double click creative insights on rich media

8/9/2019 Double Click Creative Insights on Rich Media

1/14

Creative Insights on Rich MediaWhat aggregate trends in ad serving data can tell you about

how to build better rich media creatives

This report details much that aggregate trends in ad serving data can tell you about click-

through rate, interaction rate, interaction time, expansion rate, expansion time and video

complete rate. The data are based on hundreds o advertisers, thousands o campaigns

and tens o billions o ad impressions. Most o these metrics can be used as an indicator o

success or your rich media campaigns.

In each section, well ocus on a specic metric o success. Well show you how the choice

o ad ormat and creative size can aect campaign perormance. Well end each section

with industry benchmarks and advice on how to compare your campaigns perormance to

that o other advertisers in the same industry vertical.

Some examples in this report showcase how goal setting can inuence creative decisions.

Well show you how the choice o a specic ormat or creative size can positively aect

one or two metrics, while negatively aecting a third metric.

For best campaign results, our advice is to set measurable campaign goals up-ront and

keep these goals in mind throughout the creative process. Advertisers should prioritize

each o the metrics eatured in this report and aim to beat the industry benchmarks or

their top one or two goals.

Click-Throgh Rate

Not every click in a rich media unit produces a click-through. When a viewer clicks on a

rich media unit, a number o outcomes are possible, including expanding the unit, playing a

video or otherwise interacting with the unit. The call-to-action or a viewer to click-through

competes with all other possible orms o interaction within the unit. As a result, click-

through rates can be lower or rich media than or other orms o online advertising where

the click-through is the only orm o user interaction.

AuTHOR

Sally Cole

Business Marketing Manager

Google

Click-Through Rate ............................................................................................ 1

Interaction Rate .................................................................................................. 3

Average Interaction Time .................................................................................. 5

Expansion Rate ................................................................................................... 7

Average Expanding Time ................................................................................... 8

Video Complete Rate ...................................................................................... 10

Conclusion ....................................................................................................... 12

September 2008

research

research report


2/14

2

Creative Insights on Rich Media

On average, DoubleClick Rich Media ads produce a 0.10% click-through rate. To enhance click-through

results, advertisers can make a ew simple creative choices.

Figre 1 shows that in-page video provides better click-through results than in-page non-video and

expanding ormats.

Figre 2 shows that larger creative sizes tend to produce higher click-through rates. Consider using creative

sizes such as the hal page ad, large rectangle or medium rectangle to get the best click-through results.

Consider Video or a Higher Click-Throgh RateFigre 1

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media in-page and expanding ormats only, January July 2008.

0.12%

0.10%

0.09% 0.09%

0.00%

0.02%

0.04%

0.06%

0.08%

0.10%

0.12%

0.14%

In-Page Video Expanding

Video

Expanding

Non-Video

In-Page

Non-Video

DoubleClick Rich Media Format

Click-ThroughRate

Consider Larger Creative Sizes or a Higher Click-Throgh RateFigre 2

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media in-page and expanding ormats only, January July 2008.

0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

CTR %

Creative Size

Click-ThroughRate

Area(inPixels)

Area (in Pixels)

HalfB

anne

r(23

4x60)

FullB

anne

r(468x

60)

Lead

erboa

rd(728

x90)

Rectan

gle(180

x150

)

Skys

crape

r(12

0x600)

Wid

eSky

scrap

er(160

x600

)

Medium

Rectan

gle(300

x250

)

LargeR

ecta

ngle

(336

x280

)

HalfP

ageA

d(30

0x600)


3/14

3


Campaign perormance can be measured against industry benchmarks or click-through rates.

Figre 3 shows that advertisers can expect to get anywhere between a 0.06% and 0.17% click-through

rate depending on the industry. A good measure o success is to beat the industry benchmark or

click-through rate. For example, a CPG campaign should aim to get a click-through rate above 0.11%.

