“data messing”, like blind dive in data, does not work as a strategic decision-making tool. We provide data analysis to meet strategic choices and importance, business objectives, management goals, through using various statistical analysis techniques such as multivariate analysis, visualization, algorithm, machine learning etc.

About Strategic Data Analysis

We know many cases where as a result of doing “data messing” to perform business data analysis, a large amount of analysis result material accumulates but it does not result in anything.

The way of data analysis which is really necessary for business strategy is to derive the “decision choices” and “strategy map” to handle an enterprise. Such analysis should inseparably include project purpose, potential actions, expected execution outcome.

In order to obtain the analysis result for corporate strategy, for example analyze the customer purchase history and ID information covering typically tens of millions of people, to clarify the behavior tendency of customers. Some of purposes should be satisfied instead by mass market surveys to clarify the psychological tendencies of customers and users.

To complete such analysis, we will use various methods from just simply summarizing and visualizing data to a group of statistical methods called multivariate analysis, or more complicated machine learning methods, but such methodologies should always be chosen based on the purpose. Just to use trending brand-new technology is actually quite useless for many occasions, but rather what is essential is deep understanding and extensive experience of applying data science to corporate strategy.

We now observe on one side companies that have “data driven corporate management” that can handle “strategy map” and “decision choices” through such strategic data analysis, and on the other side companies that are still “somehow” play with data without clear direction. The difference in strategic competence of these two sides is widening day by day in today’s global competitive environment.

Service configuration example

Various data can be used depending on the purpose setting and the conclusion to be issued. However in particular for corporate strategy, in order to expand sales and profits of existing businesses and verify potentials of new businesses, the most important area should be to deeply and widely understand customer persona and its diversity. Major methods includes log data such as purchase history, and a large-scale questionnaire survey.

  • Statistical analysis of customer log data
    We use the sales history database with customers / user IDs to find out customer characteristics and their distribution, which is the basis of various business strategies such as product development and promotion strategy, based on large scale real data. Definition of data for classifying customers by behavior, use of multivariate analysis method, visualization of analysis results to lead to concrete action at the business site, are essential points to determine success or failure.

    (Analysis example)
    – Typical purchasing pattern of customers: Weekday / time of a day, store type, multiple store purchasing, product category, purchased category combination, number × unit price, LTV
    – Create customer clustering based on similarity of purchasing behavior, create customer map, illustrating personas
    – High LTV customers, Low LTV customers features, growth path from newcomers to high LTV customers
    – Identification of new entrants and verification of action factors related to retention or dropout
    – Analyzing factors of dropout in marketing funnel and its each process by cross-analysis with web access log etc
    – NPS and other customer satisfaction survey data to verify the effectiveness of CS improvement and LTV expansion measures

    (Technology and know-how to use)
    – Multivariate analysis such as clustering, factor analysis. correlation analysis, selection of optimal methods from various statistical methods, parameter adjustment methods
    – Processing of customer data based on the database structure, recalculation of variables, composition of variables representing customer characteristics
    – Expression technology of data visualization for strategic decision making, derivation of interpretation
    – Indicators that should be noted according to business structure and business model, design and calculations of KPIs
    – Set action hypothesis by business structure understanding and customer’s psychological understanding

  • Market research and statistical analysis
    Many companies are satisfied with just analyzing the database in their company, but it does not provide any information about the share of customers’ wallet including customers’ use of competing other companies. Customer spending to competitors does not appear in log data. Access and purchase log also do not include any information about psychology at the time of purchase and situation of the scene.

    It is essential to acquire data of a large market survey by questionnaire method, when we want to describe the customer characteristics and its diversity as a basic map of corporate strategy, as we need customer information including competitors share and also including not only customers’ explicit behavior but also their implicit needs and background lifestyle. Implementation of market research is now easier with modern Saas services, but survey design for strategic purposes tends to fail if there is no experience-based know-how’s.

    (Analysis example)
    – Share of the entire market including own company, competitors and substitutes, creating maps to depict the situation
    – Own and competitors’ share estimates for each purchase product trend, purchase unit price, customer attribute (gender, generation, family, occupation, preference)
    – Pass rate of each marketing funnel x factor analysis, through, recognition, interests, consideration, purchase, use and recommendation
    – Customer clustering based on customer needs and lifestyle, customer map, persona depiction
    – Customer’s tendency of recognition of own company / competitors. Recognized image, recognition path validation, visualization of whole situation

    (Technology and know-how to use)
    – Survey questionnaires design know-how for grasping the real behavior and psychology of respondents
    – Setting of extraction condition of respondent population, two-step design of screening survey and main lengthy survey
    – Know-how to eliminate contamination of dust data such as dishonest answers for accurate survey data acquisition
    – Aggregation method that draws customer psychology and behavior by combining answer data of many question items
    – Expression technology of data visualization for strategic decision making, derivation of interpretation
    – Utilize action hypothesis based on business understanding and customer psychology understanding from the stage of designing survey questionnaires

Service process
  • Status assessment, Method selection
    When you contact us, firstly we will ask you about the current situation. We will examine what is necessary now, what kind of method is best for the current purpose, etc.
  • Analytical design
    We explain expected deliverables, required period, cost estimation, role setting etc., and if you get internal approval, we will start the actual work.
    (NDA exchange if necessary at initial contract)
  • Data acquisition
    – When using data from the internal database, firstly decide the necessary data set, then export the data, receive it, and start the analysis work.
    – When using market survey data such as questionnaires, start with preparing survey design, then proceed into coordination with research provider, survey implementation, data collection.
  • Analysis work, meetings
    Before providing all detailed analysis results in 2-3 months, we will regularly hold meetings for sharing intermediate results, discussing the direction to advance analysis, or if necessary statistical methodologies lectures can also be provided.
  • Deliverables
    Finally, in addition to the electronic file of the analysis report materials, in some cases we also provide intermediate processing data, calculation procedure manual etc. according to the necessity of subsequent development.
Experience example
  • Manufacture and retail company
    The customer membership system was introduced, and the purchase database to which the customer ID was given accumulated for several years.
    Until then, product planning was conducted following employee’s personal experience, so this time data analysis was conducted aiming for more precise product planning based on the data.
    By systematically classifying customers’ purchased products, we create customer type division by creating variables to index product preference differences. By specifically visualizing each type of persona and sharing it with product planning members, we have established a product lineup process in the company based on statistical data-driven customer understanding.
  • Food retailer
    Since purchasing database with customer ID accumulated following installment of full-fledged point card platform, we explored effective utilization for digitalization of marketing.
    Analyze purchase history data and sales item data to create customer clusters. Use the analysis results in a variety of ways, such as validation of promotion measures, customer targeting at creative creation, CRM measures through mobile application development.
  • And many others

If you are interested in our services, please let us know from the contact form.

You can also find contact information at the company information page.