Big data is cooked, have you been killed?

A few days ago, the State Administration for Market Regulation issued an announcement for public comments on the "Regulations on Administrative Penalties for Price Violations (Revised Draft)". It stipulates that for big data killing, a fine of more than 1% but less than 5% of the total sales of the previous year will be imposed, and if there is any illegal income, the illegal income will be confiscated; if the circumstances are serious, it will be ordered to suspend business for rectification, or revoke its business license.

So far, there is no clear definition of what "killing" is in the legal profession, but when dealing with such cases, appropriate rulings are made from the perspective of safeguarding the overall interests of society. As for what is the overall interest of society? I believe there will be different interpretations in different times.

To know if you have been "killed", you must first understand what "big data kills".

1. E-commerce platform operators use big data analysis, algorithms and other technical means, according to the preferences of consumers or other operators, transaction habits and other characteristics, and based on factors other than costs or legitimate marketing strategies, the same commodity or service. Set different prices under the same trading conditions;
2. In order to crowd out competitors or monopolize the market, e-commerce platform operators that do not yet have a dominant market position dumped at prices lower than the cost through subsidies and other forms, disrupting the normal production and operation order, and harming national interests or the legality of other operators. rights and interests;
3. Other unfair pricing behaviors of e-commerce platform competitors as stipulated in administrative regulations and departmental rules.

This article focuses on the first point, because the second and third points are both business strategies and not directly related to the application of technology.

Various e-commerce platforms (not just for mainland online shopping platforms, in fact, the same is true for online shopping platforms all over the world), before the term "big data" appeared, they have been collecting all customer data for basic analysis, but they have not been fully utilized. . Generally speaking, the collected customer data includes age, gender, education, work industry, income, etc., while the order-related information includes the selected item and quantity, the date and time of the order. In this way, companies can use simple data integration methods to understand their daily turnover. For the company, it's just understanding what's going on, not making further interpretations of the data.

Having said that, there will always be people who are curious about data and try to find a little more useful information to help their company grow. With limited technical support, use the data analysis platform to try to explain the reasons for this result (for example, find that a certain item is purchased by a certain major customer group). If the amount of data is small, some valuable results can be found, but if the amount of data is large, it is impossible to start.

These are all things that have happened in the past.

After the advent of big data technology, most data analysis has become very simple. Using artificial intelligence and machine learning technology, computer programs can handle these complex and tedious data analysis tasks. We must first identify one or several clear data analysis goals, set the basic parameters, and let the computer program complete the rest of the work and feedback the analysis results. What's the difference after using machine learning techniques? To put it simply, there is no problem in drawing a two-dimensional or three-dimensional chart by hand. If you were asked to use a chart to express a ten-dimensional data group, how would you interpret it? Fear not, the computer will do the work for you.

More importantly, the amount of data we generate (and collect) on a daily basis is increasing, and its forms are becoming more and more varied. To find hidden relationships from a large amount of data, big data and artificial intelligence technologies are definitely good helpers. . Using these methods, we can find more accurate and targeted customer groups.

It's time to connect the present tense with the future tense and find a way to make the same thing happen in the future.

This stage is the predecessor of "big data killing", because it is not "familiar" enough.

Think about what will happen to you before the major telecommunications and network providers complete their contracts. Three to six months ago, you will receive a renewal email and phone call to remind you that the contract is about to end, and now there is a special offer (delete a thousand words below). do you know? There are many users who agree to renew the contract immediately, but those of you who believe in "clear heart" will know that it takes a few rounds (two or three weeks as soon as possible, or two or three months as slow) to get the truly best plan combination.

Frankly, if the company knew I was willing to pay $300 per month to buy a service, how could they tell me that the same service would actually cost $250, right?

Although these renewals may only happen every one to two years, if the various e-commerce platforms already have a good understanding of my background, behavior, and affordability (or expectations) for each item or service, I will After becoming "familiar" enough, slowly start a round of fighting, how would you feel? ("Unfamiliar enough" can refer to the confidence of a machine learning algorithm to predict customer behavior.)

I saw a comment from the mainland saying, "These actions have harmed business ethics and social trust, and it is not a shame to make this kind of money." And these behaviors must be regulated.

After all, what to regulate?

One of the main purposes of big data analysis is to find out consumer preferences and habits, and to provide the most personalized product or service recommendations. These actions are above board, and it makes sense to consider these considerations as part of a marketing strategy. All that's left is price.

Some travel and food delivery platforms in the mainland offer old users and VIP users much higher prices than new registered users. This kind of "killing familiarity" is not surprising in Hong Kong. Generally, the discounts for new users are more attractive. Therefore, some users will continue to switch to new platforms, use the service as new customers for a period of time, and then switch to other platforms. Even if you register with a mobile number, there is always a way to take advantage of the welcome benefits.

Did I get killed? Of course there is! Sometimes for convenience, sometimes to save time, and sometimes to avoid trouble, I didn't think so thoroughly when making a decision. At an affordable level, being injured or even killed by "big data" is inevitable. Of course, if you feel cheated, it is recommended to share and compare with your friends.

Technology has always been just a tool. If we use technology to improve our lives, the focus is on. With the rise of artificial intelligence and big data, the results are also good and bad. "Big data kills familiarity" is an example of how technology affects our lives. While enjoying the convenience brought by technology, we do pay a lot of price silently. Don't expect those regulators to provide you with adequate protection for everything in your daily life, being a high-profile consumer is king.

Facebook: https://www.facebook.com/drjackeiwong/
Instagram: https://www.instagram.com/drjackeiwong/
YouTube: https://www.youtube.com/@drjackeiwong