Now, I know that most of the video ads that brands and agencies will launch for the holidays have already been shot, edited, reviewed, and approved, and they’re just waiting to be uploaded and published. But, it’s not too late to learn a few last-minute Christmas video marketing lessons from John Lewis, a chain of upmarket department stores operating throughout the United Kingdom.

For example, check out “John Lewis Christmas Advert 2016 - #BusterTheBoxer.” Uploaded to YouTube on November 10, 2016, it has generated 16.7M views, 13.8M in the first 3 days alone!

A second version of the video, “#BusterTheBoxer,” was uploaded directly to Facebook on the same day. It generated 31.6M views, 30.1M in the first 3 days after it was published. This puts #BusterTheBoxer significantly ahead of the pace set by “John Lewis Christmas Advert 2015 - #ManOnTheMoon.” Uploaded to YouTube on Nov. 6, 2015, #ManOnTheMoon generated 12.3M views in 7 days, 10.4M of those views in the first 3 days.

A second version of “John Lewis Christmas Advert 2015 - #ManOnTheMoon” was uploaded directly to Facebook on the same day generated 6.5M views in the first 7 days. So, the two versions of #BusterTheBoxer got a total of 48.3 million views over their first 7 days, compared to a total of 18.8 million views for the two versions of #ManOnTheMoon over their first 7 days.

Comparing Video Campaign Metrics

Now, video marketers already know that comparing views on YouTube and Facebook is like comparing apples and oranges. But, I’m using different metric to compare a basket of apples and oranges from last year to a basket of apples and oranges from this year. And this year’s basket of views is more than 2.5 times larger. So, whatever John Lewis is doing to get more views for its annual Christmas advert is something that other brands and their agencies will want to understand as soon as possible.

Tubular Labs, the only analytics software dedicated to giving publishers and content marketers the most comprehensive and actionable video intelligence, introduced two new metrics in 2016 - V3 (views generated in the first 3 days after upload), and V7 (view generated in the first 7 days after upload). V3 and V7 enable us to benchmark YouTube and Facebook videos from 5.3 million publishers over two important milestones, respectively. Using Tubular, you can also use V30 to compare and contrast your videos with those from competitors over their first 30 days.

bustertheboxer-video-adBut, views isn’t the only metric that matters. That’s why Tubular also introduced new metrics for engagements over the first 3, 7, and 30 days. For example, the ER3 and ER7 metrics measure a video or publisher’s engagement rate over the first 3 and 7 days, respectively, benchmarked across all content (e.g. 2.2x more engaging than average). Using Tubular’s new engagement rate metrics, the YouTube version of #BusterTheBoxer had an ER3 of 1.1 x and an ER7 of 1.4x, while the YouTube version of #ManOnTheMoon had an ER3 of 0.9 and an ER7 of 1.1x. Meanwhile, the Facebook version of #BusterTheBoxer had an ER3 of 1.6x and an ER7 of 1.7x, while the Facebook version of #ManOnTheMoon had an ER3 of 1.8x and an ER7 of 1.9x.

So, this year’s Christmas advert by John Lewis generated roughly similar cross-platform engagement over the first 3 and 7 days than last year’s one over the same period of time. Since a V7 that’s 2.5 times higher can’t be explained by a significantly higher engagement rate, this indicates that John Lewis may have put a bigger ad budget behind promoting this year’s Christmas advert. Not that there’s anything wrong with that. Brands can get views organically by creating engaging content that people like, comment on, and share. Their agencies can also get views by leveraging paid video advertising to put that engaging content in front of the right audience. And in the world of online video, the target audiences are far more segmented than the ones reached by television. So, the media targeting process involves a lot more strategic choices than selecting which shows on ITV currently have the right reach and composition for your annual Christmas advert.

The Impact of Mobile on Video Marketing

In addition, mobile phones have created new ways for consumers to engage with brands, often blurring the physical and online worlds. For example, people might see your Christmas advert on TV and then go online to learn more. Or, they might watch a mobile video while they’re commuting on the London Underground (aka the Tube) that inspires them to shop when they’re back home on their laptops. Or, they might watch your annual Christmas advert online and then go into your department store to buy that Plum Space Zone II Trampoline & Accessory Pack that was featured in Buster the Boxer. So, if there are some last-minute lessons worth learning from John Lewis, then one of them is this: Creating emotional content that’s even more engaging than last year’s memorable hit is hard.

