Digital Publishers: Turn on Custom Audiences, Ramp Up CPMs

Digital Publishing:Adding Value Through Data

New Research Findings

Stefan Tornquist, VP Research, EconsultancyKathy Menis, SVP of Marketing, Signal

Our speakers

Stefan TornquistVP, ResearchEconsultancy

@SKTornquist

Kathy MenisSVP, MarketingSignal

@KMenis

Today

The study and some contextA look at the digital publisher study behind this talk

FindingsA look at top findings from the study

Case Studies and QuestionsData-driven publishing in the real world and your questions

The study

• September 2015• Highly screened (127 in final cut)• 60% have offline publishing operations• Revenues $50M+ ($200M-$500M 31% / $500-$1B

29%)• Director+ (EVP/C-suite/Board 40%)

Headwinds on a rising tide – display in context

/01

The state of publishing

Current trends in display advertising revenue

Shrinking quickly

Shrinking slowly Staying about the same

Growing slowly Growing quickly0%

5%

10%

15%

20%

25%

30%

35%

40%

7%

15%18%

34%

26%

Challenges

None of the above

Slow and disconnected marketing technology

Unable to target in real time with advertisers' or partners' first-party data

Unable to target in real time with 3rd-party data

Unable to use live intent data to target 'in-market' consumers

Inability to recognize consumers across devices

Insufficient scale to recognize individuals online and across channels

0% 10% 20% 30% 40%

18%

24%

29%

29%

30%

32%

33%

Top display advertising challenges – 2 choices allowed

0%

20%

40%

60%

80%

100%

45% 38% 32% 30%

35% 41% 52%37%

13% 17% 11%27%

6% 5% 5% 6%

Strongly agree Agree Disagree Strongly disagree

Running with the alpha dogs

The addressable user /02

Current capabilities

We cannot target based on any of these

We can target by live visitor intent (behavior triggers ads or actions)

We can target with shared advertiser data

We can track visitor identity across devices/platforms

We can target by content category

We can target with applied third-party data

We can target by audience type/persona

We can target registered subscribers

0% 15% 30% 45% 60% 75%

4%

41%

41%

47%

49%

51%

57%

61%

The desktop display challenge

0%10%20%30%40%50%60% 52%

38%31% 28% 26%

The mobile display challenge

0%

15%

30%

45%40%

38%

31% 30%28% 28%

PUBLISHER O&O

ADVERTISER DATA

W E B S I T E & M O B I L E A P P

C U S T O M E R F I L E , W E B S I T E & M O B I L E A P P

MEDIAPLATFORMS

S U P P LY S I D E

D E M A N D S I D E

NEUTRAL ID PLATFORM

S E G M E N TAT I O N U IHASHED IDs

HASHED IDs, ATTRIBUTES & BEHAVIORS

CUSTOMAUDIENCE

SEGMENT(S)

SYNC PUBLISHER IDs

COLLECT ADVERTISER DATA

SET-UP AUDIENCE SEGMENTS

• HANDLES OPT-OUT

• SET-UP ALIGNS WITH PUBLISHER PRIVACY POLICY

Neutral onboarding + activation platform

Defending CPM/03

Capabilities and premiums

We can target with applied third-party data

We can target by content category

We can target with shared advertiser data

We can target by audience type/persona

We can target by live visitor intent (behavior triggers ads or actions)

We can target registered subscribers

We can track visitor identity across devices/platforms

0% 20% 40% 60% 80% 100%

22%

25%

26%

32%

33%

35%

37%

41%

34%

34%

43%

33%

38%

41%

16%

32%

25%

13%

28%

18%

10%

22%

9%

15%

13%

7%

9%

12%

22%

9%

15%

13%

7%

9%

12%

High premium Good premium Some premium Limited premium No premium

Applying audience data

AdvertiserData

Custom Segments

TailoredAds

77% “Quite” or “Highly Valuable”

A competitive advantage : the best of both worlds

Data-driven identityMaximum Relevancy

for Advertisers

Controlled,journalistic content

Looking ahead/04

Moving forward

0%

20%

40%

60%

80%

100%

38% 34% 31% 31% 26%

34% 47% 49% 44% 48%

22% 16% 16% 18% 25%6% 3% 4% 7% 2%

Highest priority Significant priority On the horizon Not a priority

Lessons of high-growth publishers

• High growth publishers (HGPs) aren’t waiting to be where their audience is.

• Targeting by live intent data is seen as fundamental. • HGPs are better able to take advantage of advertiser

data as a targeting variable. • To meet the alpha-publisher challenge, HGPs are

emphasizing partnership and technology. • The ability to target known subscribers at scale is

their highest priority.

People-based marketing outcomes

/05

YAHOO! Japan is not a Walled Garden

“Our data strategy is open, that’s the key word here. Our policy is about being completely open about our data.”

--Toru TakataSVP & Chief Product Officer

For Marketing Solutionsat Yahoo! Japan

GOALSA leading financial services company wanted to improve their digital reach for an upsell advertising campaign.

The goals were to increase conversions and reduce CPA’s

SOLUTIONSignal’s platform enabled the advertiser to combine their data with YJ’s “online” data for cross-channel and cross-device matching. The advertiser was then able to target an audience based on specific purchase or intent behaviors to increase upsell conversion.

YAHOO! Japan case study

6x

Client Conversions

778

129

+94X

Ad Impressions

126,458

11,944,961

+76x

Publisher Revenue

￥ 15,926

￥ 1,210,945

Retargeting Custom Audiences

YJ Custom audience case study results: higher conversions

Retargeting Custom Audiences Retargeting Custom Audiences

Solving scale/06

Achieving people-based marketing at scale

Difficult to do with only one source of data

Data Collection Technology

?Publisher

But what if?

Publisher1

Advertiser2

Publisher2

Advertiser1

Stop diminishing CPMs and grow premium ad inventory

Premium Direct

Remnant/Open Exchange

Content-based

• C A T E G O R Y / C O N T E N T • S E G M E N T S ( D F P )

Programmatic Premium

Premium Direct

Programmatic Premium

Remnant/Open Exchange

• C A T E G O R Y / C O N T E N T • S E G M E N T S ( D F P )

• C A T E G O R Y S E G M E N T S ( D F P )• D M P C A T E G O R Y

I N T E R E S T S E G M E N T S

• C A T E G O R Y / C O N T E N T S E G M E N T S• R O S / R E M N A N T ( O P E N E X H A N G E )

• C A T E G O R Y / C O N T E N T S E G M E N T S• R O S / R E M N A N T ( O P E N E X H A N G E )

• C A T E G O R Y S E G M E N T S ( D F P )

• C U S T O MA U D I E N C E S

• D M P C A T E G O R Y I N T E R E S T S E G M E N T S

People-based

Questions?/07

Questions / Answers

Stefan TornquistVP, ResearchEconsultancy

@SKTornquist

Kathy MenisSVP, MarketingSignal

@KMenis

Thank you!