He also repeats a common error. Specifically, Pakman says:
“First, we spend 86% of mobile time in-app. The idea that the mobile web is a credible channel through which to reach consumers is largely disproven at this point.”
Nope. Mobile web and mobile in-app behaviour are not binary. When users are in the facebook app, they spend a tremendous amount of time accessing the mobile web through facebook’s own in-app browser. The same for twitter and others. We enter social apps for discovery and then access the mobile web while still in-app. It is a mistake to conflate time spent on the mobile web with time spent in a traditional browser.
This is why when media sites talk about the astonishing growth of mobile they are generally not talking about their own apps where traffic behaviour tends to show a loyal but small and slow-growing audience. Instead the traffic that is swiftly breaking the 50% of total traffic mark is mobile web traffic of which more comes from social sources than anywhere else, and most of that is in-app (and depressingly for twitter, by an order of magnitude mostly facebook).
A more valuable analysis of whether the mobile web is a credible channel to reach consumers would be to:
– Separate time spent within the facebook/twitter/whatsapp/etc mobile browsers from native in-app behaviour and then combine them with traditional browsers to get a true picture of the mobile web’s share of attention.
– Remove from the pie chart the categories where, by their nature, brands do not have a meaningful opportunity to reach consumers (utilities/productivity/non ad-supported gaming etc).
Then one could analyse what share of the available channels for brands to reach consumers on mobile the mobile web represents. I’m willing to bet Mr Pakman a good steak dinner at the restaurant of his choice that it will surpass the credibility bar quite comfortably.
Now all this may change. The mobile web experience is a shit show. Facebook is pushing hard with its instant articles, Snapchat is experimenting with Discover. However, today, to analyse the mobile web without accounting for in-app mobile-web browsing is about as useful as trying to understand national infidelity rates by sampling Ashley Madison users.