Archive for the ‘Measurement and metrics’ Category

What's Dwell and why do you care?

Thursday, May 6th, 2010
Source: Eyeblaster Research. Data: Q1 2009 to Q4 2009. Worldwide.

Source: Eyeblaster Research. Data: Q1 2009 to Q4 2009. Worldwide.

In a previous post, we talked about approaches to measuring display ad effectiveness that go beyond clickthrough.

In this post, we’ll look at one of those approaches, the concept of Dwell, in a little more detail as it’s described in an Eyeblaster Benchmark Insights paper. And we’ll touch on how you can measure Dwell performance for your own Xspot campaigns.

What exactly is Dwell?

Dwell as a performance metric was developed, in response to research by Eyeblaster, Microsoft Advertising, and comScore, to capture the branding effect of display ad campaigns.

You calculate Dwell in two different ways:

  • Dwell Rate measures how successfully an ad captures viewers’ attention. Dwell Rate equals the number of viewers who physically touch an ad divided by the number of ad impressions served.
  • Dwell Time measures how long viewers were exposed to an ad. Average Dwell Time equals the total amount of time all viewers spent engaging with an ad divided by the total number of viewers.

What does Dwell affect?

Based on comparisons of high and low Dwell campaigns, researchers found that viewers exposed to high Dwell campaigns are significantly more likely to:

  • Search for brand-related keywords.
  • Visit advertiser websites and, during those visits, spend more time and view more pages.
  • Convert.

What produces better Dwell?

Given Dwell’s affect on branding and conversion, advertisers may want to take concrete steps to improve Dwell performance for their campaigns. The research data provides advertisers with some specific actions to take:

  • The Dwell research found a direct correlation between the length of time an ad is presented and Dwell Rate. This means advertisers are likely to see higher Dwell Rates with ads placed on pages, such as news sites, where viewers spend more time.
  • Video ads perform better than ads without video. On average, adding video to ads increases Dwell Rate by 29 percent and also nearly doubles Dwell Time.
  • Polite and expandable banners typically have relatively high Dwell Time but relatively low Dwell Rate (as compared to commercial break and floating ad formats). To increase banner ad Dwell Rate, make the starting images that initially load as engaging as possible.

How can I measure Dwell for an Xspot campaign?

Dwell Rate In the Mixpo platform, Active View Rate is equivalent to Dwell Rate. Active view rate is calculated by dividing the number of times viewers play or replay an ad by the number of impressions served.

Performance statistics in the Mixpo Dashboard

Performance statistics in the Mixpo Dashboard

Average Dwell Time In the Mixpo platform, you have three ways to measure time spent with an Xspot:

  • Avg. Viewed = percentage of an Xspot viewers are watching averaged across all viewers.
  • Percent Completed = percentage of viewers who watched the entire Xspot.
  • Exposure = the total number of minutes all viewers of an Xspot spent watching it.

Tags: , , , , ,
Posted in Agencies, Measurement and metrics, Publishers, Research and industry data | No Comments »

Measure to optimize AND discover

Wednesday, May 5th, 2010
Video ad views and clicks by geographic location

Video ad views and clicks by geographic location

In a thought-provoking article in the New York Times Magazine, Gary Wolf examines how technology has made it feasible for people to measure and evaluate their most basic habits.

For example, enthusiasts that Wolf describes as “trackers” are measuring how much they sleep, exercise, and eat, how productive they are, where they’re located, and even how their moods, alertness, and spiritual well-being fluctuate.

Wolf acknowledges that “technical people are often painfully aware how much of human behavior is a mystery. People do things for unfathomable reasons.” And yet, he points out, trackers believe that they can gain self-understanding, not just by interrogating inner worlds through talking and writing, but also by using numbers. “They are constructing,” Wolf says, “a quantified self.”

In a similar way, advertisers construct quantifiable campaigns. People buy things for what may, in fact, be “unfathomable reasons.” But, as Stephen DiMarco points out, in Campaign Attribution Isn’t A Zero-Sum Game, advertisers rely on data to refine campaign strategies and maximize each dollar they spend.

