Poetic justice – the entwined nature of paid, owned and earned

We’re always up for a bit of good old segmentation, us media types. Especially when it breaks down into a groovy acronym or a new set of ‘business bingo’ buzzwords. So, what’s not to like about POE – the segmentation of media into paid, owned and earned?

Of course, there is a fine line between segmentation and silo thinking, and this neat approach to classifying media based on a paid-free and owned-mediated matrix encourages us to see these three categories as separate entities, with their own unique benefits and contributions to the bottom line.

The implicit argument behind the segmentation has always been that advertisers should try to reduce the contribution of paid media investments and increase the role of owned or earned channels. After all, they don’t cost as much (do they?) and they have greater credibility amongst consumers.

First of all, a couple of caveats to the Thinkbox research. It is not exactly representative of the entire advertising inventory – with 36 brands representing just three sectors; retail (6 brands), finance (8 brands) and drink (with a whopping 22 brands). I know from painful experience, though, how difficult it is to get a wider range of advertisers participating in a study like this.

Secondly, this kind of analysis is always a bit ‘black box’, especially to non-statisticians, and so is not subject to the normal scrutiny we would expect to provide for consumer research. And, finally, my own frustration with this type of research is that it provides outcomes and identifies relationships, but it falls short on actionable insight, apart from ‘spend more on medium A’ or ‘spend less on activity B’.

That said, the POEtic research is based on an impressive database covering word of mouth (via Keller-Fay’s ‘Talktrack’ research), brand buzz (YouGov’s ‘Brandwatch’ data) and investment and response data from the brands themselves.

This provides a cumulative analysis base of more than half a million data points. As such, we should certainly take its conclusions seriously.

The main conclusion appears to be that you get what you pay for!

Owned media – measured by website traffic – is heavily dependent on marketing activity; almost half of all traffic is generated in this way. Not surprisingly, paid media dominate the influence – estimated to generate more than two thirds of all additional traffic.

TV by itself generated almost half (47%) of all additional traffic, followed by non-TV paid media (22%), then quite a gap to ‘earned offline’, ‘brand PR & events’, ‘earned online’ and ‘owned’.

Regarding earned media – which is often measured by positive word of mouth conversations about the brand – the results are pretty startling. An econometric analysis of brand conversation and buzz metrics suggests that around 40% of brand WOM is generated by media activity – the vast majority paid-for media.

Without that constant boost to their buzzability, many brand conversations would possibly wither on the vine.

The media-generated activity is dominated by television, reinforcing the traditionally strong correlations Brandwatch has recorded between TV campaigns and buzz metrics. Overall, TV advertising directly accounted for more than a fifth (22%) of all brand conversations across the sampled brands and more than 50% of all incremental conversation. (Other paid media combined accounted for less than half the level and brand PR even less than that).

TV also has a significantly stronger carry-over effect, meaning that brand conversations decline much more slowly after the advertising campaign finishes. Anybody familiar with the concept of adstock will know that TV often retains its impact for much longer; indeed, the PriceWaterhouse Coopers payback study from a few years back demonstrated that TV advertising generates almost as much sales impact a year after the campaign ceases as when it was in full flow – something no other media can match.

In fact, I find it fascinating that the carry-over into online WOM via TV is four times higher than for the equivalent level of online spend.

That one fact sums up this research; paid, owned and earned are highly interdependent, with paid media (especially TV advertising) having a disproportionate influence on the so-called ‘free’ segments of earned and owned media, at least if website traffic and word of mouth are a good barometer.

We know from past experience that most other earned/owned media metrics move in the same direction (take search as an example). In some of the most creative campaigns recently, there would be precious little owned and earned media traffic without the effectiveness of paid media.

In fact, maybe we should think of owned and earned media as a means of maximising the effectiveness of paid media, rather than pitching them as viable alternatives.