How are content marketing and data related?

What exactly constitutes a "good plan" in content marketing? In advertising, the goal was straightforward: "to generate buzz." However, content marketing aims for two things: "to generate buzz" and, crucially, to prompt diverse target audiences to take various actions that connect to the company's business.
When working in advertising, ideas often take center stage while analysis is seen as secondary. It's easy to fall into that habit when looking at content marketing too. But when you want specific actions from your target audience, data and analysis become inseparable from the content itself. Moreover, when truly aiming for the optimal plan, relying solely on website access data can sometimes be insufficient. How can we solve such challenges when they arise? I spoke with Katsumasa Yagi from the Marketing Solutions Bureau and Satoshi Saeki from the Integrated Data Solutions Center, who are currently working through these challenges together in their day-to-day operations.

 

Gunji: First, shall we start by discussing which areas each of us primarily handles?

Yagi: For the entire content marketing team, we focus on products with longer consideration periods. We tailor communications based on users' consideration stage and level of interest, creating and running systems that deliver satisfaction tailored to each user's situation.
Within that, I handle the area of segmenting users based on their consideration stage and level of interest, then building and operating a marketing system that tailors initiatives to these segments, accumulates data, and continuously improves initiative effectiveness.
Mr. Gunji handles communication planning and content development tailored to the defined user segments. Mr. Saeki analyzes the data accumulated from the implemented communications and creates the foundation for insights to inform the next initiatives. Broadly speaking, this is how the roles are divided.

 

Gunji: When you seriously pursue content marketing, you need both communication planning and a database. As someone on the content strategy and planning side, I'm acutely aware of this now. That essentially means we need a "marketing system," right...?

Yagi: I agree. What companies expect from content marketing is figuring out what kind of content will keep delivering results, based on factors like individual interest in products and their attitudes at any given time. From the perspective of my team, which builds the overall framework, I believe one solution is to organically integrate with teams like Mr. Gunji's, which has expertise in how to communicate with end-users, and Mr. Saeki's, which has expertise in analyzing large, unwieldy datasets. This allows us to simultaneously run the entire process from planning and implementation of initiatives to the analysis of that data.

Gunji: On the other hand, we're also learning that when you publish content, you usually get some kind of reaction. That reaction can provide new insights. Site behavior is an obvious example, but there's also qualitative data like user comments, or email addresses collected in exchange for contest entries. From the perspective of understanding the user's situation, I've come to realize recently that content and data are actually inseparable.
If that's the case, then content creators could likely intentionally plan content to gather specific information they want. It's not just about creating good content and seeing if it's well-received; there's also this way of leveraging content. It's about using content as a means to an end, pushing it to its ultimate potential.
But isn't that exactly what we're doing right now?

 

Saeki: Right now, when we expose about 10 pieces of content, we start analyzing the readers for each piece, right?

Gunji: How has it been going so far?

Saeki: We can perform analysis comparable to ad effectiveness or site access analytics, but since the exposure volume per piece of content is lower, we need to be creative. Broadly speaking, content effectiveness boils down to: ① Who saw it and how? and ② What happened after they saw it? For ①, we can measure overall volume and the content's inherent ability to engage readers through metrics like read volume and completion rate. However, for ②, it becomes quite challenging when content exposure is low. Since methods like surveying exposed individuals in a panel to ask about attitude changes—as done in ad effectiveness analysis—are difficult to apply, we're cross-referencing with the large data held by DMPs. We're attempting to analyze, from the vast DMP data, who viewed what and how within each DMP segment, whether they actually finished reading, and changes in site access or behavior after content exposure.
However, since changes in mindset remain difficult to observe, we're considering that a combined approach is necessary: integrating behavioral data analysis with traditional survey methods like creative testing. This would create a comprehensive content evaluation that combines actionable insights with direct questioning.

Yagi: When trying to understand in detail how effective each piece of content is, you naturally want to track an individual continuously to identify content that suddenly boosts effectiveness and the optimal timing for its delivery. How do you approach that?

Saeki: Personally, I think that's difficult. Why? Because even if you track one person, they encounter so many different pieces of content or ads. We remember seeing a billboard or transit ad, but we can't capture all that data. So even if you could track one person, I think there are still limits.
Moreover, it's difficult to fully capture things like the emotional impact of an ad, serendipity (chance discoveries), or shifts in feelings.

The Gap Between Data and Creativity

 

Gunji: In Saeki's circle, since advertising tends to be the focus, would you say they value high-CTR creative?

Saeki: In the realm of performance advertising or direct advertising, we often focus on hard numbers like CTR or CPA. But as the world of content expands, I believe some things can't be evaluated solely by direct effect metrics. This includes indirect effects, or post-impression effects. We're trying to understand this through behavioral data, but I feel advertising tends to get undervalued. It's quite challenging. It depends on the product or category, but people don't always act immediately. Especially online.
Obviously, I firmly believe creative and ideas absolutely impact consumer behavior. Since much of this impact isn't quantifiable, I think analysts need to interpret it with a certain "good sense" – treating the unmeasured parts as latent variables, so to speak. That approach seems to yield more positive results.

 

Yagi: While CTR and CPA are important, if you just push hard based on numbers alone, you'll end up in a shrinking equilibrium unless the market is growing. To avoid that, you need elements like exceeding visible needs or reaching new people with information. These might seem a bit vague in the data, but without that creative aspect, growing while looking at the data is harder than you might think.

Saeki: Exactly. Quantifying things like emotional impact through data will become explainable soon enough. When that happens, our creators will surely be delighted (laughs).
The ideal creative evaluation I'm aiming for involves not just emotional impact, but something that resonates with business objectives like sales figures. It should have enough impact to influence a company's marketing investment decisions. We'd mobilize the entire team to achieve that. That's the vision I want to pursue.

Gunji: What I've been thinking about all along is exactly what Mr. Saeki just said. The power of ideas inherent in creativity, and the ability to bring them to life—in today's society where people are driven by information, there must be a strategic way to leverage that.
Until now, it was difficult to plan in ways that moved people toward the direction management and business aim for, but with data, it might gradually become possible.

Saeki: In the future, as we gain the ability to track biometric information and quantify things like emotional impact, we'll see the emergence of approaches like data-driven creativity. This means digital marketing will encompass not just digital advertising, but also the roles traditionally fulfilled by conventional advertising. For us, within this world of data × creativity, we want to discover the true value and evaluation criteria of content.

 

---

【Gunji's eye】

Content marketing is often discussed in easily understandable terms like operating owned media or owned communities. This time, however, we explored how content and data interact within the broader perspective of marketing as a whole. Our exploration continues in the next installment.