Recently I was asked by one of my clients to evaluate the usage of video demand-side platforms (DSPs) for branding campaigns. This is interesting because DSPs have always been associated with performance/direct-response campaigns. When I think of DSPs, I think e-commerce retargeting, automotive leads acquisition, or FMCG member acquisition campaigns. The reason for this association is due to the fact that direct-response campaigns are purely based on results, and only better results can justify the claims that many DSPs makes: “accurate audience buying,” “millions of audience tags,” etc…
However, this started to change when local DSPs started to acquire online TV (OTV) inventory. iPinYou was one of the first local vendors to get into the OTV space, and their rationale was simple: OTV is by far one of the hottest sectors in China, and video inventory can get them a bigger piece of the pie compared to performance campaigns. So, can video DSPs effectively be used for branding campaigns? I’d like to evaluate this from three dimensions below: Inventory, Targeting, and Trading.
- Advantages: Even though the top four OTV channels in China (Youku, iQiyi-PPS, Tencent, Sohu) seem to procure different content for their users, a large portion of the content does overlap. This means that consumers will rarely stay loyal to one single OTV channel; they will simply follow the content. Hence, if the branding objectives are to hit a large number of UVs with a reach target, then there’s going to be tons of repeated impressions. This is an area where video DSP shines, because their inventory expands into many smaller OTV channels like LeTV, 56, Ku6, etc… So a branding campaign with the same budget will hit a larger UV target using video DSP than going with the four big channels.
- Disadvantages: We all know that regardless of whether the format is display or video, most DSP inventory is going to be remnant. So if the branding campaign was doing any sort of TV show buying based on brand affinity, then using DSP inventory will certainly be a no-go.
- Advantages: Targeting is the core selling point of DSPs over direct-buys. There’s the more traditional targeting methodology like retargeting based on site visits or banner interactivity. But there’s also the more interesting mechanisms like targeting based on video tags. An example would be targeting mothers based on tags like “baby videos.” This type of targeting successfully utilizes the UGC inventory that’s largely considered non-premium by brands. Another targeting-related feature is cross-media frequency control; as ad serving is not prevalent in China, brands can only frequency cap pre-rolls on a single channel. But with video DSP, it essentially takes the place of an ad server and caps your reach goals accordingly across the different media channels.
- Disadvantages: The downside of these new targeting methodologies is twofold. First, the brand must have complete trust in the DSP vendor data, because all targeting is based on a black box. There are also no third-party DMPs in China, so this in itself creates lots of transparency issues as to where exactly the DSPs get their cookies. Second is that the new targeting methodologies simply don’t fit with the traditional TV buying methodology of measuring eyeballs (GRP/iGRP). Let me explain: Currently, OTV buying in China is usually in conjunction with TV. The methodology goes that if we mix OTV inventory using iGRP (with netizen as the universe) along with TV, then we can get to the same reach goal with lower investment, as OTV inventory is generally cheaper than TV. The iGRP target reach is calculated the same way as TV target reach using primarily offline panels. This traditional methodology assumes that some of your impressions are wasted because the consumer that was exposed was essentially not your target. But the DSP audience buying methodology claims that every impression is essentially on target. So a DSP-run branding campaign has an effectively 100 percent target reach. How do you verify this? The fact is, you can’t! You simply have to believe. Hence, brand marketers have to make a big leap of faith in following the DSPs’ targeting methodology. There are certainly other ways to integrate DSPs with the existing iGRP methodology (by treating DSP as a single channel instead of an aggregator of different channels), but I won’t go into much detail here.
- Advantages: Lower CPMs! Aside from the obvious pricing advantages, video DSPs can also serve as a supplement to direct-buy OTV channels. From a trading perspective, Chinese OTV channels often way oversell their available inventory because the demands from brands are too much. This is especially the case in first- and second-tier cities like Beijing and Shanghai. Hence, in order to hit the brand’s impression key performance indicator (KPI), they can only make up the impressions several weeks if not month after the campaign has already ended. Now, this will have a severe impact on products that have strict seasonality, such as ice cream. What media agencies can do to rectify this problem is to estimate the volume that’s unfinished by the end of campaign for a certain market, and then pull the investment from direct buy into video DSPs.
- Disadvantages: Even though the CPMs are lower, there are certain political trading issues with media that brands and their media agencies must solve if they’re to buy inventory from the video DSPs. Many bigger brands in China invest more than 50 to 70 percent of their digital budget on OTV channels. With this level of investment, media agencies can negotiate lower CPMs based on a pre-committed annual investment level. But when comparing these negotiated CPMs with video DSP CPMs, DSP vendors still come out lower simply because their inventory is not as premium. Now that’s great from a brand’ perspective, but the media block investment from video DSPs because they do NOT want the brand to be getting a lower CPM through another source. Even if the brand exerted enough pressure on the media to sell their inventory through the DSP, the media will certainly not count that investment toward their annual commitment. This presents a dilemma for brands, as it impacts their pre-negotiated CPMs – do they pull a percentage of investment out from direct into DSPs for a lower CPM at the cost of impacting their direct-buy rates?
From looking at these three dimensions, video DSPs certainly have both distinct advantages and disadvantages. In order for brands to successfully utilize this new channel, there need to be many fundamental shifts in measurement methodology as well as investment mindset. So I urge all Chinese marketers to carefully evaluate the above three dimensions for your own brands.
Image via Shutterstock.
Header bidding is a programmatic technique that allows publishers to offer their inventory through multiple ad exchanges before they serve up ads from their ad server.
All top Chinese retailers, banks and internet companies share mobile data in earning releases. None of the top 10 US retailers do, nor does Google. US banks and Facebook are better.
American Apparel's chief digital officer discussed the future of retail, the importance of delivering value to the consumer, and strategies for an IoT and omnichannel world.
With so many bloggers and blog posts storming in every day, it is hard to stand out. The secret to creating an outstanding blog is to cultivate a community around your blog.