• by Steve Smith , Staff Writer @popeyesm, June 6, 2013

As someone who only watches video occasionally on my smartphone, I had to get a reality check on this phenomenon from my most trusted source -- my 21-year-old daughter. “All the time,” she said when I asked if she views much video on her phone. “Mainly in apps. When I see something in Reddit that looks interesting I’ll usually watch it right away. And I am always using it to find game hints and walk-throughs.” Her summer project is going through the original PlayStation versions of "Tomb Raider." And her old man cannot remember exactly how he defeated those T-Rex bosses near the Great Wall of China in the second game.

But the point echoes and confirms something I wrote about yesterday regarding BuzzFeed seeing half of its video views now coming from devices. Today GigaOm reports that YouTube is seeing 40% of all of its video views in the U.S. coming from devices. On a worldwide basis it is now 25%.

And the money is following the usage in this case. Bloomberg reports that YouTube’s advertising sales from mobile devices have tripled in the last six months alone. Video going to devices may be responsible for about $350 million in revenue.

The mobile usage is coming especially from the app, Bloomberg reports. Citing Nielsen numbers, they report 70 million people in the U.S. using the app in March -- up 42% from a year ago.

For the most part, YouTube seems to be fairly polite in its use of pre-rolls on mobile. Most of them that I’ve seen allow you to skip, and many are short enough to endure. The inventory clearly is selling well, because almost all of the YouTube videos coming from trusted content sources have some pre-roll attached.

I am eager to see the ad effectiveness metrics around mobile video compared to online video. For many of us, the tortuous 30-second spot in front of a three-minute news report on the desktop is a good occasion to do a quick email check and not look at the ad. That is less of an option on devices, which command full attention. Theoretically at least, it seems to me that mobile video should be performing better on basic brand lift metrics like recognition and message Association, if not purchase intent.