In the process of tracking his work hours, Gary Wolf learned, much to his chagrin, that the amount of uninterrupted time he was able to muster in a given work day totalled about 3 hours. Once he got over the humiliation, he realized his measurement experiment had turned him into a “mean-spirited, small-minded boss” who was making an unnecessary concession to a worthless stereotype. “Does anyone believe” Wolf asks, “that long hours at a desk are a vocational ideal?”

Like other trackers, Wolf discovered that the true value of a tracking system is as a source of critical perspective, not on one’s performance per se but rather on the assumptions about what it is important to track.

“Self-tracking,” Wolf concludes, “is not really a tool of optimization but of discovery.”

In arguing for an integrated approach to measuring the effectiveness of advertising campaigns, DiMarco makes a similar point. He advises against falling in love with 1 or 2 specific metrics and for going beyond ad server and site data to include direct marketing emails, organic search, and even consumer panels.

“Experimentation” says DiMarco, “is key to lasting success. Attribution isn’t a zero-sum game; it’s your best way to learn more and earn more.”

Tags: , , , , ,
Posted in Agencies, Measurement and metrics, Publishers | No Comments »

Be realtime relevant with dynamic video ads

Monday, May 3rd, 2010
Grungy social media icons from webtreats.com

Grungy social media icons from WebTreats etc.

Kimbal Musk, CEO of OneRiot, claims that brands or websites that want to connect with users in a social or realtime environment need to advertise in realtime.

According to Musk, realtime ads are ads that “are created as they become relevant across the realtime web. Realtime ads” he says, “can be Tweets, search ads or display ads that contain content that is relevant right now.”

Indeed, one of the keys to the success of the Scott Brown for U.S. Senate campaign was the campaign’s ability to immediately capitalize on audience reaction to the final Brown-Coakley debate.

Immediately after the debate, without retrafficking any ad tags, the campaign replaced the creative in a video ad running on BostonHerald.com with “Momentum” creative that stoked the building enthusiasm.

Voters chose to watch Brown’s video ad nearly twice as often as the industry average for video ads. They engaged twice as often as viewers engage with video ads overall. The campaign’s click through rate exceeded the industry standard for display ads by almost double.

Read a case study about the Scott Brown campaign.

Learn more about the realtime power of dynamic video advertising.

Tags: , , , , , ,
Posted in Agencies, Case studies, Measurement and metrics, Publishers | No Comments »

When is a video ad campaign effective?

Monday, April 26th, 2010
Chart showing viewer engagement over the length of a video ad

Chart showing viewer engagement over the length of a video ad

As studies increasingly show (for example, see comScore’s Whither the Click in Europe?, this .Foxnet-comScore study, this VideoEgg-comScore study, and these reported results of a Nielsen IAG-Microsoft study), engaging video ads, even in the face of minimal clicks, effectively drive visits to advertisers’ websites and increase the number of trademark searches.

Given this data, click-through rate is not a terribly meaningful measure of an online video ad campaign’s success. But if not click-through, then what?

Here are some ways researchers are answering this question.

Dwell time

Using data sets from comScore and Eyeblaster, Microsoft found a “clear connection between the amount of time that a user spends actively engaging with an online ad and a consumer’s subsequent online behaviour with that brand.” Higher dwell scores mean more branded search queries and more visits to and engagement with the brand website.

Engagement mapping and view-through conversions

For several years, online marketers have been struggling with the disconnect between the knowledge that consumers visit multiple sites before conversion and metrics that focus on the last-seen or last-clicked ad.

For example, a study conducted by Microsoft’s Atlas Institute of advertising campaigns run by 250 advertisers across thousands of sites showed that most consumers make contact with advertisers months before they actually convert. Using a 90-day history window, consumers experienced a median of 18.5 ad events.

In response, Atlas introduced Engagement Mapping technology that allows advertisers to fold all types of events, including passive events such as views and impressions, into their ROI calculations.

Google recently announced view-through conversion reporting on the Google Content Network, which allows advertisers to tie conversions to display ad views within a 30-day window even when no clicks occur.

Engagement rate and conversion activity

In a June, 2009 benchmarks review, DoubleClick suggested that “the objective of rich media ads isn’t always to drive clicks.” Therefore, engagement metrics, such as interaction rates and times and video completion are important success measures.