So, kudos to John Lewis for creating yet another Christmas advert that’s more engaging than average. And more kudos to the company for adding “Discover Buster's Garden in 360” this year. It enables viewers to look around and explore the 360 world of #BusterTheBoxer and his friends on their mobile, tablet or desktop devices.

Now, if viewers are watching on a mobile phone, they’ll need to have the YouTube app installed on their device to get the full 360 experience. Or, to explore Buster's Garden in a Google Cardboard device, they can simply touch the cardboard icon, insert their phone, and view the world in 360. These requirements may explain why Buster’s Garden, which was also uploaded Nov. 10, 2016, only has a V3 of 71,100 and a V7 of 87,200 views. And it has an ER3 of 0.5x and an ER7 of 0.6x, which indicates that John Lewis is still trying to figure out the best practices for making an engaging 360-degree video.

It's still early days for 360-degree video, so it’s critical to place clear instructions at every turn: in website copy, in the content itself, even in the voiceover. So, maybe the little girl called Bridget, who loves to bounce, could have guided viewers through the experience. A guided tour might have served to engage a higher percentage of viewers.

What Brands Need to Do After Uploading a Holiday Video

Another lesson worth learning from John Lewis is this: Even if it’s too late for other brands and their agencies to put a bigger ad budget behind promoting their Christmas videos this year, they still have plenty of time to review and revise their target audiences before the holidays are over. Now, I know their digital media plans have already been finalized. But, these plans are based on assumptions on which target audiences will be engaged by the content. And, to paraphrase an old military strategist, no plan survives contact with the customer.

So, what can brands and their agencies do 3 to 7 days after their new video has been launched? For YouTube, you have a number of options for improving your video advertising results without increasing your ad budget. For starters, everyone fixates on views, but you should double-check early and often to see which video ad format is getting better engagement rates: TrueView in-stream ads or TrueView video discovery ads, the format formerly known as TrueView in-display ads.

Next, double-check the campaign-level targeting that you’re using. For example, you can target your video ads by geographical location. So, even if you’re only running your ad campaign in the United Kingdom, are you getting better engagement rates in different cities, regions, or postal codes? At the campaign level, you can also target ads by device. So, are your engagement rates higher after people have watched your video on a desktop, tablet, or mobile phone? And, if a mobile user is near one of your stores, what’s the offline impact. Do your online ads lead to store visits and drive in-store transactions?

At the ad group level, you have a wide variety of additional targeting methods available. So, are you getting more robust engagement rates using demographic targeting, which lets you choose the age, gender, parental status, or household income of the audience that you want to reach? Or, does keyword targeting, which shows your video ads based on the search terms or phrases that your audience is interested in, ensure that don’t miss the grandparents, cousins, friends, and co-workers who are also planning to purchase Christmas presents for Bridget?

What happens when you target your video ads to specific topics on YouTube and the Google Display Network like Mass Merchants & Department Stores? Or, are you getting above average engagement rates when you target affinity audiences like Shopaholics or in-market audiences like Consumer Electronics or Home & Garden to reach people interested in certain topics, even when they may be visiting pages about other topics? You can also target unique channels, videos, apps, websites, or placements within websites. So, do you get superior engagement rates by using Google Preferred, which gives brands access to established creator channels in categories like Family and Parenting, or by using the Breakout Videos package, which lets brands advertise against the hottest and fastest-rising videos on YouTube? 

In addition, you can show personalized ads to millions of viewers across YouTube and video partner sites, based on their past interactions with your videos, TrueView ads, or YouTube channel. This is known as “video remarketing.” So, if someone has liked, commented on, or shared your Christmas advert, what new or existing campaign should you show then next?

