If you're looking to increase video plays, Melissa Worden points out some valuable statistics provided by the head of BBC Interactive.

Pete Clifton said that when video was embedded into the article page, up to 40 percent of people reading the story also played the video. When users had to click a link to watch the video, only about 2 percent of people actually played it.

That's a huge difference. At first glance, we should all run out and jam video into the top of article pages, where the BBC runs it. But Clifton has a qualifying statement:

What irritates the hell out of people is if they click a story which says 'Britain buys 100 new tanks for the war in Afghanistan' they then click on the video and it's just a bloke standing in Whitehall saying 'they’re going to buy 100 new tanks for the war in Afghanistan'. The viewer could say 'you've wasted my time.'

We have done a lot of that. We have put up hundreds of pieces of video on the news site and too often they have replicated what the story has already said.

We should think more about what that page does in the round and come up with a piece of video that absolutely complements the text.

So the next question becomes, who is going to record all of this complementary video? Easier said than done. With a TV news station, videographers could be required to bring back extra footage that won't air as part of their story package.

Without a TV station, turn to your print photographers. Arm them with cameras. Train them. And then require the photo editor to return video with most stories. (Notice I said the photo editor, not the photographers.)

It's true, you're not going to get the best possible footage. When something big is happening, the print photogs are going to have their regular camera in hand. But for a moment, they'll put it down and tape something that complements the story.

Or, hire a bunch of Web-only videographers . . . Good luck getting that approved.

Related story: What the BBC learned from video experiments, found via Martin Stabe