[Data Analysis from Scratch #4] What beginners need to know about data analysis, what management expects from data analysis and analysts

Implementing business improvement actions based on the results of data analysis takes time and money, so you will need to get the go-ahead from the organization's management before you can implement them.

When you think, "I need to get the management involved!", the first thing you should do is:It is important to correctly understand what management wants from data analysis.

In this article, I would like to explain what management wants from data analysis and what they expect from data analysts.

↓ Click here for a list of articles on "Data Analysis from Scratch"

#1 "8 steps of analysis" that beginners should know first
#2 Three analytical techniques that data analysis beginners should remember
#3 Communication tips to get management involved that data analysis beginners should know
#4 What data analysis beginners need to know: What management expects from data analysis and analysts
#5 4 tips for field staff who are new to data analysis to get management and the organization involved

What do executives want from data analytics?

The "principle of victory" required for data analysis

In a nutshell, what management wants from data analysis is"The Rules of Victory".

So what exactly is the formula for victory?

First of all, the premise is that "data analysis" and "human thinking" are two wheels that go hand in hand.
Data analysis alone, or human thinking alone, will not work.

For example, imagine placing a jump platform at the base of a mountain and trying to jump over a high mountain.

In this case, the jumping-off point is data analysis (= logic), and the jump is human thinking (= creativity).

The higher the jump platform, the greater the chance of jumping over it.Data analysis plays a role in raising the jump platform.

By building up logical correctness through data analysis, human creativity can soar to greater heights.

What management wants from data analysis is a springboard that allows people to jump over mountains with confidence. In other words,Logic and strategies to increase your chances of winning

Two suggestions for data analysis

I explained that what management is looking for from data analysis is the "rules for winning."

For executives who want the "rules for victory," it is not enough to simply present the analysis results. So, what specific proposals do executives want to see through data analysis?

There are two main proposals that management is looking for:

1) Causal hypothesis
②Next Action

1) Causal hypothesis

The 1 one isCausal hypotheses behind the results.

Causation means cause and effect.
A hypothesis about what caused the results of the data analysis.

For example, let's say you work for a beer manufacturer.
After analyzing the data, we found that the coronavirus pandemic is actually boosting our company's sales.

In such cases, it is extremely important not to rest on the analysis results but to go a step further and pinpoint the cause (formulate a hypothesis) such as "Maybe it's because the culture has changed to drinking alcohol at home amid the stay-at-home demand," in order to take further action.

We would not be able to notice these causal relationships if we relied solely on the human brain.Data as facts allow us to notice problems that we had not seen beforeThis is where data analysis comes in.

② Next Action

The 2 one isA feasible next action with a high chance of winning.

After understanding the causal relationship of the data analysis results,What action should we take?This is the next action.

Here's another example.

A data analysis of regional disparities revealed that "increasing public subsidies to local areas will not narrow the gap"(*1).

We hypothesize that the following causal relationships may be responsible for these results:

Subsidies are public funds and must be used within a set period of time. This means that local governments have no choice but to make short-term investments without being able to fully formulate business plans (cause).
As a result, it ended up just constructing buildings and holding typical events, and did not lead to the fundamental correction of regional disparities.

What next actions could be considered here?

For example, if we consider the topic of "the type of subsidies that can enhance the brand power of local areas," the following actions could be considered:

In order to increase the competitiveness of local areas, it is important to accumulate stock assets throughout the region (things that will increase the region's brand power and attractiveness in the medium to long term).
How about asking them to create medium- to long-term business plans and considering a subsidy system that can be used to foster intangible assets such as designs and brands that may not produce short-term results but will increase the value of the region in the future?

in this way,Consider causal relationships based on data and use human creativity to suggest the next action.

These are the two suggestions that management is looking for in data analysis.

(*1)See
Chuo Online "Will providing subsidies to rural areas lead to regional revitalization?" https://yab.yomiuri.co.jp/adv/chuo/research/20220224.php
Cabinet Office Regional Revitalization Promotion Office "Report on the Effectiveness Verification of Regional Revitalization Promotion Subsidy Programs (R2_houkokusyo-suishin.pdf)"https://www.chisou.go.jp/sousei/pdf/R2_houkokusyo-suishin.pdf

What do executives want from data analysts and marketers?

Next, what qualities do management expect from people who put into practice the "proposals that management wants" that we have discussed so far?"What executives expect from data analysts and marketers"I would like to organize this information.

Here, we will introduce three skills that a data analyst must have (must skills) and three skills that we think would be great if they could achieve them (excellent skills).

