10 Key Measurement Metrics for Video Marketers (Part 2)

In the first installment of 10 Key Measurement Metrics for Video Marketers, we looked at earned media views, content sharing and subscriber and sign-up conversions.  In this week’s post we’re going to explore three more key video analytics data points for marketers and advertisers: (A) click-through rates (CTRs) and referral traffic, (B) audience retention (view-to-completion), and (C) social conversation.  Let’s dive in.

4. CTRs

Marketers we talk to seem to be split about 50-50 on whether to track CTRs and referral traffic from videos as a campaign metric.  On one hand, it’s worth asking the question of any digital marketer – “why wouldn’t you want to track clicks?” – considering how focused many are on email, call-to-action button and social media post CTRs.  The flip side however, is that measuring referral traffic and CTRs from video tends to be challenging: primary social video hubs like YouTube and Vimeo offer very limited native click-tracking support. In the full-service version of our video A/B testing tool ZoomTilt Analytics, video experiments can be customized to track video click-through-rates for videos where YouTube annotation links have been added.  ZoomTilt Analytics can also estimate and compare CTRs for any type of video by capturing closed-loop audience data, but doing so currently requires additional custom client integration.

Video A/B Testing Software

If you’re not using ZoomTilt Analytics, there are still a few options to track CTRs on your video.  One option is using trackable links (like bit.ly) within YouTube annotations then calculating your CTR manually with bit.ly’s analytics data.  For enterprise customers using Brightcove or Wistia for video hosting, both platforms also offer click and conversion tracking insights on individual video call-to-actions within their respective analytics reporting.

Overall it’s helpful to monitor and A/B test video-specific CTRs, as well as track video referral traffic from YouTube and other content hosting destinations.

5. Audience Retention (View-to-Completion)

Video Audience Completion

One of the best video measurement metrics is, of course, audience retention or the video’s view-to-completion percentage.  The higher the average, absolute percentage, the more engaging, relevant and/or enjoyable the content is to audiences.

What’s a good baseline view-to-completion benchmark for branded video? One data set from Tubemogul indicates that, on average, only 15-25% of skippable online video ads are viewed to completion. In most cases, when viewers are allowed to skip pre-roll, they’ll jump ahead to the content they want to watch.  When video viewers are forced to watch ads, additional data indicates 85-95% view-to-completion rates for long-form videos and 65-70% completion rates for short-form (<5 minute) videos.

Since social video and branded entertainment is clearly the right content strategy in digital video, our view is advertisers should be sure to open their content to choice-based viewership and evaluate content traction with people who can freely skip or share it.  Audience retention rates won’t be as high as forced pre-roll or mid-roll ads, but that’s exactly the point: content marketing should be pull-based because of its value, not push advertising that’s forced into audiences viewing sessions.

When it comes to measuring view-to-completion, we see a new layer of valuable data emerging in demographic and audience-profile-specific retention data, a key feature within ZoomTilt Analytics.  For example, a branded entertainment spot aimed at working mothers might have 100,000 views and an overall view-to-completion rate of 30%.  That’s directionally helpful to a marketer, but it would be great to know what % of those views came from working mothers and what their subsegment-specific audience retention rate was.  Was it higher than the mean? Lower versus the mean? That’s a critical campaign insight that deserves to be tracked.

6. Social Conversation

Another metric we recommend video marketers track is the volume of social conversation around their video (where, at the risk of repetition, it helps to develop content worth talking about). Generally, the best places to track conversation is in comments sections (YouTube, Vimeo, Facebook, blogs where the video has been embedded) and on Twitter.  Generally, our baseline recommendation is to measure three conversation dimensions:

  1. Total number of comments, responses or social mentions for your video or campaign
  2. General sentiment of comments (positive, negative or neutral)
  3. Distribution of comments across owned media properties and other channels (is your content most discussed on your YouTube channel? more popular on Facebook? Why might that be?)

By understanding (and participating) in the dialogue around your video, as a marketer you have another great feedback tool to measure what works or doesn’t work with different audiences. Moreover, by encouraging your viewers to take a follow-up action by including a call-to-action that incentivizes user-generated content responses or conversation, you can help amplify your message and brand experience well beyond your own content.

=======================================================

In next week’s Part 3 we’ll continue our discussion of key measurement metrics for video marketers, looking at viewer sentiment, redemptions and the viewership distribution curve, so stay tuned!

