Everything you need to know about viewability & AppNexus' new viewable marketplace

1. Everything You Need To Know
About Viewability &
The New Viewable Marketplace
1 AppNexus Inc. - Confidential – Not for Distribution

2. Document Guide
A Visual Overview of Viewability
An Explanation on Why Vendors Provide Different
Viewablity Results
Answers to Your Most Common Questions (Including
Which Verticals and What Sizes are the Most Viewable)
Introduction to the Viewable Marketplace & Why You
Should be Using it
Benefits of Viewable Deals for Buyers & Sellers
2 AppNexus Inc. - Confidential – Not for Distribution

3. Ad Viewability 101
A Visual Guide
3 AppNexus Inc. - Confidential – Not for Distribution

4. In theory, ad-viewability is a simple concept
4
AD-VIEWABILITY
WEB PAGE
VIEWABLE BANNER
SCREEN

5. Reasons for
non-viewability
are numerous
5
AD-VIEWABILITY
OUT-OF VIEW HIDDEN BROWSER TAB HIDDEN BROWSER WINDOW
UNLOADED BANNER HTML OVERLAY NO USER

6. Measurement is made with a script
6
AD-VIEWABILITY
{}),_=d.prototype=0>b?{}:new
oc[b],_.cM=c);for(var
e=3;e<arguments.length;++e)arguments[e]
.prototype=_;d.cZ&&(_.cZ=d.cZ,d.cZ=null)}
function r(){}q(1,-
1,{});_.eQ=xc;_.gC=function(){return
this.cZ};_.hC=yc; _.tS=function(){var
a=this.cZ.d+"@",b;b=this.hC();var
c,d,e;c=zc(Ac,-
1,8,1);d=(Bc(),Cc);e=7;if(0<=b)for(;
SCRIPT ACTIONS:
Banner detection
Viewability measurement
Data transfer
1
2
3
This Script runs on the user’s computer and measures every ad banner that is served with the script.

7. iFrames greatly challenge viewability measurement
7
AD-VIEWABILITY
Iframe requires a specific viewability measurement
iFrame: opaque wall around the banner
Script: no access to the page information

8. A robust
measurement
requires huge
effort
8
AD-VIEWABILITY
DOZENS OF
BROWSER
VERSIONS
HUNDREDS OF
BANNER
FORMATS
THOUSANDS OF
DEVICES, SCREEN
SIZES, ETC.
MILLIONS
OF WEBSITES
280K+TESTS
AUTOMATICALLY
PERFORMED

9. Pixels and Durations
9
AD-VIEWABILITY
1 second Y seconds
50%
OF BANNER PIXELS
X%
OF BANNER PIXELS
IAB DEFINITION CUSTOM DEFINITIONS
Viewable impressions take into account two independent variables. The first dimension is pixel area and the second is
duration (how long pixels are in view). Both of these factors are dependent on the user’s behaviors.
The IAB definition states that 50% of banner pixels must in view for at least one second.
Definitions can vary since you can measure ‘X’% percent of banner pixels in view during ‘Y’ seconds.

10. IAB Definitions
10
AD-VIEWABILITY
1 second
50%
OF BANNER PIXELS
STANDARD DISPLAY
BANNER
2 seconds
50%
OF BANNER PIXELS
VIDEO AD
1 second
30%
OF BANNER PIXELS
Sometimes large banners can be very high in pixels (> 242,500 Pixels),
and may not actually fit on the user’s screen,
so this definition is slightly different.

11. Vendor Discrepancies
Why do viewability vendors provide different results?
11 AppNexus Inc. - Confidential – Not for Distribution

12. Comparing vendors can sometimes be puzzling
12
AD-VIEWABILITY
Vendor-A AppNexus Vendor-B
Purchased impressions Unknown 54 million Unknown
Analyzed impressions 43 million 54 million 33 million
Measured impressions 36 million 43 million 28 million
Measurement rate 83% 80% 85%
Measurement rate
(/ total imps)
67% 80% 52%
Viewability rate 56% 47% 50%
Viewable impressions 20 million 20 million 14 million
30 campaigns | 54 million impressions

13. Let’s compare viewable impressions per campaign
13
AD-VIEWABILITY
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
-80% -60% -40% -20% 0% 20% 40% 60% 80%
DIFFERENCE OF VIEWABLE IMPS VS APPNEXUS
Vendor-A / AppNexus
Vendor-B/AppNexus
EXAMPLE:
• Vendor-B counts 54% fewer viewable
impressions than AppNexus
• Vendor-A counts 11% more viewable
impressions than AppNexus
If vendors had the same results,
all campaigns would be here
At first glance, it seems that viewability results are not consistent among vendors