Interaction Rate

Interaction rate is a popular measure o rich media campaign perormance. The metric places value

on interactions within a unit, even i they do not result in a click-through. (See Appendix A or more

inormation on how interactions are calculated.)

Advertisers can make simple creative choices to enhance interaction rates.

Figre 4 shows that in-page video has the highest perorming interaction rate in comparison to in-

page non-video and expanding ormats. In addition, in-page non-video has a slightly better perorming

Aim to Beat the Indstry Click-Throgh Rate BenchmarkFigre 3

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media in-page and expanding ormats only, January July 2008

0.17%

0.12% 0.12%0.11% 0.11%

0.10% 0.10% 0.10%0.09%

0.08% 0.08%0.07%

0.06%

0.00%

0.02%

0.04%

0.06%

0.08%

0.10%

0.12%

0.14%

0.16%

0.18%

Click-ThroughRate

Industry Category

Financial

Services

Classifieda

ndLocal

Travel

Telec

om

Auto

W

ellness

Miscella

neou

sB2

BCP

G

Brands

Techn

ology

Retai

l

Media

/Entert

ainme

nt

Consider In-Page Video or a Higher Interaction RateFigre 4

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media in-page and expanding ormats only, April July 2008.

2.45%

1.79%

1.61%

1.30%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%


Interaction

Rate

Expanding Non-VideoExpanding VideoIn-Page Non-VideoIn-Page Video


4/14

4


interaction rate than expanding ormats. Consider using in-page ormats, especially in-page video, to

achieve a higher interaction rate.

Figre 5 shows a strong relationship between ad size, as determined by total pixel area (pixel width x

pixel height), and interaction rate. To get a better interaction rate, consider using the larger creative ad

sizes, such as the 300x600 hal page ad, 336x380 large rectangle, 300x250 medium rectangle or 160x600

wide skyscraper. Note that the 728x90 leaderboard has the lowest interaction rate, despite taking up

over 60,000 square pixels o screen real estate. We believe this is due to viewer atigue o leaderboard

creatives caused by the high number o impressions and routine page positioning.

Campaign perormance can be measured against industry benchmarks or interaction rates. Figre 6 shows

that advertisers can expect to get anywhere between a 1.52% and 3.27% interaction rate, depending

on the industry. A good measure o success is to beat the industry benchmark or interaction rates. For

example, a Travel campaign should aim or an interaction rate above 1.91%.

Consider Larger Creative Sizes or a Higher Interaction RateFigre 5


0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

5.00%

Half Page Ad

(300x600)

Large

Rectangle

(336x280)

Medium

Rectangle

(300x250)

Wide

Skyscraper

(160x600)

Skyscraper

(120x600)

Full Banner

(468x60)

Rectangle

(180x150)

Leaderboard

(728x90)

InteractionRate

Creative Size

Area(inPixels)

Interaction Rate % Area (in Pixels)

Aim to Beat the Indstry Interaction Rate BenchmarkFigre 6


3.27%

2.55%2.47%

2.35%2.21% 2.17%

2.04% 1.99%1.91% 1.87% 1.84%

1.56% 1.52%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

Miscellaneo

us

Wellness

B2

B

Retail

Technolog

yCP

G

Classifie

dand

Local

Media/Entert

ainme

nt

Travel

Fina

ncial

Services

Brands

Auto

Telec

om

Interaction

Rate

Industry Category


5/14

5


Average Interaction Time

Average interaction time is a measure o how long viewers are interacting with an ad unit. (See Appendix B

or the ull denition.)

The amount o time users spend interacting with DoubleClick Rich Media units is just over 11 seconds

on average. There are very ew gains to be made to average interaction time by controlling the ormat or

creative size o the ad units. Rather, interaction time is dependent on how interesting and engaging your

creative is.

Figre 7 shows there is only a hal-second gain to be made by selecting an in-page non-video ormat over

an in-page video ormat.