Must Skills

• Deep understanding of company-wide objectives and issues and a high level of perspective / Thinking based on company-wide objectives

Data analysts at companies need to have a deep understanding of the company's overall goals and issues, and think about things from that perspective.High Perspective is required.

Sometimes, people start their data analysis from what they want to do, such as "I want to use social media, so let's analyze it." However, such data analysis is not essential, and it is often difficult to put it into practice.

• A neutral stance as an analysis and marketing professional

Even if you belong to a company,Approach data analysis with a neutral stanceIf you give priority to the company's policies or the wishes of the management, there is a risk that the accuracy of the analysis will be compromised, incorrect decisions will be made, and ultimately, significant damage will be done to the business.

As professional analysts, I would like you to take on the role of conveying the correct judgment without any bias, saying, "The president put his heart and soul into creating this commercial, but it hasn't had any impact at all!"

• Ability to correctly explain analysis results

Many of the analysis results are difficult to judge or ambiguous. Among them, the management is troubled by:Should we take the results positively or negatively?, That is the point.What to judge and teach as a professional.

そ し て,If there are any concerns about a hypothesis or action, they will communicate them correctly..

This is what managers who want to make the right decisions with the highest chance of success want:Ability to correctly explain analytical results.

This also relates to the second point, "a neutral professional stance," but the analysis results may match the desired hypothesis, but the sample size may be insufficient, so it is best not to trust the accuracy. This happens often. In such cases, I would like the background to be properly communicated.

Excellent skills

• Changing the language for non-professionals in data analysis

As an analysis professional, it is important to have specialized knowledge. However, management and members of other departments may not know the technical terms of data analysis.Language shift.

For example, when you ask management to make a decision, you probably won't understand if they say, "President, the adjusted coefficient of determination is very good."

If you have the skill to translate your ideas into language that even an analytical layman can understand, your proposals will be more likely to be accepted.

• Speedy implementation of PDCA cycle

The business world continues to change at an incredible speed.

I think that many people in charge of analysis originally aimed to become scholars or researchers and write papers. When writing a paper, you spend 8% of your time and aim for 2% accuracy, but in the business world,2% of the accuracy in 8% of the timeI want you to ask for this.Have a sense of speed that suits the business worldIt is important.

• Management skills to get people involved in projects

Ability to identify suitable personnel and key peopleThen, at two important times (*2) when the hypothesis is designed and when the interpretation is considered, get everyone involved.The ability to dynamically manage projects.

Having this kind of project management ability is expected to greatly advance the entire business. I think that someone who has this kind of driving force in addition to high professional data analysis skills is an extremely valuable asset.

(*2) See
[Data analysis from scratch #3] Communication points that beginners in data analysis should know to get management involved | XICA Co., Ltd.

Free downloads of related materials

The 8 Essential Steps for Turning Data Analysis into Business Results
~What are the important points to keep in mind when analyzing data that can be used in business settings? ～

Download Now

Recommended books for those who want to learn data analysis from scratch2 selections

We have talked about what management wants from data analysis and what they expect from data analysts. Finally, we would like to introduce two books that we would like you to read from the perspective of "involving management and the organization."

Koji Mitani (2013) "Complete History of Business Strategy" Discover XNUMX

This book will help you to have a higher perspective as a data analyst.
This book is aimed at beginners who are wondering, "What is management?", and systematically teaches the history of management strategy. You can learn about the history of management from the past to the present, and raise your perspective from the field of marketing and analysis to management.

Stephen G. Blank (2016) "Entrepreneur's Textbook" Shoeisha

The basic premise is that this book shows you how big an impact verification can have on your business. This book is a must-read for entrepreneurs and those in charge of launching new businesses, and it teaches you the steps to create a business while verifying it, allowing you to imagine the magnitude of the impact.

*Link to Amazon product page is provided.

↓ Click here for a list of articles on "Data Analysis from Scratch"

#1 "8 steps of analysis" that beginners should know first
#2 Three analytical techniques that data analysis beginners should remember
#3 Communication tips to get management involved that data analysis beginners should know
#4 What data analysis beginners need to know: What management expects from data analysis and analysts
#5 4 tips for field staff who are new to data analysis to get management and the organization involved

↓ The business media "PIVOT" explains how to utilize data science in business.

Yoshiaki Hirao, CEO of XICA Corporation 

After his father's bankruptcy, he strongly wanted to "eliminate the hopeless sadness in the world." He encountered statistical analysis while studying at the Faculty of Policy Management at Keio University, and founded XICA Co., Ltd. in February 2012, just before graduating. Before founding the company, he also had a unique career as a band member.