In addition, DoubleClick argued, conversion activity can happen within an ad unit itself. For example, while watching a video ad, viewers might open and submit a lead capture form, take a poll, locate the nearest retail outlet, or play another video.

A wealth of data

In the Mixpo platform, you have access to a wealth of real-time data related to the performance of video ad campaigns. For example, you can:

  • See how much of your video ad viewers watch on average and what percentage of viewers watch the entire ad.
  • Track clicks, views, and interactions individually, or see an overall engagement rate that divides the total of all of these measures by the number of impressions to calculate an engagement percentage.
  • Track conversions, either by counting actions viewers take within a video ad or subsequent visits they make to the advertiser’s website.

How do you see it?

Let us know what you think. What results should advertisers expect from video ad campaigns? What metrics should they use to judge campaign effectiveness?

Tags: , , , ,
Posted in Agencies, Measurement and metrics, Publishers, Research and industry data | No Comments »

Video ad exposure drives brand engagement

Monday, April 19th, 2010

A recent .Fox Networks and comScore study shows that exposure to video and display ads, even without clicks, increases consumer:

  • Visits to advertiser websites.
  • Searches for advertiser trademarks.

Compared to consumers exposed to display ads, consumers exposed to video ads were 28 percent more likely to visit brand sites and twice as likely to conduct trademark searches.

This study, which examined 4 2009 campaigns that delivered 300 million impressions to UK Internet users:

  • Proves the value of the video advertising format.
  • Demonstrates how important it is to measure campaign effectiveness using behavioral metrics that go beyond clicks.

Download a PDF about the Value of the Mixpo Xspot.

Tags: , , , ,
Posted in Agencies, Measurement and metrics, Publishers, Research and industry data | No Comments »

Google on the future of display ads

Monday, April 19th, 2010

In mid-March, Google began a series of posts on The Official Google Blog about the future of display advertising.

In the first post, Susan Wojcicki, vice president of product management, introduces the series by describing Google’s belief “that display advertising, as a category, can grow dramatically,” a belief that has spurred a number of Google acquisitions, including YouTube, DoubleClick, and Teracent, as well as the development of new and enhanced advertising technologies.

Wojcicki attributes the renaissance in display advertising to the evolution of display ads “from a series of simple, static images to the incredible creative units that we see today,” units that include complex animations, stunning videos, interactive and social elements, and more.

Google’s display ad investment is supported by 4th quarter 2009 online advertising data from comScore that shows display ad spending growing at 4 times the rate of search spending.

comScore’s Gian Fulgoni speculates that perhaps “the search industry has begun to reach maturity, pricing has become an issue for advertisers, or they realize that clicks on ads don’t produce relevant metric….”

Now is the time, as Wojcicki points out, for agencies, advertisers, and publishers to take advantage of technology that makes it easy to run media across the web at scale, create engaging ad formats, measure the impact of ad campaigns, and deliver relevant ads to the right audiences in real-time.

Tags: , , , ,
Posted in Agencies, Measurement and metrics, Publishers, Research and industry data | No Comments »

Synergy: TV + online more than sum of parts

Sunday, March 28th, 2010

Double an ad campaign’s brand and message recall without spending any additional money? Sounds like a snake oil sales pitch.

As the recent Yahoo Advertising Blog post called A Smarter Mix of TV and Online demonstrates, however, it’s anything but.

From 2004 to 2009, TV viewership remained flat. During that same time period, Internet usage among North American households increased 117 percent. By taking advantage of these trends, advertisers can increase a campaign’s reach while lowering the cost of reaching target audiences.

Based on its analysis of Nielsen IAG data, Yahoo uses an auto maker promoting a new hybrid vehicle as an example. By devoting its entire $10 million advertising budget to a TV campaign, the auto maker could reach 68 percent of its target market. By shifting just 10 percent of the $10 million from TV to online, the same advertiser could reach 70 percent of potential buyers at a $3K+ lower cost per point (CPP).

No snake oil there. Just a strategic allocation that builds on the synergy between TV and online.

Read more or learn how easy it can be to extend your TV campaigns to the internet.