Format Choice Plays Little Role in Average Interaction TimeFigre 7


11.9 11.8 11.6 11.4

0

2

4

6

8

10

12

In-Page Non-Video Expanding Video Expanding Non-Video In-Page Video


AverageInteractionTime(sec)


6/14

6


Figre 8 shows that creative size selection can produce gains in average interaction time o up to 1.9

seconds. While this is our times the hal-second gain made possible through ormat selection, its hardly

enough o an improvement to dictate creative size selection.

I interaction time is an important goal o your campaign, we suggest the use o highly interactive eatures

such as polls, games and sweet-spot video.

The benchmarks or average interaction time show little deviation rom the norm. Figre 9 shows that

advertisers can expect to get between a 10.4- and 13.1-second average interaction time. Advertisers can

assume that their DoubleClick Rich Media campaign will have an interaction time within this range.

Choice o Creative Size Also Plays Small Role in Average

Interaction Time

Figre 8


12.47 12.21 12.1111.68 11.57

11.0010.57

0

2

4

6

8

10

12

14

Skyscraper(120x600)

Leaderboard(728x90)

LargeRectangle(336x280)

Half Page Ad(300x600)

MediumRectangle(300x250)

WideSkyscraper(160x600)

Full Banner(468x60)

Creative Size

Average

InteractionTime(sec)

Indstry Average Interaction Time BenchmarksFigre 9


13.06 13.0312.65 12.51

11.64 11.6011.23 11.21 11.13 10.88 10.70 10.54 10.39

0

2

4

6

8

10

12

14

FinancialS

ervices

Teleco

m

Travel

Technolog

y

Auto

B2B

Retail

Media

/Entert

ainmen

t

CPG

Miscellaneou

s

Classifie

dand

Local

Wellnes

s

BrandsA

verageInteractionTime(sec)

Industry Category


7/14

7


Expansion Rate

Expanding ad units contain rich content and eatures that can only be accessed by viewers when they

expand the ad. (See Appendix C or the denition o expansion rate.)

On average, 1.94% o expanding ad impressions served get expanded by viewers. Advertisers can make

simple creative choices to enhance expansion rates.

Figre 10 shows that the expansion rate improves rom 1.85% to 2.18% when comparing expanding

units without video to expanding units with video. To achieve the highest expansion rate possible,

consider using video in your expanding units.

Figre 11 shows that use o the 336x280 large rectangle results in the highest expansion rate o the six

creative sizes measured at 3.90%. To improve expansion rates, try using more large rectangles.

Consider Video or a Higher Expansion RateFigre 10

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media expanding ormats only, January July 2008.

2.18%

1.85%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

Expanding Non-VideoExpanding Video

Expansion

Ra

te


Choose Creative Sizes that Prodce a Higher Expansion RateFigre 11


3.90%

2.33%

2.10%2.01%

1.79%

1.46%

0.0%

1.0%

2.0%

3.0%

4.0%

Large Rectangle(336x280)

Skyscraper(120x600)

Leaderboard (728x90)



Wide Skyscraper(160x600)

Exp

ansionRate

Creative Size


8/14

8


Campaign perormance can be measured against industry benchmarks or expansion rates. Figre 12

shows that advertisers can expect to get anywhere between a 1.24% and 4.74% expansion rate

depending on the industry. A good measure o success is to beat the industry benchmark or expansion

rates. For example, a Wellness campaign should aim to get an expansion rate above 2.75%.

Average Expanding Time

To make the most o expansion activities, advertisers can aim to increase the amount o time viewers

spend within the expanded creative unit. (See Appendix D or the denition o average expanding time.)

Figre 13 shows that or best average expanding time, avoid the 468x60 ull banner.

Together, Figure 13 and Figure 11 show that one creative size cant always address multiple perormance

goals. In Figure 13, the wide skyscraper shows the highest expanding time at 9.5 seconds. In Figure 11,

Aim to Beat the Indstry Expansion Rate BenchmarkFigre 12

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media expanding ormats only, January July 2008 .

