For Marketing Leadership Professionals

August 23, 2013

Solving The Cross-Platform Targeting Riddle

Why Marketers Should Not Be Satisfied With Todays Mechanisms
For Targeting Individuals Across Platforms And What They Can
Do About It
by Joanna OConnell
with Luca S. Paderni, Samantha Merlivat, and Collin Colburn

Why Read This Report

As consumers increasingly embrace personal mobile technologies, the third-party cookie, the most
prevalent desktop-based targeting mechanism, is showing its limits at identifying and addressing your
target audience let alone at reaching the same user cross-platform. Todays targeting challenge for
marketers is threefold: 1) reliably identifying, in a persistent manner, an individual consumer; 2) reaching
that consumer across an ever-increasing number of devices and platforms; and 3) partnering with vendors
whose infrastructure is still built around traditional targeting mechanisms. Read this report to make
sense of todays cross-platform targeting ecosystem, understand the pros and cons of each targeting type,
develop your own plan for cross-platform user targeting using this knowledge, and determine where you
want the ecosystem to improve and fast to meet your specific targeting needs.
evolving consumer behaviors reveal flaws in TRADITIONAL TARGETING
In the past few years, increasing adoption of tablets, smartphones, and other connected devices has driven
the growth of perpetually connected consumers (PCCs) in every part of the globe.1 Forrester defines the
perpetually connected consumer as one who owns and personally uses at least three connected devices
and accesses the Internet multiple times a day from multiple physical locations, at least one of which is
on the go. Through their everyday actions, these consumers are changing the terms of their relationships
with marketers, demanding greater relevance in every interaction.

Perpetually connected consumers are mainstream. In the US, by the end of 2012, 42% of online

adults met the Forrester definition of PCCs, up from 38% in late 2011.2 Globally, Forrester predicts
that by the end of 2013, close to half of online adults will be perpetually connected. This opens up the
opportunity for marketers to address consumers at any time and in different contexts and, at the same
time, poses a challenge, as interactions at each touchpoint need to be more data-driven and targeted to
be relevant.

PCCs value utility, personalization, and relevance. These types of consumers want to be addressed

by content designed around their needs and wants.3 Forresters Mobile Mind Shift Index estimates that
22% of consumers now demand mobile utility: They expect any desired information or service to be
available on any appropriate device, in context, at their moment of need.4 These expectations call for
individual rather than segmented marketing. Marketers need to develop the ability to recognize and
target individual consumers, on whichever device they use, at any point in time.5

Headquarters
Forrester Research, Inc., 60 Acorn Park Drive, Cambridge, MA, 02140 USA
Tel: +1 617.613.6000 Fax: +1 617.613.5000 www.forrester.com

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

Cookie-Based Targeting Falls Short In The Multiplatform Era

The cookie, a tiny piece of code associated with a computers browser and originally designed for
very basic desktop functions, has become the core underpinning of much of the audience targeting
done today. Our 15-year relationship with the humble third-party cookie is finally showing signs of
strain, however, in an increasingly mobile and privacy-aware world, throwing the door wide
open for alternative solutions. But in the current state of limited market adoption and with new
technologies still in early stages of development, there remains no clear winner in the multiplatform
targeting game. The three primary options available for user targeting, whether within one channel
or across several, are cookie-based targeting; person-based targeting; and inference-based targeting,
and each has strengths and weaknesses (see Figure 1 and see Figure 2):

Cookies are the most scaled targeting option but the least accurate across platforms.

Cookies, typically belonging to third parties such as DSPs, ad networks, or targeting vendors,
are a powerful targeting tool for desktop-based audience targeting they are nimble, easily
deployable, scalable, and widely available. Third-party cookies fall significantly short, however,
when asked to bridge the gap across a users desktop and her mobile device(s), facing serious
limitations that hamper their utility for multiplatform marketers. Heres why: 1) Mobile OS
platforms are more plentiful than their desktop counterparts and often block third-party
cookies by default, as Apple does, and 2) just as in the desktop world, cookies are specific to a
browser they do not travel across devices, making it very difficult to link a mobile user to her
desktop persona. And while first-party cookies address the cookie-blocking issue, they are not
immune to the cross-browser problem.