Advertisements

10 Key Measurement Metrics for Video Marketers (Part 1)

Digital Marketing Measurement

Campaign measurement and metrics tracking is a must for digital content marketers. But, unlike other brand communication and advertising channels like PPC/SEM and blogging, where success is typically dictated by clicks and follow-on signup or purchase conversions, a common core objective in video marketing – particularly social video marketing – is to create content that encourages a strong enough emotional connection or response from a viewer that leads them to (1) share the video, (2) display a more positive affinity toward the brand, (3) explore additional brand properties (potentially further down the conversion funnel at a destination like a landing page or e-commerce store), or, even better, (4) perform some combination of all three. In other words, social video typically reaches people outside or at the very beginning of the typical inbound marketing funnel, a place where identification and attribution analysis can be most challenging.

As a result, video marketing success must be measured both quantitatively and qualitatively (with a heavy emotional component to the latter). Because of this, and with a goal of clarifying some common video marketing misconceptions, we’ve put together a practical guide to help digital marketers get started measuring the results of their video efforts.

1. Earned Media Views

Although a simple view count is the typical conversation starter when it comes to video performance, outside of directionally indicating content quality an impression count, total views are primarily a vanity metric that convey more symbolic significance than bottom-line results. In fact, probably the single most important thing a video view count establishes is social proof (or lack thereof) that the video might be worth looking at. All other things equal, from a consumer psychology standpoint, a video with 1,000,000 views will get a viewer’s click over a video with 10 views simply because of the perceived “potential interestingness” the higher view count conveys.

Why? For starters, views can be paid for. While there’s nothing wrong with this practice (in fact, quite the contrary, we actually encourage reasonable video distribution spending to help effectively seed marketing videos and help quality content get discovered online), the simple fact is that anyone can get a million views on their video with a large enough budget backing it. Second, a high number of views can still translate to low engagement from a minutes viewed and viewer retention standpoint if the video has a high, steep drop-off rate during watch sessions.

Our recommendation: Track your video’s absolute earned media (e.g., non-paid) video views from social media and search engine referral views.

Earned Media = Total Views – Owned Media (views from your website, Facebook page, etc.) – Paid Media (CPV or other promoted views)

If 20% or more of your video’s total views aren’t coming from earned media from social and search, you aren’t reaching audiences beyond your existing follower base, and it’s likely you should explore improving both the quality and the discoverability of your video content. Videos that are shared are 3 times more likely to be watched and are watched 3 times longer than a video that a user finds himself.

2. Shares

Over half a million branded videos are shared every 24 hours. What kind of share rates is your brand getting?

Based on ZoomTilt research and branded content performance data we’ve observed, a quality baseline share rate for social video is approximately 0.25%-0.50% direct shares (from the player or player-embedded page, excluding re-shares from social networks) as a percentage of total content views. In most cases, even smash hit digital video ad campaigns only see circa 2.5% direct share rates.

Be sure to track how your content is being shared. And, just as importantly, consider and try to benchmark why your content will be shared during the creation.

Why do people share things online according to the New York Times?

Psychology of Social Sharing

3. Subscriber Conversions

As marketers, we all recognize the importance of subscriber lists. In many cases, email lists are the backbone of web commerce companies like Fab, Karmaloop, Birchbox and Rue La La. Video marketing is no difference: there’s much less value in a one-time view or share than a subscriber who sticks around for more.

In fact, if you’re using a hosting provider like Wistia, you can even capture email conversions within your video player:

Wistia conversion

However, with the bulk of social video activity happening on YouTube (and driven by shares of YouTube embeds on blogs and social networks), YouTube channel subscribers is a critical metric for the ongoing success of your video content marketing efforts.

An interesting case study example is “The Beauty Inside” video campaign by Intel & Toshiba. Released as an interactive web series, “The Beauty Inside” reached close to 7 million YouTube views and 14 million Facebook page interactions, according to Intel. At a high level, clearly a success, particularly in light of the fact that a concurrent Toshiba laptop campaign running through Best Buy saw sales increase from 200 units a week to 900 units a week during the campaign’s early release.

But one place the “The Beauty Inside” campaign seems to fall short is cultivating YouTube subscribers. As most social media marketers know, a Facebook page like just isn’t what it used to be (more specifically, it’s about 10% of what it used to be), so getting high Facebook engagement from a social video campaign is nice, but less impactful than adding new YouTube subscribers. Moreover, because Intel and Toshiba released “The Beauty Inside” on a special branded channel titled TheBeautyInsideFilm, the 16,000 subscribers they collected don’t subscribe to Intel or Toshiba’s YouTube pages, which TheBeautyInsideFilm doesn’t even bother to link to. In some way, while insulating the artistic integrity of “The Beauty Inside” from overt branding, both brands sacrificed their opportunity to build their YouTube subscriber base – a mis-step in our view.

Our recommendation: pay attention to and encourage subscriptions (channel, email) within your social video campaigns. Building a robust subscriber list on YouTube is one of the best ways to drive recurring brand engagement with your video content.

=======================================================

In next week’s Part 2 we’ll continue our discussion of key measurement metrics for video marketers, looking at click-through-rates (CTRs), view-to-completion, and more, so don’t miss out!

