using website header

Connected: Display Connected: Media Landscape Connected: Regional Connected: AV Consumer Surveys Connected: Direct LinkedIn LinkedIn logo icon Twitter Twitter logo icon Youtube Youtube logo icon Flickr Flickr logo icon Instagram Instagram logo icon Mail Mail icon Down arrow
Dominic Mills 

Mary, Mary, still quite contrary

Mary, Mary, still quite contrary

After the legendary and influential analyst, Mary Meeker, delivered her latest Internet trends report, Dominic Mills gives his views on what she got wrong, what she got right - and what should concern marketers the most.

It's that time of the year, when Mary Meeker, Queen of Silicon Valley (and with a day job at leading tech venture capitalists Kleiner Perkins Caufield Byers) offers up her annual state-of-the-digital-world address.

As with the Delphic oracle, we pour over her pronouncements, made last week, with a mix of awe and wonder. This year, perhaps, there is less to gasp at, but still some highly relevant nuggets for the media and marketing communities to get their teeth into.

You can see the full 164-slide deck here, and as a thought-provoking primer that opens our eyes to the wider world it's well worth a look. But for those with less time, here's a summary of the bits that concern us.

1. Data - we all know it's gold-dust, the new oil etc, and as Meeker points out, the explosion in connected devices and installed sensors will only increase the volume of data.

It already has: we're up to 4 zetabytes (slide 65), a year-on-year increase of 50% (a zetabyte, by the way, is 1 million petabytes, which in turn is 1 million gigabytes - in other words, quite a lot).

But here's the issue. According to IDC, about a third of this data is actually useful, but only 7% is tagged and 1% analysed. All of which suggests that, for all the myth peddling of Big Data, we have no idea if we're actually using the right bits.

It brings to mind a modern-day version of Lord Leverhulme's mantra: "I know a third of my data is useful, I just don't know which third. And I haven't even got round to looking at 99pc of it."

2. The trend towards mobile is irresistible (slide 9, below). Intriguingly, as a percentage of all web traffic, it's much higher in emerging markets (37% in Asia) than developing markets. Even so, in Europe, it doubled to 16% between May 2013 and 2014, and accounts for almost 20% of all traffic in North America.

Of course, she wouldn't be Mary without banging on about how mobile isn't getting its fair share of advertising - even though it grew 47% last year. As usual, she complains that mobile's share of revenue at 4% under-performs its share of time spent (at 20%) (slide 15) in the US, and has a go at print for outperforming (5% share of time spent versus a 19% share of revenue).

Slide 9

What she consistently fails to grasp, of course, is that this is an apples and pears (i.e. false) comparison. For reasons we all know, a mobile ad does not equate to a print one (or most other media, come to that) in terms of impact, scale or, as often as not, dwell time.

It's strange, too, that nowhere does Meeker feature OOH in her analysis, all the more so as it becomes a digital, geographically precise, medium. Maybe there are no poster sites in the bubble that is Silicon Valley.

3. She thinks there's still plenty of growth left in the tablet market (slide 7, below), with sales up 52% in 2013 and selling at a faster rate than PCs ever did. Nevertheless, while it may be a short-term blip, the latest figures show a slowing down in shipments. Nonetheless, global tablet penetration is only 6%, compared to 22% for smartphones.

Sloide 74. One of the big questions Meeker raises is the impact of narrow-cast messaging services like WhatsApp, Snapchat, or China's Tencent on marketing. Usage of WhatsApp, for example, is growing at a phenomenal rate: 50 billion messages a day, up 178% year-on-year; Snapchat now records 1.2 billion messages a day; in five years, they'll have over 1 billion users combined.

Their appeal is clear: they're instantaneous and personal. Naturally, this is something the advertising community will seek to tap into, and the services themselves will wish to monetise.

But how? And how do you avoid antagonising the user base, for whom one of the attractions of such services must surely be that they constitute an unimpaired environment? For media agencies, all sorts of issues suggest themselves: currency, metrics and our old favourites, reach and frequency.

5. Visual culture will dominate the social landscape soon, suggests Meeker, pointing to the rapid growth of Vine, Instagram, Snapchat and Pinterest (slide 38, below) as platforms for sharing pictures and video.

She calculates 1.8 billion images are uploaded and shared daily. Of course, advertisers are already tapping into these platforms, but the message here is that they're going to have to up their game (and their budgets) quickly.

slid 386. Renting is in, owning is out - as far as music is concerned, anyway - which explains why Apple is buying Beats. As Meeker points out, sales of digital music in the US fell 6% last year, while streaming grew 32% (slide 50). And while Spotify hangs its hat on the subscription model, competitors might naturally follow the ad-funded model.

7. Screen proliferation is growing, Meeker says, and linear TV is dying, replaced by internet TV accessible via on-demand apps (like BBC iPlayer, for which about 46% of requests come from mobile and tablet devices). Meeker is not the first US citizen to say this (e.g. Google's Eric Schmidt) and globally, as represented by screen time spent with TV versus desktop, smartphone and tablet, the figures back her up.