Person-based targeting brings the highest accuracy at the expense of reach. A whole host

of companies have collected personally identifiable information (PII) about users for a variety
of reasons: Yahoo creates and maintains users email addresses; Apple has customer phone
numbers; and marketers have databases filled with customer (and sometimes prospect) data.6
These persistent, user-specific data points can serve as a key to identify, map, and ultimately
target a given user across channels, formats, and devices. Person-based targeting has a highly
promising future but struggles today with: 1) access challenges marketers must either own
the PII key directly or find partners that are willing to share it; 2) scale issues there are
simply fewer person-based profiles than cookies today; and 3) privacy concerns there is a risk
of privacy infringement if this is not handled with extreme care.

Inference-based targeting shows promise but remains limited in utility and adoption.

Several companies such as BlueCava and Drawbridge have staked their fortunes on
statistical inference to make accurate guesses about both an individual device and that devices
relationship with other devices, using a range of data points about the devices themselves, the
users browser(s), and more. Other vendors, including several in the data management platform
(DMP) space, are aggressively exploring this option, having run into the limitations of the
third-party cookie firsthand.7 This method also shows promise but faces very real issues today,

2013, Forrester Research, Inc. Reproduction Prohibited

August 23, 2013

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

including: 1) very limited scalability relative to other options; 2) criticisms about its accuracy;
and 3) an air of privacy-unfriendliness particularly in marketers confusion about persistent
device-based versus more ephemeral inference-based targeting that scares many marketers.
Figure 1 Marketers Can Choose Among Three Primary Targeting Options
What it is:
Cookie-based

A small piece of data stored in a users browser by a visited website to

help the site recognize that browser in the future. There are two types:
Third-party cookies associated to third-party domains, e.g., a DSP
First-party cookies associated to the host domain, e.g., the brand
website

Person-based

Relies on personally identifiable information (PII). PII is information that

can be used on its own or with other information to identify, contact,
or locate a single person or to identify an individual in context, such
as physical address, phone number, or email address.

Inference-based

Relies on statistical inference based on one or more parameters to make

a strong guess about a unique user. These can include device, cookie,
and browser characteristics.

Source: Forrester Research, Inc.

92561

Figure 2 Each Targeting Method Presents A Different Set Of Strengths And Weaknesses
Third-party
cookie-based

Personbased

Inferencebased

Performance
Accuracy
Scalability
Persistency

Strong
Moderate
Weak

Ease of deployment
Privacy
Platform
Display
Mobile
Cross-channel
92561

2013, Forrester Research, Inc. Reproduction Prohibited

Source: Forrester Research, Inc.

August 23, 2013

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

marketers need to shift from audience to relationship management

Current targeting challenges, however, mustnt be taken as an excuse not to pursue more targeted
communication. Its time for marketers of all stripes, not just the customer relationship management
(CRM) team, to embrace the idea of relationship management continuously endeavoring to
positively engage with users across the customer life cycle from acquisition through retention and
use that to drive their broader approach to targeting. Mastering the consumer data flow (gathering
it, making sense of it, and applying it) is the key to this novel approach. In a targeting environment
still lacking the perfect solution, successful marketers should enrich their arsenals with new tools
and techniques, with the goal of developing new ways to begin and maintain conversations based on
high relevancy and a compelling value exchange.
Build Your Way Into Data-Driven Relationship Management
Marketers who are serious about starting the journey toward more granular multi-touchpoint
targeting and, ultimately, relationship management, must look for every opportunity to gather and
store user-level data for future targeting purposes. This can come from a variety of first-, second-,
and third-party sources: your own website, your offline database, direct publisher relationships, and
external vendor data that you license. Marketers leading the charge here are:

Deploying a DMP to unify their data-gathering efforts. The past twelve months brought an

explosion of interest in DMPs from marketers explicitly looking for a more unified mode of
data collection and targeting. With the help of DMPs such as BlueKai, companies such as HP
are now capturing a wide range of previously unknown data points on site visitors to use in
their site and ad-based targeting efforts, with a long-term goal of making every communication
relevant. Unsurprisingly, major advertisers are pushing their DMPs to solve for cross-platform
challenges, leading DMP vendors to invest innovation and release new features to address this.

