Shun Maekawa

Shun Maekawa
Data Technology Center
General Manager

Sohei Mitani

Sohei Mitani
Data Technology Center
Senior Analyst

Data clean rooms allow long-term data applications and integration of marketing

The use of cookies, Identifier for Advertisers (IDFA), and other personal identifier data have become increasingly restricted in recent years as a result of revisions to related laws amid growing global awareness of the need to protect personal information.

With the coming arrival of a post-cookie era, the arena for digital marketing is shifting from the cookies-based OpenWeb platform to anonymized customer data supplied by major platform providers, such as operators of loyalty points programs.

Data clean rooms have emerged along with this shift. A data clean room is a marketing environment provided to corporate clients by platforms such as Google, Facebook, and Amazon. Within a secure cloud location, first-party data held by corporate clients and data stored on the platform can be shared, analyzed, and used for distributing advertisements and other information. As such, data clean rooms can be seen as a means for securely utilizing data while ensuring the privacy of consumers.

Further, a vast amount of ID data already approved by users is stored on the major platforms, meaning this data can be used for many years to come. In contrast, the use of the same cookie IDs tracked through cookies has, in practice, been limited to individual marketing campaigns.

In addition, data clean rooms allow data at the individual ID level to be analyzed all at once, including online data such as keyword searches, offline data from loyalty point programs and point-of-sale purchases, as well as cross-device data from smart phones, personal computers, and other devices. Various types of data can also be combined for analysis, including first-party data held by companies, user-approved ID data stored on platforms, and television viewing data.

Benefits of using data clean rooms

Dentsu has focused on data clean rooms since early on, and built systems for using and running more than 500 data clean rooms. Having taken the initiative ahead of its rivals, it offers a multitude of enhanced functions through these systems, which have earned an excellent reputation among many platform operators.

Four cases in which advertising was optimized through the use of a data clean room are presented below.

Case1: PDCA cycle design based on consumer interests, tastes, and attributes

To help a corporate client identify customers with potential needs, we used Google Analytics to conduct a preliminary analysis of highly probable attributes of users using a large amount of anonymized customer data related to consumer tastes and interests.

The targeted attributes specified through this process were then measured using the anonymized customer data in Ads Data Hub, Google’s data clean room environment. We then clustered the client’s customers according to interests and tastes in computer games, travel, and automobiles. Advertising plans tailored to each cluster were carried out, and a plan-do-check-act (PDCA) cycle was implemented.

Case 2: Single-source panel analysis conducted to create an omnichannel at low cost

Since in order to integrate online and offline data into an omnichannel, a company that owns stores must make huge investments to develop, and cover the operational costs of, such a channel, we used user-approved IDs in a customer database as a large, single source of data. While limiting costs, doing so allowed us to collect data on purchases after website visits and, at the same time, created a model for predicting the likelihood of purchases.

Case 3: Short-term PDCA cycle enabled by converting strategic cluster data

The client was a producer of food products that was experiencing sluggish sales due to the COVID-19 pandemic. It did not have quantitative data for understanding what types of customers had stopped buying products and the probability of customers resuming purchases. Although panel data for the market as a whole was available, it had not been integrated with the client’s customer data.

We thus conducted a cluster analysis to determine the structure of the customer base from the panel data, and to strategically target certain customers. We then converted the cluster data for a data clean room provided by Loyalty Marketing, Inc. (operator of the Ponta loyalty program). This allowed us to distribute ads to individual IDs across various types of media, and carry out a PDCA cycle over the short term.

Case 4: Shopping experiences enhanced, marketing optimized with registration-free campaign

Ideally, in order to lower barriers for becoming involved in digital marketing, proof of purchase data from consumers’ ordinary shopping activities should be obtained, while loyalty point awards and cash rebates should be offered. If these shopping activities are too ordinary, however, raising consumers’ awareness of brand offers and encouraging them to become fans of a brand can be difficult.

In this case, to increase the scale of the brand and foster customer loyalty, we implemented a registration-free campaign that offered consumers a point rebate just by making a purchase using the PayPay electronic settlement service. The impact of the campaign was measured using a data clean room, which allowed us to analyze how this digital marketing campaign performed and the know-how gained from its implementation.

Undoubtedly, the use of data clean rooms in ways described above has the potential to bring about major advances in corporate marketing. However, solutions are still needed for some issues, such as the many limitations of machine learning and other advanced analytical tools. Data clean rooms by themselves cannot provide all the answers.

How companies respond to the inevitable arrival of the post-cookie era remains to be seen. It is important for them to recognize this major shift as an opportunity rather than a calamity. By starting to use data clean rooms as soon as possible, companies can capitalize on the advantages ahead of their competitors.

Dentsu will work to turn new data generated by data clean room environments into assets, and help realize a genuine digital transformation of the marketing industry.

Authors

Shun Maekawa

Shun Maekawa

Data Technology Center
General manager

After planning and measuring results of television and digital integration initiatives as a data analyst, in 2015 Shun Maekawa was appointed to lead a project to develop Dentsu’s STADIA marketing platform to integrate TV commercials and digital ads. Currently, he is mainly involved in supporting the development of data clean rooms, provided by platform operators, and their adoption by the advertising and marketing industries. In the post-cookie era, clean rooms will provide the basis for the generation of new data.

Sohei Mitani

Sohei Mitani

Data Technology Center
Senior Analyst

Sohei Mitani has managed direct advertising sponsors and assisted with the digitalization of marketing processes. He is currently developing tools and promoting advances in digital marketing based on quantitative performance results.