What is the job of a data analyst to support management decision-making?
The work of XICA's data analysts goes beyond simple data analysis. They see data analysis as a means to solve problems, and are particular about "usable data" that can be used to inform management decisions. What kind of work do XICA's data analysts do? What do they value? We spoke to Nishizu Taira, head of the ADVA Analysis Department, who is responsible for XICA's data analysis.
Professionals in "advertising x statistics" who propose data that can be used in business
──First of all, please tell us the role of a data analyst at XICA.
Analyze advertising data at a high levelProfessionals in "advertising x statistics"But, as a data analyst at XICA, the analysis results are not the end of the story.It is important to link this to management decision-making and actionTo that end,The role of interpreting the results, such as why the results were obtained, what the expected results were, what the likely causes are, and what kind of data should be analyzed in the future.I will carry the.
──Please tell us about the organizational structure of the ADVA Analysis department.
The ADVA Analysis Department is divided into three organizations: the MAGELLAN Division, the IES Division, and the Research Division.
The MAGELLAN Section is a team that uses data science to work alongside customers to solve the problems they have heard from them. The IES Section works with the IES Division, which handles advertising buying and creative production, to perform in-depth analysis of TV commercials for customers who have issues with advertising, particularly TV commercials. The Research Section is responsible for systematizing the knowledge accumulated by the Magellan Section and the IES Section, turning it into reproducible know-how, and for researching and creating new analysis methods when new problems emerge that cannot be solved by current solutions.
How does a data analyst carry out their work?
Our work is carried out in pairs, with one data analyst working face-to-face with a consultant.
First, we listen to the issues the client wants to solve, and ask in detail about what they want to know and what actions they want to take with the analysis results. Next, we agree on the purpose of the data analysis based on the information we hear from the client. After that, we ask the client to collect the necessary data, which we then analyze. Finally, we submit a report tailored to the issues the client wants to solve, and together with the consultant, we propose an action plan.
This is the basic PDCA cycle, and this series of steps is repeated with a sense of swiftness.
Motivating people with proactive proposals
──It seems like you have a lot of opportunities to communicate with customers.
That's right. I think some data analysts like to stay alone and tinker with numbers in silence, but the data analysts at XICAWe will even take on a role similar to consulting, proposing effective ways to utilize data while interacting with customers..
So,"Speaking science students" who are good at communicationThat's a good fit for me (laughs).
--I imagine the ultimate goal is for customers to take action to resolve their issues, but what is needed to achieve that?
It’s difficult to motivate people. It may be surprising, but what is absolutely necessary isindependenceI think the trigger for people to take action isAmbitious proposals that are thought out ahead of time, and deep suggestions that come from a deep understanding of the customer's situation.If that aspect is shallow, it's difficult to get analysis results that will move customers.
How much knowledge of statistics is required?
Of course, a certain amount of knowledge is necessary, but it is also something that you will naturally acquire while working.The ability to proactively think about your customers is more important than knowledge of statistics.
Speaking of skills,When it comes to data analysis, setting goals is the most important thingSo, I think it's enough if you have at least one experience where you say, "For this purpose, I made this hypothesis, did this analysis, and as a result, my actions changed like this before and after the analysis," regardless of the industry or analysis.
How will you get started in your work once you join the company?
New employees are always assigned a senior employee who acts as a mentor. After that, they undergo a full month of training. In the second and third months, they are involved in various projects as part of OJT to help them learn the detailed work. From the fourth month onwards, they are given their own projects with senior employees and learn how they are actually carried out.We have an onboarding system in place that provides thorough training over a span of around six months.So it's okay even if you don't have much skill or experience.
Build up your skills through a wide range of learning and practical opportunities
--As data science professionals, how do you understand your customers' business challenges?
You work with a consultant who will pair you up.
The role of the data analyst is to compile the results of the analysis and come up with suggestions and interpretations from the results that can help solve the client's problems, while the role of the consultant is to consider how to propose the suggestions and interpretations that the data analyst has come up with so that the client can use them.
Our consultants, who mainly work face-to-face with clients, listen in-depth to their real issues on a daily basis. Through thorough communication, we come up with analyses that can be used for business decisions.
──How do you keep up with information necessary for your work, such as analytical know-how, advertising trends, and business models in various industries, outside of the projects you are in charge of?
Share reports on all projects at department meetings attended by all members. Many projects already have know-how within the company..
later,Study sessions will be held once a month with consultants.Here, we will learn about specific analysis cases. In addition, there are many other opportunities to learn, as independent study meetings for case studies are actively held.
Thinking on the move: speed and logic come together
──The ADVA Analysis Department, mainly the Research Division, is also responsible for providing analytical know-how to customers. What kind of benefits do customers get from having analytical know-how?
In fact, what corporate advertising and marketing people are most concerned about is"What are the trends in this industry?" "What are the latest advertising trends?"That's what I'm talking about. Perhaps this is something people want to know more about than the effectiveness of their own advertising.
We have conducted data analysis for over 100 national clients so far. We have accumulated a track record of comprehensive analysis that is unbiased in a good sense. When we want to respond to requests to know industry trends, the data that we have systematically compiled from the analyses we have accumulated is very useful knowledge.
Understanding industry trends and advertising trends will help you understand your own company's analytical trends and dig deeper into implications. Advertising also goes in and out of fashion, soThis will also help you determine the timing of when to do something and speed up the implementation of measures.Right.
──Please tell us what the ADVA Analysis department values.
ADVA Analysis DepartmentAn organization that will continue to hone its data science technology, which is the root of XICA, while also taking on new challenges..
To overcome this, there's a teaching called Things to think about while runningIt's important to have your own logic, think about what would happen if you tried to do something like this, and move forward with a sense of urgency, correcting your course as you go.Data science where speed and logic coexistI would like to continue updating it.
◆ Related Articles
Using data analysis as a management tool. ADVA Analysis Department launched | XICA Corporation
https://xica.net/career/recruit/000097-2/
*The affiliation and title of the interviewee are those at the time of the interview.