Taking advantage of publishers direct control of first-party consumer relationships.

Premium publishers know that their ability to connect with consumers via their content across
platforms is a highly valuable asset. As Lori LeBas, the SVP of business operations, sales, and
marketing of ESPN, notes, Measures are available to publishers like ESPN to target a verified
audience (across devices), but few marketers today are ready to pay the price premiums
associated with these techniques. As DMP adoption is increasing, so too are new kinds
of business relationships between marketers and publishers, including mutually beneficial
data-sharing arrangements that give both parties direct, controlled access to each others
users. Acxiom, a leading marketing service provider (MSP), is investing in the technology and
publisher partnerships necessary to stitch together a massive pool of targetable consumers
across platforms, using its Abilitec product, its universal match key for consumers.

Working with third-party data providers for a mix of ephemeral and persistent signals.
Increasingly, marketers are using third-party data to enrich existing user profiles for more-

2013, Forrester Research, Inc. Reproduction Prohibited

August 23, 2013

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

accurate targeting and more-relevant message delivery. Today, a whole range of data providers
are at the ready to round out a marketers targeting strategy, from traditional players such as
IXI and Polk (in the financial and automotive sectors, respectively), to digital natives such as
DataLogix, which offers targeting based on users previous shopping behaviors. To keep track
of it all at the user level, DMPs such as Adobe Audience Manager are constantly synching with
third-party providers to maintain a completely up-to-date read of users characteristics for
insights and targeting.

r e c o m m E N D AT I O N S

Push the envelope with whats available and focus on innovation

Ultimately, in spite of the myriad challenges that marketers face with cross-platform user targeting,
the real cost will come from not upgrading their targeting strategies. Marketers who continue
to use the same old tried-and-true methods of targeting risk driving consumers into the arms of
savvier competitors. Not accepting the industry status quo is, in this case, a business imperative
marketers must demand better solutions and push for more innovation from partners while making
the most of whats currently available. In the meantime, as youre building out your multiplatform
user-targeting strategy, dont forget some basic, but hugely important, principles:

Develop your measurement strategy ahead of execution. Thankfully, while the targeting

piece still remains a major challenge, multichannel/platform measurement systems have

been steadily improving year over year. Advanced attribution companies such as Adometry
and Visual IQ use both positive and negative signals from users to understand channel
interplay. Before you launch a campaign targeting users across platforms, be sure the
relevant tracking is in place you may not get to a perfectly clean answer, but youre well
served to be prepared in advance to glean meaningful insights post-initiative (or even midinitiative).

Ensure that your privacy policy is in tune with your brand values. Marketers have long

ignored the explicit role a consumer plays in the targeting value exchange for fear of rocking
the boat, both internally (e.g., with legal teams) and externally (e.g., with a front-page
story in the Wall Street Journal). In the age of perpetually connected consumers who are
more aware than ever that their data is out there for the taking, this is no longer a viable
option. Marketers must focus on trust, transparency, and control in their dealings with
consumers and their data. As Ned Brody, CEO at AOL Networks, puts it, The reality is that
many things have been done for the consumers benefit but not with the consumers choice.
Beyond your own privacy and consumer data-related efforts, be sure you understand any
potential partners approach to data collection and consumer opt-out, particularly with
respect to more cutting-edge techniques such as device fingerprinting.

2013, Forrester Research, Inc. Reproduction Prohibited

August 23, 2013

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

Invest in creative strategy as much as in targeting. Marketers who want to deliver on the

promise of relevant message delivery will need to drive deeper integration of creative and
media-buying decisions to be able to act quickly enough. This may mean developing formatspecific executions, building more creative concepts to appeal to a range of very different
audience segments, or investing in dynamic creative optimization (DCO) technology
ideally, it means doing all of these things. The payoff is there, says Yahoos Peter Foster, vice
president of solutions development and mid-market sales, In early results, were seeing that
the combination of audience solutions, in addition to data-driven creative that personalizes
the ad elements to each individual, can yield up to 100% lift in conversion.

