More eyeballs on more sites does not mean more customers
Reducing the number of sites your ad displays on from 400,000 to 5,000 wouldn't make a single dent in performance. Path 51's Simon Larcey explains why.
A good media brand is like a trusted friend for a marketer or media buyer – you’ve got a level of established trust that means you can rely on them to get the results you need, and you can believe they aren’t going to steer you wrong.
To continue the analogy, an unknown media destination is a stranger. That’s not to say they’re necessarily no good, but since you have no connection to them, you’d be a fool to blindly consider anything they say without further investigation.
And yet, in these days of highly fractured audiences, we’re seeing far more media being bought from these ‘strangers’.
When planning media in the old-school days, before the ‘data revolution’, you would choose a magazine, programme, location or newspaper where you knew a majority of your target audience would be, and advertise there to achieve your goals.
Fast forward to the present, and that simple, basic strategy – which kept media owners in business for years – is out the window.
Now, it’s all about data, and those media owners who have not embraced data are struggling to stay afloat. In order to capitalise on this sea of numbers, media owners have lumped up multiple brands to create audience networks, hoping to attract advertisers driven by audience data.
It’s a case of delivering a million sets of eyes, but doing so by showing an ad on a million different sites if required.
But the thing is, while the ability to apply data like this may be impressive, it’s generally not necessary – and, in fact, comes with a notable drawback.
In March this year, JP Morgan Chase went from advertising across 400,000 sites per month to 5,000, with the bank’s chief marketing officer, Kristin Lemkau, telling the New York Times: “We haven’t seen any deterioration on our performance metrics.”
Going down to 5,000 wasn’t an arbitrary decision either – it was a number of websites that could ultimately be checked by a human.
By now, you’ll probably have drawn the line between Chase’s decision and the brand boycotting of YouTube that occurred in March and April, as a result of a number of companies seeing their ads appearing on videos using hate speech, or promoting terrorism or anti-Semitism.
Chase decided they wanted to whitelist their advertising, whereby, rather than make a list of what content and websites they didn’t want to see their company associated with – an all-but impossible task, given the billions of websites and videos out there – they drew up the 5,000 web addresses they were happy to be on.
And, in fact, it has seen such an improved overall result for the company as a result of being far more efficient that they have decided to reduce that number further, to 1,000 sites.
Which begs the question: just because a website isn’t on a blacklist, does that mean you should just blindly run your ads on it?
In a world where we see continued growth in digital advertising, why isn’t there more digital spend on trusted, well-known publishers?
Maybe we can’t buy the mass numbers we think we need, but surely it’s about results?
How about understanding the type of results we can expect from a particular media brand?
In a recent article, Mumbrella’s Tim Burrowes described beautifully what has happened in the magazine sector – in a nutshell, lumping everything together has not resulted in the success that was expected.
Magazines are successful because they target the passionate fans of a particular subject. Tim refers to GQ as a hugely successful online brand off the back of an equally successful offline brand.
It’s been the case for many magazines under the Conde Nast umbrella, the publishers of GQ. The century-old media company has had great success converting titles to an online proposition by simply keeping true to their brand and audience.
This may be obvious, but the success of a campaign is generally defined by the environment where the advertising appears. Chasing audiences using predominately inferred data is questionable, but targeting by environment will guarantee a large chunk of the advertising is seen by the type of person you are looking for. Overlay audience data on that site and you create an even more compelling proposition.
With the rise (and fall) of programmatic, we are starting to see more and more advertisers focus on private marketplace deals to ensure advertisers only appear on specific sites at specific costs to guarantee best results.
Which seems to show we have nearly come full circle, and that the power of a media brand is starting to become top priority again.
Therefore, advertising results will be better if you stick to trusted, well-known brands that hold strong relationships with their audiences and advertisers.
In short, while the occasional stranger may become a friend, more often than not you’re better sticking with someone you already trust.
Simon Larcey is managing director at Path 51.
Very well said.
User ID not verified.
“…the success of a campaign is generally defined by the environment where the advertising appears. Chasing audiences using predominately inferred data is questionable, but targeting by environment will guarantee a large chunk of the advertising is seen by the type of person you are looking for.”
Is this satire?
The placement of an ad is just another data point. You’re right that using inferred data is questionable and that’s exactly what you’re doing when you use the placement to define an audience.
User ID not verified.