4.74%

3.98%

3.69% 3.65% 3.62%3.39%

2.83% 2.75%2.61%

2.16%

1.52%1.32% 1.24%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

Retail Miscellaneous Travel Technology CPGMedia /

Entertainment Telecom

ExpansionRate

Industry Category

Creative Sizes to Consider or Best Expanding TimeFigre 13


9.45

7.386.89 6.79 6.75

6.50

2.77

0

1

2

3

4

5

6

7

8

9

10

WideSkyscraper(160x600)



Leaderboard(728x90)

LargeRectangle(336x280)

Skyscraper(120x600)

Full Banner(468x60)

verage

xpan

ng

me

sec

Creative Size (Pre-Expansion)


9/14

9


the wide skyscraper shows the lowest expanding rate at 1.46%. While a relatively ew number o viewers

are expanding the wide skyscraper, once they have expanded it, viewers are spending more time in the

unit than any other creative size. This exemplies the importance o deciding up-ront the perormance

metrics that will best drive the overall success o the campaign.

Figre 14 shows that the use o video in an ad unit provides a 2.8-second increase to average expanding

time. For best expanding times, use video.

Campaign perormance can be measured against industry benchmarks or average expanding time.

Figre 15 shows that advertisers can expect to get anywhere between a 4.5- and 13.0-second average

expanding time. A good measure o success is to beat the industry benchmark or average expanding

time. For example, a Telecom campaign should aim to get an average expanding time above 8.1 seconds.

Add Video to Improve the Expanding TimeFigre 14


9.37

6.63

0

1

2

3

4

5

6

7

8

9

10

Expanding Video Expanding Non-Video

AverageExpandingTime(sec)


Aim to Beat the Indstry Expanding Time BenchmarkFigre 15


13.00

10.92

9.85

9.06

8.25 8.05 7.81

6.92 6.686.30 6.27

4.864.47

0

2

4

6

8

10

12

14

Classifie

dand

Local

Brands

Travel

Media

/Entert

ainme

nt

Auto

Telec

om

Technolog

y

Miscellaneo

us

FinancialS

ervice

sCP

G

Retai

l

Wellness

B2

BAvera

geExpandingTime(sec)

Industry Category


10/14

10


Video Complete Rate

The video complete rate is a measure o how many o video impressions served and commenced to play

were watched through to the end o the video advertising message. (See Appendix E or inormation on

how video completes are measured.)

Advertisers can make a ew simple creative choices to enhance video complete rates.

Figre 16 shows that 40% o video plays are completed in the expanding video ormat. An even larger

55% o video plays are completed in the in-page video ormat. These numbers speak to the eectiveness

o video at delivering an advertisers message to audiences.

The choice between auto-play and user-initiated video will aect video complete rates. Our aggregate

data skews toward auto-play video. We believe that advertisers using user-initiated video should expect

lower video complete rates than those in Figure 16.

Consider In-Page Video or a Higher Video Complete RateFigre 16

Source: DoubleClick DART or Advertisers, U.S. advertisers, DoubleClick Rich Media in-page video and expanding video ormats only,

January July 2008.

55%

40%

0%

10%

20%

30%

40%

50%

60%

VideoCompleteRate%


Expanding VideoIn-Page Video


11/14

11


Figre 17 shows three creative sizes that garner, on average, video complete rates above 50%. These

creative sizes include the 336x280 large rectangle, the 160x600 wide skyscraper and the 728x90 leaderboard.

It also shows the 300x600 hal page ad with a below average video complete rate. I you recall rom

Figure 2 and Figure 5, the hal page ad was the best perorming creative size by click-through rate and by

interaction rate. This highlights the importance o dening goals early on in the campaign, since the same

creative size may be a good choice or a click-through or interaction goal and a bad choice or a video

complete goal.

Campaign perormance can be measured against industry benchmarks or video complete rate.Figre 18

shows that advertisers can expect to get anywhere between a 32% and 64% video complete rate. A good

measure o success is to beat the industry benchmark or video complete rate. For example, a Financial

Services campaign should aim to get a video complete rate above 60%.

