded and less likely to be the subject of a spam complaint.