14. And now viewability rates
14
AD-VIEWABILITY
y = 1.0755x - 0.004
R² = 0.89215
0%
20%
40%
60%
80%
100%
0% 20% 40% 60% 80% 100%
AppNexus
Vendor-B
APPNEXUS VS VENDOR-B
y = 1.0075x - 0.0691
R² = 0.8365
0%
20%
40%
60%
80%
100%
0% 20% 40% 60% 80% 100%
AppNexus
Vendor-A
APPNEXUS VS VENDOR-A
Viewability measurement technologies provide consistent results,
So where do the differences come from?

15. Last, measured impressions
15
AD-VIEWABILITY
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
-70% -60% -50% -40% -30% -20% -10% 0% 10% 20% 30% 40%
DIFFERENCE OF MEASURED IMPS VS APPNEXUS
Vendor-A / AppNexus
Vendor-B/AppNexus
Example:
• Vendor-B measured 78% fewer
impressions than AppNexus
• Vendor-A measured 16% fewer
impressions than AppNexus
The main difference is not in measurement methodologies, but in measurement set-up

16. Answers To FAQs
How viewable is programmatic?
Can I find viewable inventory?
What formats are the most viewable?
16 AppNexus Inc. - Confidential – Not for Distribution

17. There is a lot of very viewable inventory
17
The placements that
represent the most viewable
quarter of impressions
(first quartile) have the
same viewability as
premium sites!
AD-VIEWABILITY
50%
78%
View Rate First Quartile View Rate

18. Whatever content category, quality is there
18
AD-VIEWABILITY
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
View Rate First Quartile View Rate

19. Whatever format you need, quality is there, too
19
AD-VIEWABILITY
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
300x250 728x90 120x600 468x60 970x250 336x280 970x90
View Rate First Quartile View Rate

20. An Introduction to the
Viewable Marketplace
Get used to it. This is the future of online advertising!
20

21. Viewable CPM is not controlled when you buy CPM
21
Viewable CPM is the
cost per mille for
viewable impressions.
It depends on both
CPM and viewability!
It is hard to control both.
AD-VIEWABILITY
0
0.5
1
1.5
2
2.5
3
3.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
EVOLUTION OF VIEWABLE CPM PER DAY

22. Viewable deals: only pay for viewable impressions
22
AD-VIEWABILITY
VIEWABLE
DEAL
Campaigns target the
viewable deal
BID
$10 VCPM
Viewable
$10
$0
Final Payout
IMPRESSION 2
VIEWABILITY MEASUREMENT
IMPRESSION 1

23. A win-win system
23
Lower VCPM for buyers and higher CPM for sellers!
AD-VIEWABILITY
VIEWABLE DEAL
(VIEWABLE CPM = $10)
PREDICTED
VIEWABILITY
EFFECTIVE CPM
(VIEWABLE CPM
X PREDICTED
VIEWABILITY)
OTHER
COMPETITORS
CPM BIDS
BID
WINNER
VIEWABLE
IMPRESSION?
PRICE PAID
(X1000) BY BID
WINNER
IMPRESSION 1 75% $7.5 $4
Viewable
deal
Yes $10
IMPRESSION 2 50% $5 $1
Viewable
deal
No
$0 (non
viewable imps
is not paid)
IMPRESSION 3 25% $2.5
$5 - $4 5$ CPM
bid
N/A
$4 (bid
reduction)

24. Benefits of
Viewable Deals
24

25. Benefits of Viewable Deals for Buyers
25
AD-VIEWABILITY
PUBLISHER
RELATIONS
BRANDING
CAMPAIGNS
BUYING
EFFICIENCY
Provides a key
performance indicator
Real exposure of contacts
Real ROI of campaigns
Zero waste
Zero risk
Fewer intermediaries costs
Publishers’ targeting
capabilities
Buy BTF!
Leverage direct relation

26. GAIN
EFFICIENCY
HIGHER
YIELD
PREMIUM
No intermediaries
New demand for
your inventory
Higher CPMs
Access to premium buyers
Differentiate your
inventory
Get branding budgets
Available for
header-bidding
No discrepancies
No need for reconciliation
Benefits of Viewable Deals for Sellers
26
AD-VIEWABILITY

27. Viewability is a market changer.
For more information on how you can partner
with AppNexus and access the world’s only
open, independent viewable marketplace,
contact us.
27