The Digital Signage Insider

Earlier this month, the Online Publishers Association (OPA) released a study on the effectiveness of various types of ads within streaming video sites. The study's conclusions are interesting but not exactly unexpected, and by themselves aren't something I'd normally spend much time writing about. However, in the course of reading about them (and learning some of the ins-and-outs of the online advertising world), I was constantly reminded of the many ways that retail media differs from online, television and other formats. Spot placement, content type and even the use of creative assets vary dramatically between in-home and out-of-home spots, as well as between the real world and the virtual (online) world. However, the study does point out a few areas where we can learn a thing or two from our online video counterparts.

What's the most effective way to show video ads on the web?

This particular study focused on a few key aspects of online video ads:

• Placement of the ads: before, during or after the "main event" (often called pre-, mid- and post-roll ads, respectively)
• Duration of the ads: 15-second vs 30-second clips
• Source of the creative content: whether it was reused TV footage or custom-built to be a web ad
• Presence of other marketing messages: whether "companion ads" appeared on the web page at the same time as the video ad itself

The researchers studied consumer behavior based on three different goals: driving "breakthrough" (what we call the "awareness" stage in our marketing funnel diagram), changing people's opinions (perception), and persuading people (preference). To drive awareness, marketers found it was best to use 15- or 30-second pre-roll ads (again, those that come before the main content spot), and run companion ads elsewhere on the web page while the video ad is playing. To improve perception and change people's opinions, 30-second ads were found to be superior to 15-second ones, but no preference was given to location or content type (i.e. re-purposed TV commercials worked just as well as built-from-scratch web spots). To persuade people and change their preference, the best results were obtained with custom-built 30 second spots that leveraged the "halo" effect of the main event content.

Does retail media have much in common with online video ads?

The OPA's findings are interesting, but we can't apply them to our networks as-is, since virtually none of the content in the study would work in a retail setting. Why not? Well, take the goal of persuading a viewer to choose a brand or make a purchase decision. The online data clearly indicate that longer is better, and a 30-second spot will outperform a similar 15-second spot. But in the retail world, we've seen plenty of examples that contradict this. For example, I'm pretty sure that Virginia Cargill from SignStorey has indicated that longer spots work better for their advertisers. However, Axel Vera at Televisa (responsible for Wal-Mart Mexico's in-store TV network) believes that shorter is better, usually opting for spots around 7 seconds long (more on this topic in coming weeks). Both store environments are reasonably similar. Both content clips are intended to persuade. And both formats perform well.

Another problem arises when trying to convert the concept of pre-, mid- and post-roll advertising to the retail media world, since our digital signage networks don't really have a "roll" of any sort. Sure, there's plenty of interesting content to go around, with everything from news and weather to custom cooking/entertainment/fashion/hobby clips being thrown onto screens to improve customer satisfaction, but none of them can hope to draw viewer attention like an episode of CSI or American Idol does. In fact, unlike any TV show or online video, customers don't initiate the media consumption behavior themselves when in a store. Instead, the screens in these locations are always playing, always ready to take advantage of the few seconds of attention that the average shopper is going to spend on them. And whether they opt to watch or ignore the screens, consumers still can't choose what's being shown.

How can we apply these findings to in-store networks?

With the length of in-store viewing likely to be much shorter than online, designers should always assume that a clip will only be viewed for a few seconds -- even if the clip needs to be longer than that. If P&G's research on the First Moment of Truth has any legs, that means somewhere between 1.5 and 3 seconds of attention will be given to anything that catches the shopper's eye. Consequently, a 7-second spot will need to communicate its core message around 2-5 times, and a 30-second spot may need to repeat the message up to 20 times. It would get a pretty annoying (and maybe seizure-inducing) to flash a message on screen 20 times in a 30-second period, so designers need to cleverly ensure that the product's logo, complementary branding, tag line, key benefits and call to action are always featured on the screen during the longer clip. This approach also helps ensure that those who are engaged by the initial message won't get tired of watching the rest of the ad, which encourages further viewing.

Also, while we don't have anything that's exactly equivalent to pre- and post-roll time slots, we do have control over a digital sign's playlist. And even though there's still some controversy over if and when news, weather and other informational content should be displayed on these networks, we have plenty of anecdotal and hard data showing that people do watch it. Thus digital signage operators might consider this content like an attraction and try anchoring ads before and afterwards to reproduce the effects that the OPA found. Finally, while some digital signage networks do partition the screen into multiple areas (and thus would have space for traditional companion ads), many opt to play full-screen content only. Regardless of which camp you fall into, try using traditional POP displays around the store to reinforce your message. As we'll talk about in an upcoming article, multi-channel marketing campaigns that feature complementary digital and static displays can be extremely powerful, and they present an attractive, cohesive image inside the venue to boot. This also plays into the larger visual merchandising strategies that we've talked about before.

Down the road, I hope that someone like POPAI or the ISMI will give us a simple chart of general dos and don'ts for improving in-store ad performance, like the OPA has done for online ads. Alas, given the dramatic disparities between different venues and networks, I'm not sure if this is even possible, let alone forthcoming. But the results of the OPA's research are logical and make sense to me, so extending them out to the retail media space (with the caveats and considerations mentioned above) seems like a reasonable thing to do. In order to survive, it's essential that we learn from the mistakes of others. But if you want to thrive, it's probably a really good idea to pay attention to their successes as well.