Tags: , , , , ,
Posted in Agencies, Measurement and metrics, Research and industry data, TV and Xspots | No Comments »

Build political lists with Xspots

Thursday, March 11th, 2010

Never miss an opportunity to get someone’s email address. That’s the key piece of advice the article called Online Advocacy Tools: Email Lists gives to advocacy organizations and political campaigns.

“Successful campaigns,” the article continues, “make every effort to convert casual visitors into active supporters,” who, typically, become a primary source of contributions.

Using easy-to-customize calls to action, you can convert any Mixpo Xspot into a list-building tool. Because viewers who join a list through an Xspot have chosen to watch and interact with the ad, they represent a highly-qualified group of leads.

Watch this Scott Brown demo Xspot to experience the list building features for yourself.

While the Xspot plays, expand the INFO menu and click the Volunteer Today! link. Click the Become a Scott Brown Supporter! overlay to open a custom lead capture form.

Note These lead capture calls to action are for demo purposes only and weren’t actually used during the Scott Brown campaign.

Learn more about dynamic video advertising for politics.

Contact us to learn more about building lists with Xspots.

Tags: , , , , ,
Posted in Measurement and metrics, Product features, TV and Xspots | No Comments »

Brown campaign validates Mixpo model

Thursday, March 4th, 2010

Scott Brown’s come-from-behind victory in the recent Massachusetts Senate race is widely attributed to his effective online communications and advertising model (for example, see An Online Model for the GOP?).

As part of that online model, in the final days leading up to the election, the Brown campaign trafficked an interactive Mixpo Xspot across the Boston Herald online edition. During the 10-day Xspot campaign, the paper served almost 1.5 million impressions.

  • Voters chose to watch and engage with Brown’s Xspot nearly twice as often as the industry standard for online video ads.
  • The Xspot campaign’s clickthrough rate exceeded the typical rate for display ads by almost double.
  • On average, viewers watched 78 percent of the Xspot.
  • Brown’s message was communicated for a total of 267 hours.

To learn more about how easy it was for the Brown campaign to convert an existing 30-second TV spot into an interactive online Xspot and to use the Xspot to respond to breaking news, read the full case study.

Find more information and resources on our political microsite.

Tags: , , , ,
Posted in Case studies, Measurement and metrics, TV and Xspots | No Comments »

Monitor Xspot views by DMA

Thursday, December 10th, 2009

Online video advertising is increasingly local. The Kelsey Group predicted that local online video ad revenue would reach $1.5 billion by 2012.

In an interview with Fred Aun of ClickZ, Matt Booth, SVP and program director for The Kelsey Group’s Interactive Local Media practice, explained the revenue growth this way: Local video ads are effective and result in action. Kelsey’s "User View" study found that 47.3 percent of the people who watched a local video ad visited the advertiser’s website.

In keeping with these predictions, large and small advertisers are jumping on the local video bandwagon. National advertisers are increasingly turning national campaigns local by geotargeting calls to action and contact information.

With this trend toward local advertising, advertisers need to know if their ads are reaching the intended audience. That’s why Mixpo recently supplemented the Viewer Locations map that we’ve always provided with a new analytics feature that allows advertisers to monitor Xspot performance by designated marketing area (DMA).

Viewer Locations map

Viewer Locations map we've always provided

By generating a custom Xspot report in the Dashboard, you can immediately see a list of the 210 U.S. DMAs arranged in descending order by the number of views that occurred in each.

Portion of DMA list in custom report

Portion of DMA list in custom report - Click for larger image

How DMA data can shape campaigns

Suppose a state agency is running a campaign for the local lottery. By examining the DMA view data (especially in combination with the Domain filter), the agency can tell how effectively publishers are delivering impressions to the target audience. Based on the data, the agency can modify its media plan or work with existing publishers to improve targeting.

Suppose a travel agency is running an Xspot campaign to promote discounted fares to the Caribbean. Based on click data from the DMA list, the agency can identify cities where the campaign generates the most activity. This gives the travel agency valuable insight into shaping future campaigns by directing impressions toward the most active markets.

More information

Mixpo subscribers can learn more about monitoring performance by DMA by visiting the Reports page, and then clicking Can I monitor Xspot views by DMA?

Tags: , , , , ,
Posted in Measurement and metrics, Product features | No Comments »

« Older Entries
Newer Entries »