Data has become central to how we run our businesses today. How can we get useful insights from data instead of taking it at face value? Why is it important to build data literacy? We invited Irene Chen, AI Team Lead at iKala to talk about the innovative application of data with Co-founder of Re-lab, Yu-Hsuan Liu.
Here are some highlights from their conversations:
What are the differences between Data Visualization and Information Design?
Yu-Hsuan: Data itself has a wide definition. It may include words, numbers, or professional field knowledge. Generally speaking, Information Design is about supplementing the transmission of information with design, to make it effective and efficient. On the other hand, the definition of Data Visualization is relatively clear. Since we can't deal with so much data in a short period of time, we need to visualize the data. It is like an interface, so that we don't have to memorize a lot of information, but can focus more on the relationship between data.
When we are talking about Data Visualization, the most important part is "data". We should be as objective as possible. If we are preconceptional, or keep thinking about the conclusion, we might ignore the most valuable part of the data, or the hidden story in it. On the other hand, in Information Design, you have to think from a human perspective. You have to put the problems to overcome and the value to create in mind, with a specific purpose.
Share the innovative application of data in both companies.
Irene: There are a lot of data-driven products at iKala. One interesting example, you might have heard about before, is KOL Radar. It analyzes massive amounts of real-time data from mainstream social platforms like Facebook, YouTube, Instagram and TikTok. In the past, when marketers were looking for KOLs, they usually decided with their experience and intuition, and there was no way to deal with so many cases at a time. With KOL Radar, we help them to build up precise and outstanding performance influencer marketing with data.
Yu-Hsuan: People come to us with very different challenges of communication. Sometimes it may be a brand new product that does not exist in the market yet, so they may need to start from scratch to think where their customers are. Based on some of their previous information, or the work they've done with other teams, we then start with making hypotheses. Then design the materials based on these assumptions. How can we communicate with this group of people? What stories, copies and images are more likely to attract them? Through at least two stages of analysis, we can figure out what kind of person will be attracted? What are the reasons they are attracted? If possible, we'll also tap into some of the reasons that might be psychologically relevant, and use these insights to build up further communication strategies.
Why is Data Literacy important?
Irene: I find the topic of Data Literacy really interesting. I've been facing it in my work. I firmly believe that not only we working with data should have data literacy, but those who raise the questions and who we need to communicate with in the final stages, should possess data literacy.
Here are some important steps of data science projects. I actually have to spend a lot of time doing user interviews and defining business problems. Without data literacy, you may keep talking about something big and unrealistic, and this might lead to more problems.
Nowadays, everyone is talking about digital transformation, but if enterprises really want a successful digital transformation, everyone should have these mindsets. You have to frequently observe indicators and trend reports in the industry, then try to understand it, criticize it, and dig into the reasons.
Yu-Hsuan: Speaking of non-professional people, I think two things are very important. First, stay curious. If you have your own preconceived ideas, and are not willing to explore the origin of the problem, or the reasons behind it, data then loses its value. Second, critical thinking, constantly questioning, and exploring the limit of data. Whether the customer is working with us or with iKala, I think if he only wants us to provide an answer, and does not want to embark on this journey together, it would be a great pity.
Thus, when we are working with new clients, specially for cross-departmental projects, we will arrange a start-up meeting to let everyone understand the reasons, goals, and what everyone should be responsible for. We actually found that optimizing the start-up meeting will greatly improve customer's feedback to the project.
What are the key factors in digital transformation?
Irene: At iKala, we help enterprises implement a DAA flywheel framework, combining "Digitalization", "Analytics" and "Application".
In the first step, Digitalization, enterprises need to know which data should be sorted, and what information might be missed when collecting the user-behavior trajectories. Customer data might come from websites, membership systems, EDMs and so on, so how we build a complete user journey is the key to collecting data. For Analytics, not only AI modules, but many statistical reports or analytical methods play important roles in this stage. The last step, Application, is where we export data, by sending text messages, emails, or even some interactive ways to engage customers. The reason why we emphasize DAA as a flywheel is that the data generated in the final step, Application, should be returned to the step of Digitalization, so that people can understand the entire user journey and the user profile. This is the so-called omni-channel integration, in which we find the user's behavior and profile in a human-centered way.