In all developing countries, and nearly all developed ones, time spent per day with smartphones alone outscores time spent with TV (in the US, it's 147 minutes with TV versus 151 with smartphones).

But not yet in the UK or France (despite high smartphone penetration): in the UK we spend 148 minutes with TV, versus 111 with smartphones (slide 96, below). Thinkbox, for one, will be pleased to see this.

slide 968. Don't forget China, says Meeker. Eight out of ten internet users are mobile, giving it a mobile user base of 500 million - more critical mass than any other country. Alibaba, Tencent and Baidu each have more than 400 million users.

As a result, we can expect rapid development of innovation and services such as mobile commerce. Tencent's WeChat service, for example, includes a sophisticated payment platform and Alibaba has launched an online savings and investment product that already has assets of $89 billion under management.

The question is: will innovation in China spread west? Over here, we're accustomed to thinking of GAFA dominance (Google, Apple, Facebook, Amazon), but they're not inviolable. Could there be a day when the likes of Tencent and Baidu bring their innovation to Europe or the US?

Don't laugh. The Chinese have already moved successfully into devices (Lenovo), cars (Volvo) and household electricals (Haier). So why not technology? Indeed, China's national state television company and Baidu entered ZenithOptimedia's top thirty global media owners for the first time last month, the first time any company from China has done so.

To get all the latest MediaTel Newsline updates follow us on Twitter.

Leave a comment

Thank you for your comment - a copy has now been sent to the Mediatel Newsline team who will review it shortly. Please note that the editor may edit your comment before publication.

Tess Alps, Chair, Thinkbox on 4 Jun 2014
“Thanks Dominic for the article and the clarification.

I'm sure she's a charming woman (never met her) but I have to say that Mary Meeker drives me nuts for several reasons.

Firstly, her reliance on claimed media behaviour, when there are much more sound sources of data around, is perverse. You're right that she must just use the data that she wants to believe is right. The same seems to be true of that other influential global source of (non-)wisdom, Emarketer, who thinks that if you take one really robust research source and 'blend' it with 14 other rubbish ones the answer you get will be more reliable.

Secondly, her insistence that time spent with a medium should match advertising investment is barking, and I'm afraid it's one that Sir Martin often spouts to justify his ambition to have 40% of WPP coming from 'digital' (why? it must be more profitable). Even though TV ad investment would increase hugely in the UK on that basis, it's still very, very wrong.

What all these techy people are doing is conflating/confusing the medium with the device and with the technology. Should watching 4OD on a tablet be called TV, or internet, or mobile or have a separate category all of its own called 'tablet'? It's all nonsense; search is search whatever tech or device you use, and TV is TV however delivered.”
Dominic Mills, Columnist, Dominic Mills Media Ltd on 3 Jun 2014
“Other than to correct an error, I rarely comment on something I've written. Jeremy Swinfen Green raises a good point about Meeker's methodology, which is also relevant to what she says about UK TV viewing habits (and possibly those in other countries).

She says UK TV viewing is 148 minutes a day. BARB, however, has it at 232 minutes for 2013. How to explain the discrepancy?

If you look carefully at the bottom of the slide, you'll see it's is based on claimed behaviour amongst 16-44 year-olds who have access to a TV and own a smartphone or tablet.

Why she couldn't have used BARB data or Nielsen in the US, heaven only knows. Maybe it wouldn't have produced the figures she wanted.

So, well, yes, hardly surprising that the figure understates overall viewing levels in the UK and, I daresay, elsewhere. As far as I know, the US daily average is something like 5 hours across all segments of the population. Clearly, if you ask 16-44 year-olds how much TV they 'think' they watch, the answer will be lower than the actuality.

I should have looked closer myself: I can only claim 'slide' and 'stat' fatigue.

Naughty Mary! Next time round, I'll read the small print on the slides.

Simon Daglish, Group Commercial Director, ITV on 3 Jun 2014
“Dominic What a brilliant piece of analysis and fantastically written, especially your view on data. A area where we are far to quick to accept at face value and the unquestionable algorithm and its worth in the media mix. Especially as we are discovering (not through the UK press) that 'most' of it is fraud !”
jeremy swinfen green, managing partner, mosoco on 2 Jun 2014
“A great analysis: thanks. One of the things that irritates me slightly is her failure to explain what she really means by mobile. For instance chart 9 gives "mobile" as a percentage of web use. But a mobile device used at home to access the web is not really different from a laptop used at home. At home my family can access the web from the TV, PC, various tablets, various consoles and various smartphones. Is the difference in the device used to access the web significant? Not particularly. What is significant is out of home/office access to the web as that access comes with different motivational drivers and different technical constraints. Wifi freed us to surf the web on our laptops while watching TV. Smartphones and 3/4 G free us to do that on our smartphones pretty much anywhere. Now (providing you can get a data connection) that is a huge change and needs proper analysis.”