Choose Creative Sizes that Prodce a Higher Video Complete RateFigre 17


January July 2008.

58%56%

52%

44%

38% 38%

0%

10%

20%

30%

40%

50%

60%

70%

Large Rectangle(336x280)

Wide Skyscraper(160x600)

Leaderboard(728x90)

Medium Rectangle(300x250)

Skyscraper(120x600)


Creative Size

VideoCompleteRa

te

Aim to Beat the Indstry Video Complete Rate BenchmarkFigre 18


January July 2008.

64% 63%60% 59% 58%

53%50% 48%

45%

38%36%

32%

0%

10%

20%

30%

40%

50%

60%

70%

Industry Category

VideoCompleteRate

Classifie

dand

Local

Trave

l

Media

/Entert

ainme

nt

Telec

om

Auto

Retail

Misc

ellan

eous

Finan

cialS

ervice

s

Tech

nolog

y

Bran

ds

CPG

Wellne

ss


12/14

12


Conclsion

This report reviewed what aggregate trends in ad serving data can tell you about click-through rate,

interaction rate, interaction time, expansion rate, expansion time and video complete rate. You can use

each o these metrics as an indicator o success or your rich media campaigns.

In each section, we ocused on a specic metric o success and provided some key takeaways:

For higher click-through rate, consider using video and larger creative sizes.

For higher interaction rates, consider using in-page units, especially in-page video units, and larger

creative sizes.

When measuring average interaction time, expect something close to the average, which is just

above 11 seconds.

For higher expansion rates and expansion times, consider using expanding video units.

For higher video complete rates, consider using in-page units.

Some examples in this report showcased how goal setting can inuence creative decisions. We

demonstrated how the choice o a specic ormat or creative size can positively aect one or two

metrics, while negatively aecting a third metric.

For best campaign results, our advice is to set measurable campaign goals up-ront and keep these goals

in mind throughout the creative process. Advertisers should prioritize each o the metrics eatured in this

report and aim to beat the industry benchmarks or their top one or two goals.

Appendix A: Measring Interactions

Theres currently no standard way to measure interactions across the online advertising industry.

Each rich media vendor calculates interactions in a dierent way, and many vendors do not disclose details

on their methodology. As a result, it is impossible to compare interaction rates between vendors in a

meaningul way.

In lieu o a measurement standard or interactions, we recommend two key best practices or working with

interaction metrics.

Only compare and contrast data that comes rom the same system and that adheres to the1.

same methodology.

For example, the interaction methodology or DoubleClick Rich Media was upgraded on February

2, 2008. This means that any DoubleClick Rich Media campaign with creative uploaded between

February 2, 2008 and the present can only be accurately compared to other DoubleClick Rich

Media campaigns or benchmarks on the same timeline. Similarly, DoubleClick Rich Media campaigns

with creative uploaded on or beore February 2, 2008 can only be accurately compared to otherDoubleClick Rich Media campaigns or benchmarks on the same timeline. Campaigns that have

some creatives uploaded under the new methodology and other creatives uploaded under the old

methodology should not be compared to any other campaign or to any benchmark due to the use

o mixed methodologies.


13/14

13


Know the defnitions and methodology behind the metrics.2.

For DoubleClick Rich Media, the methodology or interaction metrics is as ollows:

Interaction Rate: The ratio o DoubleClick Rich Media ad interactions to the number o rich media

ad impressions displayed. This number is reached using the ollowing calculation:

Interactions Rich Media Impressions

Interactions: The number o times that a user interacts with a DoubleClick Rich Media ad.

Interactions are captured when the user does one or more o the ollowing:

Clicks an Exit link-

Makes the ad display in Full Screen mode-

Mouses over the ad or 1 continuous second-

Rich Media Impressions: The number o times that a DoubleClick Rich Media ad unit is displayed.

