Does Twitter truly reflect the UK population? Here’s what our Head of Digital Strategy, Matt Rhodes, has to say…


No, Twitter is not a reflection of the UK population. But it is still useful.

It has been much discussed, especially in the fall out from the UK’s vote to leave the EU in June this year, that what is discussed on Twitter does not seem to be reflected in discussions and decisions that take place offline. There is a danger of over-analysing this (and I have previously done so – using data to show the seeming centre-left bias of political discussions on Twitter) but the truth is actually quite simple. Twitter usage does not reflect the UK population.

A useful set of data released by Ipsos Mori in their Q3 2016 Tech Tracker report shows that overall only 19% of UK adults report having accessed Twitter in the last three months. And this is heavily concentrated in certain demographics:

  • People ages 15-24 are much more likely to be on Twitter
  • Those under 54 and in the AB social class are much more likely to be on Twitter
  • And men, in the AB social class and aged 15-34 are heavily over-represented on Twitter compared with the UK population

So Twitter is younger, more male and more affluent than the UK as a whole. It turns out it maybe is more representative of the ‘Metropolitan elite’ that has become something of a slur for those people who are more critical of the vote to leave the EU (and a view that was felt to be more prevalent on Twitter than off it).

Twitter chart

 Source: Ipsos MORI


So what does this mean? Well on one hand, it just gives us the information to better understand when and where to use the platform. For a key set of consumer types, Twitter still gives us access to a material proportion of the UK. But it is not a reflection of the UK population and when we are monitoring and analysing conversations and data from the platform we need to keep this in mind. Like with any good research, we need to understand and allow for the biases in the data before us. Maybe conversations on Twitter are more reflective of that younger, more male and more affluent audience. That does not make them useless (far from it) but it means we need to analyse what we see in this context.

By contrast, Facebook is much more a reflection of the UK population – 62% of the UK population report having visited the site in the last three months and they are well represented in pretty much every demographic of men split under 54 and of women under 64. Even for groups above these ages, a significant proportion of the population are using Facebook.

So this might make Facebook a more natural place for campaigns and activities where we are looking to reach larger, more mass-market audiences with paid media. But the quality and diversity of conversation on Facebook makes it less useful as a place for developing an understanding of what people are saying and how they are reacting to different events. It’s why the ‘trending topics’ on Twitter are a better and more current reflection of events than the same on Facebook, where discussion tends to lag behind actual events.

Facebook chart

 Source: Ipsos MORI


So what do we learn from this Ipsos data? Well it gives us a level of detail and understanding that is sometimes overlooked about exactly who is accessing different platforms. There is still a temptation to treat all social media as a single channel to customers and a single way to get your message across. Twitter, despite its problems with growth and finding a buyer, is still a place that is strong for some key consumer demographics in the UK. And it is still the best way to take a sense check of how people react to an event or discuss an issue (allowing for the biases we can identify). But Facebook is where mass market digital advertising is now at – and media plans and campaigns should be reflecting this.

Matt Rhodes

About Matt Rhodes

Head of Digital Strategy for work. Marathon runner and charity trustee for fun.

Follow Matt Rhodes