For the past few years, Twitter has been telling TV programmers and advertisers that it can bring them more eyeballs, because people on Twitter spend a lot of time talking about TV, reading about TV and watching TV at the same time.
This argument is hard to prove, but it’s supposed to work particularly well for big live TV events.
And it has a pleasing logic to it, particularly for people who use Twitter a lot, and watch their streams fill up with TV chatter every time there’s a big game, award show or political debate on the tube. And it’s a very good story for Twitter to tell Wall Street, especially while it is still proving out its ad model.
So what happened last night, during the Grammy awards? Well, according to Nielsen, which has a Twitter-endorsed tracking system, Twitter activity was up — just a bit. And according to Nielsen, TV ratings were also up — just barely.
The numbers: Last night’s show generated 13.8 million tweets in the U.S. That’s up eight percent compared to last year. And last night’s CBS telecast drew 28.4 million people. That’s the show’s second-best showing, but only up one percent from 2013.
Is that enough data to really draw any conclusions? Nope! If you’re Twitter, which worked hard to integrate itself directly into the show (hence lots of cutaways to stars playing around backstage with a Twitter mirror/camera/tweeting machine set up), you might still argue that you helped keep attention focused on a pretty meh show. And you might also note that the much-coveted 18-to-34-year-old demographic was up 10 percent last night.
But it’s still really hard to make too much of these numbers, since they bounce around from event to event and year to year. In some cases, we’ve even seen Twitter numbers increase while ratings declined.
Well, it’s not really sure what to say this time. Facebook says that 6.3 million people had 13.5 million “interactions” with the Grammys last night, which I think they are defining as generating a post, “Like” or comment on the service. But we can’t compare that last year, because last year Facebook only provided a “reach” number of 43 million.
Obviously, those numbers aren’t apples and apples. I asked Facebook to explain what changed, and got a whaddayagonnado, via a spokesperson’s email: “Our methodology has evolved over the last year, so the timeframes and keywords methodology from the past doesn’t exactly mirror what we use today.”
Translation: Come back next year. Maybe we can talk then.
Meantime, we can get ready for new data in a week, after the Super Bowl. A reminder: Last year, social metrics were way up, and ratings were up a bit.
This article originally appeared on Recode.net.