Supplemental MATERIAL
Companies Interviewed For This Report
Acxiom

Responsys

AOL

Rocket Fuel

AudienceScience

Simulmedia

BlueKai

State Farm Mutual Automobile Insurance

ComScore

The Trade Desk

Digitas

The Weather Channel Companies

ESPN

Turn

Guthy-Renker

X Plus One

Knotice

Xaxis (WPP)

Nielsen

Yahoo

Endnotes
Forrester defines the perpetually connected consumer as one who owns and personally uses at least three
connected devices and accesses the Internet multiple times a day from multiple physical locations, at least
one of which is on the go. For a full discussion of these customers and the addressability framework we
use to engage them, see the September 26, 2012, The Always Addressable Customer report.

At the end of 2012, 42% of US online adults were accessing the Internet multiple times a day from multiple
devices and locations. See the February 11, 2013, 2013 Interactive Marketing Predictions report.
Source: North American Technographics Online Benchmark Survey (Part 2), Q3 2012 (US, Canada).

2013, Forrester Research, Inc. Reproduction Prohibited

August 23, 2013

For Marketing Leadership Professionals

Solving The Cross-Platform Targeting Riddle

In 2013, the ultra-connected customer base will grow at a staggering pace, destabilizing marketing as youve
come to know it. This report identifies the changes youll see in your customers expectations, your abilities,
and the way your organization is structured and explains how you can succeed in this new age. See the
February 11, 2013, 2013 Interactive Marketing Predictions report.

Six years into the smartphone transition, customers are making a mobile mind shift. The shifted customer
expects that any desired information or service is available on any appropriate device, in context, at their
moment of need. To analyze how far people have shifted, we created the Mobile Mind Shift Index (MMSI),
which segments people into six categories: Disconnecteds, Dabblers, Roamers, Adapters, Immersers, and
Perpetuals. See the April 19, 2013, The Mobile Mind Shift Index report.

Forty-three percent of perpetually connected consumers are willing to share personal data in exchange
for loyalty points, compared with only 28% of US online adults overall. See the April 19, 2013, Marketing
Strategy For The Mobile Mind Shift report.

PII is defined in the NIST Special Publication 800-122 as any information about an individual maintained
by an agency, including (1) any information that can be used to distinguish or trace an individuals identity,
such as name, social security number, date and place of birth, mothers maiden name, or biometric records;
and (2) any other information that is linked or linkable to an individual, such as medical, educational,
financial, and employment information. A users IP address counts as PII, regardless of whether it may
or may not on its own be used to uniquely identify a person. Source: Erika McCallister, Tim Grance, and
Karen Scarfone, Guide to Protecting the Confidentiality of Personally Identifiable Information (PII):
Recommendations of the National Institute of Standards and Technology, NIST, April 2010 (http://csrc.nist.
gov/publications/nistpubs/800-122/sp800-122.pdf).

A data management platform (DMP) is defined by Forrester as, A unified technology platform that intakes
disparate first-, second-, and third-party data sets, provides normalization and segmentation on that data,
and allows a user to push the resulting segmentation into live interactive channel environments. DMPs
have risen in popularity over the past year with marketers looking to unify their digital audience data
collection efforts and operationalize targeting across a range of addressable channels, both paid and owned.
For more information on the DMP space, see the July 25, 2011, The DMP Is The Audience Intelligence
Engine For Interactive Marketers report.

Forrester Research, Inc. (Nasdaq: FORR) is an independent research company that provides pragmatic and forward-thinking advice to global leaders in business and technology.
Forrester works with professionals in 13 key roles at major companies providing proprietary research, customer insight, consulting, events, and peer-to-peer executive programs.
For more than 29 years, Forrester has been making IT, marketing, and technology industry leaders successful every day. For more information, visit www.forrester.com.
2013 Forrester Research, Inc. All rights reserved. Forrester, Forrester Wave, RoleView, Technographics, TechRankings, and Total Economic Impact are trademarks of Forrester
Research, Inc. All other trademarks are the property of their respective owners. Reproduction or sharing of this content in any form without prior written permission is strictly
prohibited. To purchase reprints of this document, please email clientsupport@forrester.com. For additional reproduction and usage information, see Forresters Citation Policy
92561
located at www.forrester.com. Information is based on best available resources. Opinions reflect judgment at the time and are subject to change.