Finally, you can improve your TrueView advertising results – without increasing your video ad budget – by using online video metrics that really matter for brand advertisers and their agencies. As I mentioned last month, Google’s Brand Lift solution gives you insights into how your ads are impacting brand awareness, ad recall, consideration, favorability, purchase intent, and brand interest, as measured by organic search activity on Google and YouTube. These metrics are available within days, enabling you to quickly optimize your campaigns mid-flight, based on these near real-time results.

For example, this year’s branding campaign not only boosted searches in the United Kingdom for John Lewis, Buster the Boxer, and John Lewis advert, it also triggered searches for John Lewis coats, John Lewis Chelmsford, John Lewis Leeds, John Lewis cashmere, John Lewis boots, John Lewis toys, John Lewis iPhone, John Lewis Hatchimals, and John Lewis Black Friday.  Don’t you think the retailer would want a heads-up about these unexpected findings as soon as possible?

And the wisdom of revising your plans after launching a campaign became more apparent in September 2016, when Google and YouTube launched new measurement products at Advertising Week, including:

  • Brand Lift for TV. And based on early tests, YouTube campaigns generate almost 2 times more searches per impression than TV campaigns.
  • Location extensions for display ads. Using Google Maps technology, you can now connect your online ads with visits to over 200 million stores globally with 99% accuracy.
  • Cross-device remarketing. This enables retailers to reach a customer with a Christmas ad on their phone during the morning commute and follow up later in the day with a limited time offer when the same user is browsing on his or her tablet at home.

But wait, there’s more! You also have a number of options for improving your video advertising results on Facebook without increasing your ad budget.

What’s your marketing objective when you create a video ad on Facebook? Most brands and their agencies select “get video views,” which is one of 15 options. But, have you tested “increase brand awareness” or “get people to visit your stores”? Maybe the results will surprise you. When it comes to audience targeting options, you have the option to target people based on their location (countries, states, provinces, cities, congressional districts, zip or post codes). But, you can also target people based on what they do off of Facebook.

For example, you can include or exclude people from your target audience based on their connections to your Pages, apps or events. Or, you can create Custom Audiences to target people you already know. They can be created from information you provide to Facebook from customer files, or from information generated by the Facebook pixel, the Facebook SDK, and engagement on Facebook. You can even target Lookalike Audiences, which are a target audience you create from a "source." It finds other people on Facebook who are the most similar to the people in the source. You should definitely double-check whether you’ve started to use Detailed Targeting, which is rolling out right now. You find targeting options such as demographics, interests, and behaviors.

And here is the key question to ask: Should our target audience be specific or broad? The right approach depends on what you're trying to accomplish and your ad budget, of course. Targeting specifically means that you’re giving Facebook a relatively strict set of parameters to work within while finding the best people to show your ad to. These parameters may lead to a potential audience that’s smaller, but also more interested in what you’re advertising.

For example, the new John Lewis Oxford Street shop in London may want to target specifically. It could target people who are parents and also have easy access to the Oxford Circus station (via the Central, Victoria, and Bakerloo lines) or the Bond Street station (via the Central and Jubilee lines).

Targeting broadly means that you're relying on Facebook to find potential customers you never would've known about otherwise. It's a good approach if you're not sure who you want to target or you’re running a large-scale brand awareness campaign.

For example, John Lewis may want to target its annual Christmas advert broadly – at least initially. However, the brand and its agency should then check Audience Insights or ads reporting to learn more about the types of people that Facebook found for them, discover how these people reacted to their ads, and decide whether or not to shift to targeting more specifically. Facebook also offers Brand Lift, formerly called brand polling, which enables advertisers to measure brand awareness and ad recall. However, Facebook Brand Lift is only available to advertisers running “eligible brand campaigns.” To find out if you are eligible, contact your account representative.

Earlier in this article, I said that John Lewis “may have” put a bigger ad budget behind promoting this year’s Christmas advert. Then again, the brand and its agency could have learned who they should target more specifically during last year’s campaign, which has enable them to get 2.5 times more bang for the same ad budget this year. I don’t know; it’s a mystery. So, let me close with this open invitation: If someone at John Lewis or adam&eveDDB, its London agency, would like to be interviewed about their latest Christmas advert, just let me know in the comments area below. I would love to do a follow-up story.