Introducing ZoomTilt Analytics

Today, we’re pleased to announce the beta release of ZoomTilt Analytics – a self-service software tool for A/B testing videos to help users identify and optimize their top-performing video content. The goal of ZoomTilt Analytics is to help businesses and video creators:

  1. Make better, more audience-targeted videos by compiling feedback and data from real, relevant viewers;
  2. Make smarter decisions about what videos to create, how to edit them and how to release them; and
  3. Increase video marketing return on investment.

The trial version of ZoomTilt Analytics, which allows experimenters to easily set up and run video A/B tests from YouTube, is now available as a free service on ZoomTilt.com. In addition, our ZoomTilt Analytics Premium service now gives brands, agencies and media companies the ability to create and customize video A/B tests around specific target audience profiles and marketing metrics.

We’re very excited to share ZoomTilt Analytics with you, we have plans to introduce lots of new features and capabilities, and we welcome any feedback or questions you’d like to share with us. Interested in learning more about the benefits of ZoomTilt Analytics Premium for your business? Contact us today to get started.

Check out the video below for a demo of ZoomTilt Analytics in action:

Solving the Equation of a Hit Film Script with Big Data

Vinny Bruzzese charges up to $20,000 to analyze and make recommendations on Hollywood scripts. He is also catching hate in the comment section of NYTimes.com, mostly from people who didn’t seem to read the entire article about his script evaluation company.

The tone of online commenters seems to suggest they think Mr. Bruzzese is taking the proverbial “room of monkeys writing Shakespeare on typewriters”, and distilling that into script writing software.

Wrong.

With Mr. Bruzzese’s methods, computers write exactly 0% of scripts. His business is highly consultative and contextual while adding years of compiled audience research into consideration. All this to save studios money at the most critical juncture, before it is spent.

Nothing uncreative about that.

Nonetheless, many leading indicators are pointing to the fact that video creative – whether it’s a movie, TV pilot, advertisement or video blog – will be shaped more and more by “big data” in the future. In contrast to Netflix’s shotgun approach, Amazon Studios has begun split-testing new TV pilots by promoting and distributing them freely on the web in order to gather audience reviews and viewer data, while over here at ZoomTilt we’re bring video creators the first self-service software tool to A/B test videos. But, in both cases, rather than using data to take away from the creative magic of screenwriters, directors and filmmakers, the goal is to make production more merit-driven based on true information, not somebody’s opinion or connections.

Welcoming to the new era of data-driven video – we happen to be pretty optimistic about it.

Solving the Equation of a Hit Film Script with Data.

An Exclusive Preview of ZoomTilt’s Upcoming Video Testing Analytics App

Video

We are proud to show you the first video demonstration of our video testing analytics app! Now you can know exactly what your audience thinks of your videos.

To learn how to get early access when the software goes live, drop us a line here: http://www.zoomtilt.com/analytics

Can You Predict a Viral Marketing Video?

Volkswagen Star Wars Super Bowl Advertisement

As a first order of business, the thunder-stealing, lead-in answer is still “probably not.” That said, this week we’re pleased to announce the launch of a new video analytics tool for ad agencies and video content marketers that brings this predictive digital marketing dream one step closer.

Video Analytics Screenshot

Our solution is a first-of-its-kind tool capable of performing targeted, simultaneous algorithm and audience-based testing of a single video, multiple videos or even multiple edits of the same branded content spot. By enabling digital advertisers, video marketers and content creators to specify audience goals such as age, gender, income level, zip-code and/or video “share rate,” then quickly test their videos against those goals and audience profile, our scoring tool brings scalable marketing automation, big data analytics and a rapid-prototyping feedback loop to video production, video audience measurement and video performance forecasting. Beneficial ways to use our new video analytics tool include:

  • A/B testing different video concepts or video edits to determine which performs best for a given campaign/activation goal or audience profile.
  • Test and compare your videos against public videos from competitors on key metrics like audience retention and viewer click-through rate (CTR).
  • Pre-release testing one or multiple pre-roll or TV ad spots earlier to proactively reduce the risk of negative brand exposure, campaign under-performance, mis-targeting and/or distribution over-spend.
  • Evaluating branded entertainment and web series pilots with richer tools, deeper insights and a faster feedback loop (so you don’t have to play the Netflix game and order seasons up front at $4.5 million an episode).

Best of all, by connecting your YouTube or Wistia video hosting account (plus more hosting platforms on the way), users will also be able to compare and back-test predicted video virality versus actual, real-world earned media rates, social media mentions and referral shares, and easily generate reports.

The tool is currently in private beta, with a broader, public release planned in May. To apply for early access, contact ZoomTilt via email or visit our video analytics signup page .