Appendix B: Defnition o Average Interaction Time

Average Interaction Time: The average amount o time, in seconds, that a user interacts with a

DoubleClick Rich Media ad. Multiple interactions with an ad during a single ad view are aggregated.

Appendix C: Defnition o Expansion Rate

Expansion Rate: Expansion rate is calculated by dividing expansion time counters by DoubleClick Rich

Media expanding ad impressions.

Appendix D: Defnition o Average Expanding Time

Average Expanding Time: The average amount o time, in seconds, that an expanding ad is viewed in an

expanded state. Any expansion times that exceed our minutes are capped. This extended expansion time

can occur, or example, when a user opens an expanding ad on his or her browser, then steps away or an

hour without collapsing the ad or closing the browser. The capping rule helps prevent skewed results.

Appendix E: Measring Video Completes

The video complete rate is calculated by the ollowing ormula and denitions:

Video Completes Video Plays

Video Completes: The number o times a video played to its completion. For example, the number o

times a 15-second video plays or its ull 15 seconds.

Video Plays: The number o impressions where a video was played.

Appendix F: Abot the Sorce Data

The data contained in this report are rom DoubleClicks Ad Serving Trend Report Application (ASTRA).

DoubleClick has built this robust sotware tool to analyze online advertising campaign activity across its

DART ad serving platorm. Presently, the tool reports click-through rates, interaction rates, interaction

times, expansion rates, expansion times and video complete rates or ads or thousands o advertisers

using the DART or Advertisers (DFA) and DoubleClick Rich Media platorms.

These data are careully normalized to reect industry norms to the best o our ability. The Advertising

Research Foundation (ARF) was consulted on the design o ASTRA and advised on aspects o its


14/14


www.doubleclick.com

2008 Google Inc. DoubleClick is a division o Google Inc. All rights reserved.

US Headquarters

76 Ninth Avenue

6th Floor

New York, NY 10011

Tel: 212-683-0001

Tel: 866-683-0001 (toll ree)

Fax: 212-287-1203

EMEA Headquarters

Gordon House

Barrow Street

Dublin 4

Ireland

Tel: +353 1 543 0100

Fax: +353 1 436 1001

Asia Pacifc Headquarters

Suite 19, Level 1

88 Cumberland Street

The Rocks,

Sydney, NSW 2000

Australia

Tel: +61 (0)2 8247 3700

Fax: +61 (0)2 8247 3701

methodological design, including the use o medians instead o mean averages or the calculation o the

benchmark metrics noted above.

Data shown here represent activity o a wide range o ad ormats or DFA advertisers in the United States.

The majority o data are rom January 2008 to July 2008. The data on interaction rates and interaction

times are rom April 2008 to July 2008.

ASTRA reports on a data set o the activity o thousands o DFA advertisers that have been categorized

by industry and country geography. Industry categories are dened by the sub-categories they represent.

To make category assignments, each DFA advertiser is assigned to a single industry sub-category. Sub-

category assignments then roll-up into category assignments. For example, the Brands industry category

is a roll-up o advertiser data rom six sub-categories: Apparel Brands, Household Brands, Luxury Brands,

Miscellaneous Brands, Ofce Brands and Sports Brands. The Miscellaneous category reports on advertiser

data rom the collection o advertisers that dont clearly t into any industry category or sub-category.

To ensure statistical soundness as well as client condentiality, minimums have been applied to the data

sets that can be reported on: at least our advertisers and our DART networks must be represented or any

metric to be reported at any dimension o the data (e.g., by product category, by ad size, by time period).

Abot DobleClick

DoubleClick is a premier provider o digital marketing technology and services. The worlds top marketers,

publishers and agencies utilize DoubleClicks expertise in ad serving, rich media, video, mobile, search

and afliate marketing to help them make the most o the digital medium. From its position at the nerve

center o digital marketing, DoubleClick provides superior insights and insider knowledge to its customers.

DoubleClick is a division o Google Inc. (NASDAQ: GOOG). Learn more at www.dobleclick.com

double click creative insights on rich media

Documents