Improving Ad Performance Online:
The Impact of Advertising on
Content Sites
Wave III
August 2009
www.online-publishers.org

Research Objectives & Method
• Objective: To understand the effectiveness of advertising placed in
different environments including Content Sites, Ad Networks, Portals and
the industry as represented by MarketNorms® data.
– MarketNorms is the industry standard method for measuring online
advertising’s effectiveness and branding impact
– As of Q1 2009, MarketNorms’ global database holds 4,882 campaigns, based
on over 7 million survey respondents
• This is the third in a series of ad effectiveness studies leveraging
MarketNorms. This report offers updates and new analyses, including
– Online advertising’s impact on decision makers and influencers
– Expanded industry ad effectiveness research, now covering Automotive and
Retail
– Effectiveness of the most popular ad sizes: Leaderboards (728x90), medium
rectangles (300x250) and wide skyscrapers (160x600)
2

Research Objectives & Method
• Ad Effectiveness Metrics: Survey questions for the ad effectiveness
metrics are typically worded like the following
– Aided Brand Awareness – Have you heard of the following brands of product
category?
– Online Brand Awareness – Have you seen the following brands of product
category advertised online in the past 30 days?
– Message Association – Which of the following brands, if any, uses the following
message in its advertising? “brand message here”
– Brand Favorability – How would you describe your overall opinion about each
of the following brands of product category?
– Purchase Intent – Next time you are looking to purchase a product, how likely
are you to consider each of the following brands?
3

Site Categories in this Study
• MarketNorms – The industry benchmark for online ad effectiveness,
with results from 4,882 AdIndex® research studies based on 7 million+
respondents
• Content Sites – OPA Members in the aggregate are a proxy for
Content Sites. OPA Members are media organizations whose focus is
creating professional content (see next slide for OPA Member roster)
• Portals – Provide multiple functions such as search, e-mail, directories
and third-party content. Examples include AOL, MSN and Yahoo
• Ad Networks – Aggregators and sellers of non-premium ad inventory,
typically across small to medium size third-party sites. Examples include
Advertising.com, Broadband Enterprises, Glam Media, Gorilla Nation,
Specific Media, Undertone Networks, ValueClick and Vibrant Media
4

OPA Members
5

Topline Findings

Ads on Content Sites Have Greatest Impact
• Ads on content sites raise Awareness, Message Association, Brand
Favorability and Purchase Intent more than Portals and Networks
• Ad Networks provide advertisers with the smallest change – including no
change in Purchase Intent (i.e., 0.2 delta is statistically insignificant)
Ad Effectiveness Deltas by Site Category
Aided Brand Online Ad Message Brand Purchase
Awareness Awareness Association Favorability Intent
A OPA BCD 3.2 5.7 3.4 2.4 1.8
B MarketNorms 2.3 4.7 2.5 1.5 1.2
C Portals 2.5 4.7 2.1 1.3 1.1
D Ad Networks 1.2 3.8 2.0 0.6 0.2
Notes: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=1,540 campaigns; MN=2,255; Portals=1,224; Ad Networks=399
7

Counter to Industry Trends, Ad Effectiveness
on OPA Sites Increased Over Time
• Portals’ experienced double digital declines across all metrics
• Ad Networks’ 2008 Purchase Intent delta was so low that even when it
doubled in 2009 it remains statistically insignificant, and 2009 Brand
Favorability is less than half of the industry average (0.6 vs. 1.5)
% Change in Ad Effectiveness Deltas, Q1 2009 vs. Q1 2008
Rolling 3 Years , Q1 2009 3-Year Deltas in Parentheses
Aided Brand Online Ad Message Brand Purchase
Awareness Awareness Association Favorability Intent
OPA +10% (3.2) +24% (5.7) +55% (3.4) +9% (2.4) +13% (1.8)
MarketNorm -4% (2.3) -4% (4.7) -4% (2.5) -6% (1.5) 0% (1.2)
s
Portals -14% (2.5) -13% (4.7) -16% (2.1) -19% (1.3) -21% (1.1)
Ad Networks -45% (1.2) -10% (3.8) -17% (2.0) +200% (0.6) +100% (0.2)
Notes: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
compared to last 3 years through Q1 2008 8

Ad Effectiveness by Demographic

Summary of Ad Effectiveness by Demographic
• Brand Opinions and Purchase Intent of self-identified Decision Makers and Influencers more
swayed by advertising run on OPA sites
– Ads on Ad Networks had insignificant Aided Brand Awareness, Brand Favorability and
Purchase Intent for this demo
– Purchase Intent on Portals was insignificant
• 18-34 year olds are more responsive to ads on OPA sites
– Brand Favorability scores are 229% greater than Ad Networks and 77% higher than for
ads on Portals
– Purchase Intent metrics are 78% greater than ads on portals
– Purchase Intent is negative on Ad Networks for 18-34 year olds
• Compared to industry averages, affluent audiences exposed to ads on OPA sites more
likely…
– Become aware of the advertised brand
– Remember the brand’s message
– Form favorable opinions about the brand
– Intend to make the purchase
• Sponsorships on OPA sites connect with up-scale consumers – with effectiveness deltas more
than triple that of sponsorships on portals
10

