It has been a few months now since Twitter rolled out its mobile ads, offering companies the chance to target its 55m mobile users.

The social network says Promoted Tweets allow brands to target their followers as well as users “that share similar interests with their existing followers”.

This sounds like a good product for marketers: ads based on location and the user’s interests, and you only pay if the user engages with the ad by clicking or retweeting.

However, the ads that have shown up in my timeline in the past two months suggest the targeting is not as accurate as Twitter would have us believe...

The Good

I should make it clear that I don’t currently follow any brands on Twitter.

Therefore all the Promoted Tweets I have seen must be based on the brands Twitter thinks I will be interested in.


While an ad for a marketing company isn’t useful for me, you can at least see some method in its targeting.

A lot of the content in my Twitter feed is related to digital marketing and a majority of people I follow work in the industry, so on that level the Eloqua ad is relevant.



As with Eloqua, I don’t actually need inbound marketing software but it’s easy to see why Twitter might think it was relevant to me.

Assemble Live and Prometheus

I tweet about movies and Lovefilm fairly regularly, so it makes sense that I would see these adverts.

SES Conference x 3

The SES marketing conference is relevant to me, however the three ads promoted in my timeline were for its Toronto event.

But I think the fault probably lies with SES for targeting its ads globally rather than in North America.

The Sunday Times

My biog says I work as a reporter and I follow several other journalists, so this tweet is relevant.

National Geographic

I assume I saw this for the same reasons as The Sunday Times tweet, however I have never engaged with National Geographic on Twitter and it isn't relevant to me.

The Bad


I used to be an O2 customer, until a dispute over broadband payments led me to tweet complaints about their service.

I’m not entirely sure whether Twitter uses brand name recognition in its targeting, but if it does it clearly doesn’t use sentiment analysis.


Friends Reunited x 2

I saw these two ads shortly after tweeting that Friends Reunited would soon be out of business.

As with the O2 tweet it seems that Twitter possibly targets ads simply based on brand name recognition.


The only reason I can think this is relevant is because I am male.

First Great Western and Chevrolet

Like every other person in the UK I occasionally travel by car and train so I suppose on that level it is relevant.

Call of Duty

I’m not a gamer, don’t tweet about games and have never engaged with a gaming brand on Twitter. I am a young male however, which could be the basis for targeting me.


Medal of Honour x 2

See Call of Duty summary.


Obviously this is just a tiny sample of the number of ads Twitter serves up, but it does raise the question of what data it actually uses to target Promoted Tweets?

The value in social advertising is supposed to be that the wealth of data they have on us allows for more effective targeting. In fact, Twitter says its targeting extends the reach of campaigns to users “who are receptive to your message”.

Yet in the case of O2 and Friends Reunited I am actively critical of their products, but still got served their ads.

Obviously targeting will never be an exact science and it is ridiculous to suggest that we should only see Promoted Tweets from companies we want to engage with.

And some of the ads I saw were definitely relevant to me based on the industry I work in.

But if some of Twitter’s targeting is no more complex than serving Promoted Tweets based on age and gender demographics then doesn’t that slightly undermine the supposed power of social advertising?

From the examples above it seems that some of Twitter’s methods are no more effective than simply pushing out banner ads to all users and hoping for the best.

David Moth

Published 1 May, 2012 by David Moth

David Moth is Editor and Head of Social at Econsultancy. You can follow him on Twitter or connect via LinkedIn

1719 more posts from this author

You might be interested in

Comments (3)

Graeme Benge

Graeme Benge, Digital Marketing Executive at Koozai

Interesting analysis David. To my mind Twitter by nature is a loud hailer and the best fit for promoted ads is in the search/trending function. Can't see at this stage the ROI in timeline placement. It's something that interests me though as I'm sure the platform will evolve.

over 6 years ago



Definitely interesting to analyse individual tweets. i have seen some very random examples as well. Thinking further, it's actually more true for people to mention certain brands on Twitter only in a negative context when complaining and other brands mostly in a positive context... so if it's TelComms I cannot imagine anyone thinking of tweeting - O2 is the best telephone provider i ever had... but more leisurely activities and hobbies will have unpromted praise - Love Frae yoghurt etc.

Possibly time for 02 and other Telcoms to get an associated entertainment platform and invest into that? Like Orange... it's almost not a telcomms provider anymore but a cinema booking service.

over 6 years ago

Nichola Finan

Nichola Finan, digital marketing at Media-input

Definitely suggests we approach 'targeting logic' in a new way, as your analysis has suggested, and the comments already posted have alluded to. This is no longer a linear kind of logic, but far more intuitive, which makes it interesting don't you think?!
Media planning more art than science, again!!

about 6 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.