insights - Playable Video

insights - Playable Video

INSIGHTS

Q4 2017

TABLE OF CONTENTS

Introduction ………..……………………………....................... 3

Executive Summary ………………………………………………… 4

Mobile vs. Desktop ………..…………………………………..…… 6

Device / Operating System …………………………..………… 7

Email Client ……..…………………………………………………….. 8

Day Part ……………..……………………………………………….…. 9

Network Connection ……….……………………………………… 10

Engagement ………….……………………………………………….. 12

Methodology ………..………………………………………………… 13

Playable Insights Q4 2017

Playable Insights Q4 2017

2

INTRODUCTION

Playable upgrades email marketing to deliver high quality native autoplay video.

This edition of Playable Insights focuses on

Dwell Time

how long do viewers

spend watching video emails?

Google recently added previews of

YouTube videos to its search results, and chose to present each of these previews as a silent six-second loop.

That’s the same duration that Vine and

Twitter decided upon for their videos too.

Facebook has decided that 3 seconds is the significant duration for which advertisers should pay for impressions.

As you will find in this report, video emails are enjoyed for much longer than

3 seconds.

Is a longer or shorter Dwell Time more desirable? The answer may not always be the same for viewers and for marketers, who both have varying objectives – sometimes wanting to maximize video content consumption, sometimes wanting to use video to inspire a call-to-action as quickly as possible.

Playable can help with both cases, for example:

• Pre-theatrical movie trailers, branding content, training and education videos where one of the key metrics is high

Dwell Time

• Travel, ticketing and retail marketing where the objective is to increase

click-to-open-rate

and

downstream engagement metrics

Playable reports Dwell Time in real-time during video email campaigns so customers can easily analyze how long their audience is watching their video emails, can compare different campaigns, and can optimize to their particular marketing objectives.

See

https://playable.video

to get started on our Free plan that includes

5,000 Plays per month.

Playable Insights Q4 2017 3

EXECUTIVE SUMMARY

How long do viewers spend watching a video email?

80% of viewers watch for up to 45 seconds

80% of viewers watch for 12 seconds or more

50% of viewers watch for 22 seconds or more

Playable Insights Q4 2017 4

The graph above shows the percentage of viewers who watch a video email for a period of time.

80% of viewers watch for 12 seconds or more.

Playable upgrades email marketing to deliver high quality native video. The video autoplays as soon as the email is opened, and loops for as long as the viewer is watching. A video duration of 5 to 10 seconds is recommended.

Half of all engaged viewers will watch a video email for 22 seconds or more, or at least two to four loops of a

Playable video.

Another way to read the data is that

80% of viewers watch for up to 45 seconds.

There is a long tail of viewers who watch beyond 60 seconds, which is not shown on the graph above.

The rest of this report compares how

Dwell Time varies according to factors such as device operating system, timeof-day, and click-to-open-rate.

Other factors such as the email client of the viewer, or their network connection, reveal little variation in

Dwell Time.

In fact the aggregate data presented here is used by Playable for the benefit of all customers, to tune its algorithms such that all viewers, regardless of email client, device or network connection, all get the best possible video experience.

There are some interesting surprises in the data.

Playable Insights Q4 2017 5

MOBILE vs. DESKTOP

Mobile

Desktop

Playable data reveals that 62% of video emails are watched on a mobile device.

Does Dwell Time vary for mobile vs.

desktop viewers?

The simple answer is No! Our research found that viewers spend just as much time watching a video email on a mobile device as they do on a desktop computer.

Playable adjusts the video payload for each recipient depending on device, screen resolution, and network conditions, to ensure that every viewer gets a great video email experience.

Playable Insights Q4 2017 6

DEVICE / OPERATING SYSTEM

Even though mobile and desktop viewers watch for the same duration on average, digging deeper into the data reveals some interesting variations when we compare iOS vs. Android, and

Windows vs. Mac OS. iOS and MacOS viewers watch less / click quicker

On average, Android viewers watch up to 30% longer than iOS viewers.

Android and Windows viewers watch more / click slower

We leave it up to you, the reader, to theorize about why this happens.

