Written by Internet

Telling the Truth About Ages

Back in 2012, James Cridland wrote a very good piece he called Truth in Numbers, which examined how Facebook marketed itself. He showed that while the Office of National Statistics showed there to be 7,482,000 16-24 year olds in the UK, Facebook was somehow selling access to 9,155,804 16-24 year olds.

I was curious to update these figures and look a little deeper across Facebook. It seems clear that while Facebook is clearly pretty popular amongst all age groups, it still dominates in younger groups.

So I decided to plot Facebook users that I can advertise against using James’ method, against the most recent ONS figures I can find – 2014 estimates.

A few notes on this chart:

  • Facebook only allows children to open accounts when they reach the age of 13. Therefore in the 10-14 category, they’re massively understated. But the chart does look a little odd. Only a tiny fraction of the audience seems to have an account. Either they already have an account (see below), or just aren’t really interested until hormones kick in as they get a little older.
  • Once you get to the 15-19, Facebook suddenly has over 100% of all people in that age group. Amongst 20-24s, Facebook reaches a remarkable 144% of adults in that category!
  • In reality, you would probably expect Facebook to reach a percentage in the high 90s, but there will always be people who don’t have an account.
  • But of course, just because a child is under 13, that doesn’t mean they don’t want to get onto Facebook. It seems likely that a lot of 10, 11 and 12 year olds over time have registered as being 13 or older just to get their accounts early. Peer pressure at school is probably enough to force this. If you have an email address, you can get an account.
  • And that skews the demographics going forward. At what point does someone “own up” to Facebook about their real age? Facebook lets you change your birthdate, but for the most part there doesn’t seem to be much incentive for correcting birth years.
  • And we must assume that there are fake accounts. There are a lot of people willing to sell you followers (and likes). We must assume that these are bots, operated through networks. And they’re likely to be pitched as advertiser-friendly younger demographics.
  • Once you get to 40, Facebook no longer claims to reach the entire adult population. I can attest to having friends in their forties who are not on Facebook. The numbers obviously slip as you move older – and this is despite your mum and perhaps your grandmother now being on the site.
  • And while 65+ looks bad, hover over the blue column because it’s way worse. I capped the chart at 6m on vertical axis. In fact there are 10.4m 65+s in the UK, of which only 26% are claimed to be reached by Facebook.
  • Facebook provides much more rounded numbers today compared to what it did when James ran his test, hence numbers to the nearest 100,000.

There may be other reasons why there seem to be so many UK Facebook users between 15 and 39, including use via proxies and so on. But it’s still a little disturbing that these numbers are being sold. But I guess that’s really just the tip of the iceberg in digital marketing!

On a separate note, there was an interesting piece on More or Less a couple of months ago. They reported that there is a significant imbalance between 16-17 boys and girls in Sweden, with 123 boys to every 100 girls, making it a greater imbalance than even China.

It seems to boil down to asylum seekers and Swedish rules which mean that if you’re under 18 and gain asylum, you have the right to bring your family into the country. If you’re 18 or over, you don’t get that right. That means that as an asylum seeker, you’re strongly incentivised to give you age as less than 18. Given that you probably arrived in Sweden without a birth certificate, who’s to know how old you really are? They don’t check, and in any case, they probably can’t.