Insights

Online Advertising for the Blockbuster Movie Season
Article

Online Advertising for the Blockbuster Movie Season

Jon Gibs

We’ve spent a good deal of time at Nielsen lamenting some rough news in online advertising. Our estimates suggest that overall spending on display advertising has recently declined. And even the more optimistic estimates from the IAB, which account for additional interactive advertising segments, including search, suggest only a slight up-tick, 2%, on a quarter-over-quarter basis. Although the Internet is not suffering like print media , let’s just say we’ve all been through better times over the past couple of years.

With that said, let’s talk about a bright spot. Historically, the late spring and early summer have been the time when major films get hyped for the summer blockbuster movie season. Previously, this hype has been the providence of TV and to some degree print. What we saw for the first time this year was that the promotion of films moved to the Internet. There are a lot of good reasons for this – where else can you show a 4 minute preview and then have a person look up a nearby theater? More importantly, however, social media and the ability to really make films viral (a key component of buzz) appears to be a key driver, and one we will certainly look into next spring. Enough with the talk, let’s get to the numbers:

Through the spring, and peaking in June, we saw increases of 50% to almost 250% year-over-year. The top film advertisers for May and June were:

Clearly Regency (“What Happens In Vegas” specifically) was a big force here and it remains to be seen if they will continue to use the Internet, or how effective they found the medium, but this might be some good news during an otherwise glum year.

The thing is, the true test would be not just an impression table like above, but rather something like this, filled in of course:

Now we’re getting somewhere. This would 1) show us the real reach and frequency of an online push, and more importantly 2) begin to value each of the line items differently by showing how each modeled against actual results. We should assume that each of these has a different value; for example, Buzz is likely a much more powerful outlet on an “impression” by “impression” basis than display. If one person I trust recommends a movie, it seems like it would be immeasurably more impactful then a single banner ad.

Through using our new BuzzReach methodology, we can treat Buzz like ad impressions and through tagging we can pick up video, so this is now something we can do. Anyone want to give it a shot?