Can the "Business Study experiment" under large data change the supermarket industry?

Condensed point of view

Lack of consumption power, profit decline, electricity business erosion, the road to the breakthrough of large supermarkets? Mastering the 5 Directions of "Commercial experimental learning" based on large data analysis will bring hundreds of millions of economic benefits to the retail enterprises.

The retail business may produce exciting but risky ideas every day. What's the effect?

"Business learning experiments" under large data

Taiwan's family of convenience stores, in Taiwan has more than 2000 stores. The survey found that consumers had a significant demand for freshly brewed coffee. For the whole family, is it a rash to offer freshly brewed coffee at all outlets, or will these coffee machines be given priority in some stores? Do you want to squeeze the sales of other coffee and beverage products at the same time that you are cooking your coffee?

In the past, experienced managers have intuitively formed strategic assumptions. Now, they can use the "experiment and learning" software based on the large data analysis of APT, more cautious data--pick up some of the different ways to test this new idea in small stores, learn from the results, and find the most successful versions of corporate revenues, net profits, ROI, and market share, To design a set of best promotional programs, and then fully promote. This systematic approach is called "Experimental learning".

First of all, APT will be more than 2000 stores of information into a data platform, and within two weeks of the past 3 years to facilitate the family of thousands of cash registers to convert the transaction into a database, within two months for each store to create a property file, including the size of the store, the surrounding population age, employment status, competitive business distance, Geography and so on all the elements of business.

After that, APT chooses a batch of shops, divides into the experiment group and the reference group carries on the experiment. The question of adding coffee machines to the family's convenience store was quickly answered by "commercial experimentation", and they found the priority of importing coffee machines and learned which locations did not require a coffee machine, and which shops upgraded their net performance after they were imported into the coffee machine.

Similar choices have been seen in the taste of race. The company plans to launch a one-foot-long 5-dollar submarine fort in the North American market in 2008, hoping to bring more business to the sale, but fears that the promotions will affect the sale of higher-priced sandwiches.

In order to test the viability of commercial innovation, they also design an experiment with APT, after collecting all the information of the restaurant, the classification tool selects the experimental and reference groups which have the near-uniform dining area, geographical location, surrounding crowd structure and the characteristics of the business district. In the experimental cycle, the overall restaurant sales performance, 5 U.S. dollars submarine Fort Sales and other sandwiches sales. After the experiment, the comprehensive Financial computing tool on the platform can provide the result in a few hours, showing that this special submarine fort really brings the performance promotion to the enterprise.

"Commercial experimental Learning" of the author of Apt (Applied predictive Technologies, Asia) was founded in 1999, the main business focused in the mature North American market, is a "experimental learning" to assist the Enterprise optimization strategy of the software company, Starbucks, game hundred flavor, Kraft Foods, Wal-Mart and other companies have applied apt products. At present, apt in the Asia-Pacific region for China, Japan, Singapore and Australia and other more than 10 major markets to provide customer service.

5 Directions of Commercial Experiment Study

In the current lack of consumption power, the cost of the rise in the case, coupled with the erosion of the electric power of the large supermarkets began to seek new development path. "For large, storefront supermarkets, hundreds of distribution networks are a valuable source of knowledge and information from which they can learn the lessons of success or failure." "APT Vice president and Asia Pacific head Li Zhanhong said. He said that "commercial experimental learning" will bring the following 5 aspects of the supermarket industry help:

1. Enhance ROI of O2O. O2O's intention was to increase revenue by selling online and attracting new customers who would otherwise not have been able to go to those stores. But the fact is counterproductive, some store new customers did not grow at the same time, but also generated a huge operating costs. Li Zhanhong said that the "online shopping, door-to-door" test for a few sample stores would reduce the risk of such initiatives if the programme was to be promoted. And then, based on the data, locking and accurate promotion of the significantly profitable storefront will improve the ROI of the program more effectively. Supermarkets have a lot of unique advantages, such as those short shelf life of food, fresh food, perishable perishable goods, are the short board of electric dealers, supermarkets may wish to do bigger and stronger.

2. Optimize the poster. Poster promotion is a supermarket idiom, but does it really bring profits? The poster design focuses on two issues, the distribution of posters and the products to be put into them. For the former, the previous posters issued by the mainstream people who are responsible for household procurement aunt, they compare the supermarket posters, buy the cheapest goods, for the latter, may be a special beer has been a huge sales growth, supermarkets may think that the promotion is quite successful. The truth is most likely that most consumers have bought only special beers without passing any other goods, and that behind the false boom is the profit of the supermarkets.

Through "experimental learning", the poster distribution to the different customer base, to attract more young people to go shopping in the supermarket will bring more profits; As to what goods to put, supermarkets need to conduct a large number of transaction data analysis to design which preferential products. For example, placing a potato chip or salad next to a beer may increase the customer's single sales amount.

3. Expand its own category. Private Brand is a big magic weapon in the supermarket operation, in this field, Wal-Mart, China Resources million, Watson, wanning and other supermarkets and chain stores have been working in recent years. According to the research report, the U.S. supermarket industry's own brand nearly 3 years of sales growth of more than 18%, is the growth rate of other brands more than twice times.

For retailers, the expansion of their own category can not be simple imitation of successful manufacturer products, which will result in a large number of product sales bad. In order to create more interest in expanding its own product line, Li Zhanhong said, supermarket managers need to strategically optimize their portfolios to avoid a decline in trade and damage margins. The supermarket enterprise may use the transaction data analysis, understands which item often appears in the high amount transaction, and is often bought by the high value customer, optimizes the own brand growth strategy.

In the case of Sam's Club, Sam's business in China aims to meet the needs of business members in the sales, office and distribution of benefits, while meeting the demand of high income individual members. Therefore, the SAM member store sold 8 pieces of "member excellent products" soap, the whole box of copy paper, the actual sales data also proved that these transformed products are the model of the success of their own brands.

4. Precision Group Promotion. By binding membership data and transaction data, supermarkets can carry out grouping promotions so that they can accurately understand the substitution effect of each promotion, the phenomenon of delaying consumption, and the variation of different consumers in various factors.

A special beer promotion, for example, could allow consumers to buy only a large amount of discounted beers, leading to a fall in future sales, thereby damaging gross profits. A small number of customer tests will reduce the risk of a significant reduction in margin and help companies understand which outlets can generate additional benefits, and which outlets are just for the benefit of customers, because customers will buy the product anyway, resulting in a decline in gross margin.

5. Measure SNS Marketing. Many supermarkets use microblogs, micro-letters and other SNS tools as a channel to deliver promotional and special offers, as they become the main customers of supermarkets. Through experiments, enterprises can compare the market performance difference between the experimental group of advertising and the control group without advertising, avoid wasting advertising investment, and focus on the more effective market and customer base.

(Responsible editor: Mengyishan)