Data-driven management is increasingly demanded, and more companies are acquiring data. However, merely acquiring data is never sufficient. It is crucial to analyze and utilize data to shape a business vision and realize it. This article features an interview with Ryota Konishi from the Data Strategy Division, Business Transformation (BX) Department at Dentsu Digital Inc., who promotes business consulting through data utilization.
In Part 1, Mr. Konishi discussed data-driven strategy formulation and examples of AI/machine learning utilization by data strategists. In Part 2, we delve into the specifics and use cases of the "AI Workshop" his team provides to numerous companies.
Company-Specific "AI Workshops" for Consulting on Data Utilization Challenges
Dentsu Digital Inc. Ryo Konishi
Q. Could you explain again in detail what exactly happens in the "AI Workshop"?
Konishi: I believe many places offer study sessions or lectures on AI. One key feature of our "AI Workshop" is that it's not an open webinar; it's conducted individually for each company. The session lasts 90 minutes. We begin by explaining what AI and machine learning can do, along with their strengths and weaknesses. We then introduce AI use cases and examples of its application in business. We also address the reality that while many companies have attempted AI projects, few have succeeded. We explain why these projects often fail and discuss strategies to mitigate failure risks.
The final 30 minutes are dedicated to a free-talk session. In this segment, after gaining input on AI and machine learning, clients discuss their vision and business challenges. The goal is to help them personalize the seminar content and connect it to solving their specific problems.
Q. So, even though it's called a "workshop," it also includes consulting elements. Is holding it individually also to delve deeper into each company's specific concerns?
Konishi: That's correct. While that's certainly part of it, open webinars accessible to anyone have the advantage of being easy to join, but they can also lead to participants having varying levels of engagement. By holding sessions individually, we believe participants can join with a more specific purpose in mind, making the input more practical. Additionally, by discussing each company's specific challenges and areas of interest beforehand and customizing the case studies we present, we can introduce examples and strengths that are particularly relevant and tailored to each participating company.
Furthermore, when holding our "AI Workshops," we ask that participants come from as wide a range of departments within the same company as possible. This is because AI implementation and strategy development very often fail if they're driven by just one department alone. Effectively integrating and leveraging AI/machine learning in business requires support from diverse departments. Collaboration is essential not only with data-related departments but also with strategy development, implementation, and funding acquisition teams. Precisely because internal awareness and coordination are crucial, we believe it's vital to have participants with varied backgrounds. Ideally, multiple attendees from each department should participate, including those with some decision-making authority.
Addressing a wide range of challenges, from advanced CDP utilization to global e-commerce strategies
Q. What types of companies have participated in the "AI Workshop," and what results did they achieve?
Konishi: Let me share three representative examples.
Case Study: Financial Services Company A financial services company consulted us about advanced CDP utilization. They expressed concerns: "We capture existing customers, but we don't know how to approach quasi-existing or potential customers," and "We want to reach people who show interest in our products or services but haven't visited our website yet."
However, they lacked the data to directly fulfill these requests. We discussed an alternative approach: collecting survey data to build a predictive model for potential customers. We then proposed a model that uses AI to calculate the probability of purchasing specific financial products and automates email outreach.
Case Study: Consumer Electronics Industry In our collaboration with a home appliance company, we established a global e-commerce analytics environment. Initially, they faced challenges: "Each global subsidiary operates its own e-commerce site, leading to scattered data" and "We want to implement a CDP for centralized management and advanced data utilization." Through repeated discussions, a consensus emerged that defining their future vision and strategic roadmap should come first. We then assisted in designing a data-driven concept for their global e-commerce strategy.
Case Study: Apparel Industry Company A company in the apparel industry consulted us after implementing a CDP but struggling to identify effective utilization methods. We held a workshop to list potential challenges solvable with AI/machine learning and prioritize them. From there, we proposed future CDP applications: a system to determine engagement levels based on customer activity, visualization of customer portfolios (distribution of existing customers across various metrics), and integration of this data with their existing email system.
AI and Machine Learning with High Affinity for the CRM Domain
Q. Listening to you, it's clear you prioritize supporting each client closely in leveraging their data. When working alongside clients over the long term, do you encounter any particular challenges?
Konishi: For us, the most crucial factor is whether we can dispel the initial uncertainty clients feel: "Can we really entrust this to them?" Even when presented with a data utilization plan, companies often struggle to visualize how it will actually strengthen or improve their business. Compared to something visually obvious like advertising, it often takes time to grasp. Therefore, a major challenge is establishing ourselves as a team worthy of their trust right from the first contact. Furthermore, when we talk about data utilization, we don't just perform analysis. It's crucial for clients to understand that projects often fail if the vision isn't established first. We put a lot of effort into explaining this point carefully and thoroughly.
Q. Many companies are now promoting data utilization. That said, some may still feel hesitant about adopting AI and machine learning. For which industries or organizations facing specific challenges would you particularly recommend introducing AI and machine learning?
Konishi: Generally speaking, the department or area where AI and machine learning tend to yield the most noticeable results is CRM (Customer Relationship Management). This is an area where customer data is readily available, and where the impact of customer nurturing is most tangible. Of course, AI and machine learning have various applications, and AI is ultimately a tool, not a solution itself. However, predicting the future is where AI truly shines. CRM excels at enabling predictions about which customers to approach and what actions to take based on customer data. We particularly encourage companies facing CRM challenges to consider this.
Dentsu Digital Inc. AI and machine learning solutions go beyond analyzing existing data. They delve into business challenges and re-examine vision and strategy. By incorporating the data strategist's bird's-eye perspective, significant business transformation may be achieved.
The information published at this time is as follows.
I began my career as a CRM/budget planning specialist at a comprehensive e-commerce company, gaining experience in designing member programs for points sites and developing/implementing LTV maximization strategies through user profiling on auction sites. After supporting corporate data utilization initiatives, I assumed my current role. My strength lies in experience at both operating companies and service providers, and my motto is to drive data utilization tailored to each client's specific situation.