Brand Favorability of Decision Makers Most
Swayed by Ads on Content Sites
• Ad Networks failed to move decision makers and influencers’ Brand
Favorability and Purchase Intent
• Portals did not significantly increase Purchase Intent in the aggregate
*
A B C D
B
B
C
D D A
D
* Respondents who identified themselves as making or influencing purchase decisions.
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=146; MN=315; Portals=149; Ad Networks=57 11

OPA Sites Drove Greater Ad Effectiveness
Among 18-34 Year Olds
• 18-34 year olds exposed to ads on Ad Networks resulted in a negative
impact on Purchase Intent -- their Purchase Intent was lower than those
who did not see the same ads on Ad Networks
• Advertising on Ad Networks failed to increase Brand Favorability
A B C D
B B B
C C C B
C B
D D D C
D D
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
12
OPA N=795; MN=1,843; Portals=668; Ad Networks=66

Young Women are More Impacted by Advertising
Run on Content Sites Compared to Industry Averages
A B C
B B B B
C C C C
C
*Ad Networks excluded due to low number of campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=408; MN=1,282; Portals=392 13

Ad Awareness, Message Association and Purchase Intent
for Young Men More Impacted by Ads on OPA Sites
Compared to Portals
• There is no change in Purchase Intent for ads run on Portals
A B C
B B
C C C
*Ad Networks excluded due to low number of campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N= 277; MN=854; Portals=285 14

Sponsorships Among All 18-34 Year Olds on
OPA Sites are More Effective than on Portals
Sponsorship Deltas, 18-34 Year Olds*
Sponsorships are marketing efforts where the objective is to connect a brand with
a separate and identifiable event, person, place, content area or promotion
A B C
B B
C C B
*Ad Networks excluded due to low number of sponsorship campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=135; MN=213; Portals=47 15

Content Sites’ Advertising Raises Awareness
and Message Association Among the Affluent
A B C D
B B B
C C C B
D D D C B
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
A/B/C/D indicate statistically significant difference between deltas at .90 CL
OPA N=653; MN=1,458; Portals=451; Ad Networks=87 16

The Affluent are More Responsive to
Sponsorships on OPA Sites
Sponsorship Deltas, $75K+*
Sponsorships are marketing efforts where the objective is to connect a brand with
a separate and identifiable event, person, place, content area or promotion
BC
A
B
C
Average Delta
*Ad Networks excluded due to low number of sponsorship campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=121; MN=191; Portals=33 17

Ad Effectiveness by Industry

Ad Effectiveness by Industry Varies Based on
Environment
• Auto, CPG, Entertainment, Financial Services, Retail, Telecom and
Travel benefit by advertising with OPA Members – especially for
the more difficult to move metrics like Brand Favorability and
Purchase Intent
• Not all industries benefit by advertising with Ad Networks and
Portals
– Purchase Intent on Ad Networks is insignificant for Financial Services
– Purchase Intent on Ad Networks is insignificant for Technology
– Purchase Intent on Portals is insignificant for Retail
– Purchase Intent on Portals is insignificant for Travel
19

On 4 of 5 Metrics, Auto Ads on OPA Sites
Outperform the Online Industry
A B C
B
B B C B C
*Ad Networks excluded due to low number of auto campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=122; MN=194; Portals=109 20

When Advertising on OPA Sites, CPG Brands’ Impact on
Awareness, Favorability and Purchase Intent was Greater
A B C D
B B B B
C C B C C
D D C D D
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
A/B/C/D indicate statistically significant difference between deltas at .90 CL
OPA N=543; MN=779; Portals=431; Ad Networks=127 21

Entertainment Brands’ Advertising is Significantly
More Effective on OPA Sites than Portals
A B C
B B B B B
C C C C C
*Ad Networks excluded due to low number of entertainment campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=52; MN=62; Portals=37 22

Impact of Aided Brand Awareness and Favorability to
Financial Services is Greater on OPA Sites
• For ads seen on Ad Networks, the impact on Purchase Intent, Brand
Favorability, Message Association and Aided Brand Awareness metrics was
statistically insignificant
A B C D
B B
C B C C
D D D D D
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=194; MN=253; Portals=154; Ad Networks=55 23

Ad Awareness and Message Association for Retail
Brands is Greater on OPA Sites than Portals
• Retailers advertising on Portals has insignificant Purchase Intent and
Brand Favorability impact
A B C
B
C C
*Ad Networks excluded due to low number of retail campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=66; MN=115; Portals=71 24

Tech Ads on OPA Sites Outperform Portals and Ad
Networks for Aided Brand Awareness and Message
Association
• Tech ads run on Ad Networks have no impact on Purchase Intent
A B C D
B B
C C
D D C
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=161; MN=256; Portals=106; Ad Networks=65 25

Telecom Ads on Content Sites Drive Greater
Ad Awareness and Brand Favorability
• Portals provide no change in Aided Brand Awareness
A B C
B B
C C C
*Ad Networks excluded due to low number of telecommunications campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
26
OPA N=62; MN=72; Portals=47