Playable Insights Q4 2017 7

.

EMAIL CLIENT

How does Dwell Time vary for different email clients?

Apple Mail on Mac OS has a slightly smaller % of viewers who watch for 20 seconds or more

Playable delivers many different video formats to various different email clients.

Gmail on Android has the largest proportion of audience who watch for 30 seconds or more

However there is not much variation in

Dwell Time between different email clients.

Playable Insights Q4 2017 8

.

DAY-PART

Playable detects the location and timezone of each viewer, and calculates the local time and day-part.

This data can be used in real-time to vary the video content. For example,

Quick Service Restaurants have no way of predicting when each recipient will open his or her email, but can use

Playable to present different timesensitive videos at breakfast vs. lunchtime vs. dinner time.

Lunchtime (12noon – 2pm) has the lowest proportion of viewers who spend 20 seconds or more watching video emails

Playable Insights Q4 2017 9

NETWORK CONNECTION

The majority of emails are now opened on mobile devices – 62% as measured by

Playable.

However, the majority of those devices are connected to the internet using WiFi rather than a cellular network.

Overall, just 19% of video emails are watched over a cellular network.

Playable detects the network connection for each viewer, so we can identify slower dialup and cellular connections and deliver a more compressed video payload, so that every viewer gets the best possible video experience.

Playable Insights Q4 2017 10

How does dwell time vary on different network connections?

Dialup viewers (less than

0.3% of all viewers) have slightly lower Dwell Times.

There is very little difference in Dwell

Time between viewers connected over cellular, cable and corporate networks.

Viewers connected over a Corporate network are slightly more likely to watch for more than 30 seconds.

In fact, this aggregate data is used by

Playable to tune its video compression algorithms such that all users get a great experience, regardless of their network connection.

Playable Insights Q4 2017 11

.

ENGAGEMENT

How does Dwell Time vary when a video email campaign has higher-than-average or lower-than-average engagement, as measured by Click-To-Open-Rate (CTOR)?

Compared to average CTOR, both higher engagement and lower engagement video emails have generally lower dwell times. This is an interesting result.

One explanation is that both extremes of engagement have a similar impact.

An audience that is not engaging well desires fewer loops of the video, and conversely the most engaging content is persuasive enough that viewers take action more quickly, requiring fewer loops of the video.

Playable Insights Q4 2017 12

METHODOLOGY

This report reflects anonymized and aggregate data from a random sample of 10 million video emails watched during November 2017.

Playable compared four methods to measure Dwell Time: a) Throttled image server. It is possible to serve a large image slowly and measure the duration until the HTTP connection is broken. However, this method is vulnerable to proxies such as used by Gmail, and intermediate image caches such as used in mobile networks. It also consumes more resources than is necessary for viewers, so we decided against using this method. b) HLS segments. Playable uses HLS

(HTTP Live Streaming) to deliver video to some viewers. It’s possible to detect which HLS segments are requested by each viewer, as a measure of how long each viewer is watching. However, we decided against this method because different email clients and devices vary their pre-fetching strategies based on different network conditions, which introduces too many confounding variables. c) CSS animations. It’s possible to include a CSS animation in an email that generates requests for images over a period of time, and these requests can be used as a measure of dwell time. However, this approach only works on a subset of email clients and is fragile to implementation changes.

Therefore we settled upon method d) measuring the time delta between

Open and Click events. This includes only the subset of recipients who click on the video in a video email, but is independent of a number of variables that are useful to analyze, in this report and by customers of Playable when comparing different video email campaigns.

Playable Insights Q4 2017 13

THANK YOU

Thanks for your interest in Playable and video email marketing.

We welcome all your feedback on this

Playable Insights Q4 2017 report, and your requests for additional analysis in our future white papers.

Our next report coming in 2018 will focus on

Content Type

… but not the

HTTP Content-Type you might be expecting!

Playable is headquartered in Silicon

Valley, with offices in Sunnyvale and

Sydney, and customer success operations in New York and

Melbourne.

For more details, please contact:

Bob Hitching

CEO

Playable, Inc.

+1 650 289 8577 [email protected]

14

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement