The technology in question, which is similar to products like AOL's Lightningcast, gives Google the ability to serve video advertisements to partner sites in their AdSense program. Video ads will appear alongside (or instead of) existing text and image/Flash ads. However, since downloading a (potentially) massive video file and playing it while a web page loads would be both annoying and intrusive, these ads all feature a "click to play" button which starts the audio and video only after a user actively clicks on it. While some have speculated that this will reduce the impact of the otherwise powerful video ads, the NYT tells us that even before the advent of Google's video ads, "[a]dvertisers have been eager to buy the relatively limited supply of spaces for online commercials at prices that equal and sometimes exceed the rates charged by major networks, as measured by cost per thousand viewers".
According to the NYT, Google will sell video ads just like it sells image and Flash spots in its AdWords system: "either as a price for each time a user clicks on the ad to visit their site or a fixed fee to show the ad to 1,000 users". The cost per 1,000 viewers, or CPM, approach is especially interesting, since it invites direct comparison with TV ads. CPM is certainly a valid measurement tool, and has been the de-facto standard, particularly for TV and print ads, for quite a while now. However, comparing CPMs for two very different advertising media can be tricky, especially comparing the "shotgun" approach of TV advertising to the much more "laser-like" accuracy of web marketing. And in this case, it would actually be more like opt-in web marketing, which typically carries the highest of premiums.
Reading this news reminded me of an I, Cringely column from January, where Cringely speculates that Google is preparing a tour de force entry into the TV commercial market where they will deliver localized, targeted ads to individual viewers based on the massive computational power of their global database and affiliate arrangements with lots and lots of local TV stations. From his musings (and taken with a grain of salt, of course):
Google imagines a world where only single people see match.com ads, and people who can't drive see ads from taxi companies where others see Toyota campaigns. Where fraternities see ads for strip clubs, beer, Cancun weekends and LSAT prep courses, and only seniors (and their adult children) see ads for Alzheimer's drugs. What would be the value of that increased efficiency, capitalized into present dollars? Ten billion? Fifty billion? I say the value is $100 billion -- 25 percent of the total U.S. advertising market and 15 times Google's current size.Google is going to let the telco and cable companies burn their capital building out IP-TV, knowing that Google will still be the only game in town for the crux of the whole thing: the ability to show every viewer the specific ads that companies will pay the most to show him at that specific moment. What Google wants to do with these trailers is SERVE EVERY TV COMMERCIAL ON THE PLANET because only they will be able to do it efficiently. Only they will have the database that converts those IP addresses into sales leads, only they will have the servers and disk space close enough to the viewers to feed the ads. Only Google will have the chops to run a constant, real-time auction for the next ad every consumer is about to see, and then serve that ad at the moment the program goes to commercial.Suddenly, everybody can (and, really, must) advertise on TV, because it'll be so specific...and so dynamic. If you start shopping for a new WiFi access point in the morning, Google will know, and that night when you watch Two and a Half Men, your ads will be from D-Link, Linksys and Belkin. And, further, they'll know that an intelligent buyer lives at your IP, so your ads won't show you a hot model demonstrating how they're plug-and-play, but will instead feature a quick recommendation from the SveaSoft guy about which AP's the best one for hotrodding.The potential value of such fine-grained control is massive, and even if we are a decade or more away from such a possibility (if it's even possible at all), it's this kind of thinking that makes people sit up and notice when Google introduces something as apparently innocuous as click-to-play video ads on a limited number of affiliate sites.
For me, the really interesting (and relevant) question is: what steps does Google take on their way to controlling the TV commercial world? They already have ad-serving technology. They're a trusted vendor with a killer ad serving platform and an obvious technological advantage. They're flogging Google Maps (formerly Google Local) like it's going out of style. In my mind, all of this puts them in a great position to enter the place-based media industry. While the potential rewards aren't nearly as big as the national TV industry, it would make an excellent stepping stone towards an eventual entrance into TV by giving them the chance to try out essential new technologies like local video caching, content distribution and collecting statistics from the multiple disparate networks playing Google-served video ads. As for localization, it would also give Google a way to get inside all of those stores that they've only been able to mark up on Google Maps thus far. And of course, this is in addition to all of the regular benefits of using digital signage to reach consumers at the point-of-sale.
So as they continue to look for ways to supplement the income from their AdSense program, it seems to make sense that Google would use their expertise in targeted marketing to start offering place-based media services. With their technological skill and mountains of cash, Google could choose to alter the digital retailing landscape in many different ways, from building out a massive Google-owned and operated digital signage network (which I think is just about as likely as them buying up a bunch of retail stores), to offering an AdSense In-Store program for serving contextually-relevant video content to retail media networks owned by third parties (which actually makes a lot of sense to me). This would also have the interesting side effect of giving us near real-time visibility as to the "value" of screens placed in lots of different venues, which could potentially cause problems for others who make their living selling in-store media. Suddenly, stores with retail media networks would act just like websites in Google's network. Those venues participating in the AdSense (or StoreSense, or AdStore, or whatever) program would receive relevant ads at the time and position they choose. More desirable screen locations in store would naturally demand a higher premium. More relevant content would be automatically selected to play more frequently. And local, regional and national brands would compete for slots on an equal footing (which is another point that Cringely brings up in his Google/TV domination scheme). When you think about it, this would fit perfectly into their existing model, but instead of partner websites, they'd simply have partner retail networks. Where might they begin? I would look squarely at the current unsold inventory across larger US retail networks, especially those using signage software that allows for real-time content feeds from web URLs (as opposed to just pre-stored video or Flash content).
So that's my speculation, built atop somebody else's speculation, and taken in the context of an only tangentially-related news article. But in light of what we've heard about Yahoo! And MSN trying to get more involved with different advertising channels, and Google's quest for utter domination of the advertising industry, I don't think that any of these possibilities are too outlandish or out in left field. Another six months or so should tell us how well Google is doing with its video ad serving technology, and by then it will probably be obvious whether they're ready to leave the confines of the web browser to deliver content into real world venues. Many people are (legitimately) skeptical that Google will ever be able to make such a transition. From my point of view, it's just a matter of time before "AdSense for Stores" hits the marketplace.