Travel Ads are Most Effective When Run on
OPA Sites
• Travel ads in general (as reflected by MarketNorms) and on Portals do
not impact Purchase Intent
A B C
B B B B
C C C C
*Ad Networks excluded due to low number of travel campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=87; MN=102; Portals=53 27

Ad Effectiveness by Size & Format

Ad Effectiveness by Ad Size and Format Varies
Based on Environment
• The most popular ad sizes – leaderboards, medium rectangles and wide skyscrapers –
perform better on OPA sites than MarketNorms on most metrics
– Ad Networks provide insignificant Purchase Intent deltas on these most popular
ad units
• Video ads on content sites – represented by OPA Members – have greatest impact
– Video ads on OPA Members’ sites outperformed Portals across Awareness,
Message Association, Brand Favorability and Purchase Intent metrics
• Rich media ads more effectively engage these content site visitors – making such sites
better platforms for alternatives to banners
– Purchase Intent and Brand Favorability is insignificant on Ad Networks
• Consumers are more involved with interactive ads on content sites
– Purchase Intent is insignificant on Ad Networks
29

Leaderboards, Medium Rectangles and Wide
Skyscrapers are Most Popular
Source: Nielsen Online AdRelevance (May 25-May 31, 2009), n=94,885, house ads excluded
30

Leaderboards on OPA Sites Outperform
MarketNorms on Upper Funnel Metrics
• Leaderboards run on Ad Networks have no impact on Purchase Intent,
Brand Favorability or Aided Brand Awareness
A B C D
B
C B
B D C C
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=450; MN=1,360; Portals=511; Ad Networks=152 31

Medium Rectangles on OPA Sites Outperform Portals
and Ad Networks on Three of Five Metrics
• Online Ad Awareness, Message Association, Brand Favorability and
Purchase Intent have significantly increased on OPA sites
• Medium Rectangles on Ad Networks had no impact on Purchase Intent,
Brand Favorability or Aided Brand Awareness
A B C D
B B
C C B
D D D C
D
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=349; MN=1,430; Portals=629; Ad Networks=119 32

Wide Skyscrapers on OPA Sites Outperform
MarketNorms Awareness and Message Association
• Wide skyscrapers on Ad Networks have no impact on Purchase Intent
or Aided Brand Awareness
A B C D
B B
D D B
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=198; MN=682; Portals=136; Ad Networks=60 33

Video Ads Are Most Effective on OPA Sites
A B C
B B B B B
C C C C C
*Ad Networks excluded due to low number of video campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=59; MN=240; Portals=132 34

Counter to Industry and Portal Trends, Content Site
Video Ad Effectiveness has Increased on Most Metrics
% Change in Video Ad Effectiveness Deltas,
Q1 2009 vs. Q1 2008 Rolling 3 Years
Aided Brand Online Ad Message Brand Purchase
Awareness Awareness Association Favorability Intent
OPA +27% +8% +20% +4% -3%
MarketNorms -33% -29% -42% -19% -21%
Portals -41% -25% -49% -33% -40%
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
compared to last 3 years through Q1 2008 35

Rich Media Performs Best on OPA Sites
• Rich media ads on Ad Networks have no impact on Purchase Intent or
Brand Favorability
A B C D
B B B B B
C C C C C
D D D D D
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA N=1,120; MN=1,854; Portals=979; Ad Networks=333 36

Interactive Ads Perform Better on OPA Sites
than Portals
• Interactive advertising on Ad Networks had no Purchase Intent impact
A B C D
B B
B
B C C B C
D D D C D
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
A/B/C/D indicate statistically significant difference between deltas at .90 CL
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA=476; MN=825; Portals=418; Ad Networks=94 37

Large Ads on OPA Sites Have Greater Impact
on Brand Favorability and Purchase Intent
A B C
B B B B
C B C C C
*Ad Networks excluded due to low number of full and half page campaigns
A/B/C indicate statistically significant difference between deltas at .90 CL
Note: Ad effectiveness deltas in red are statistically insignificant (i.e., there is no change)
Source: Dynamic Logic’s MarketNorms campaigns over last 3 years through Q1 2009
OPA=40; MN=99; Portals=21 38

Executive Summary
For ad effectiveness, environment matters –
and Content Sites help advertisers ‘move the needle’
• OPA Member sites’ ad effectiveness consistently outperform other sites –
including Portals and Ad Networks, even during an economic recession
• OPA sites have largely improved their ad effectiveness over time, unlike
Portals and Ad networks
• These Content Sites generally do a better job of integrating advertising
into professional video, as well as rich media and interactive ads
• Ad Networks’ Purchase Intent delta is statistically insignificant – that is,
brands get the same result if they do not advertise on ad networks as they
do by advertising on networks
39

For More Information
Research Questions:
Stuart Schneiderman
Director of Research
Online Publishers Association
stuarts@online-publishers.org
PR Inquiries:
Aimee Yoon
Dotted Line Communications
aimee@dottedlinecomm.com
646.678.4980
40