Big data prediction | Is Xiaoming Bicycle’s strategy of charging 10 cents for a bicycle feasible?

Gartner, a research organization, gave the following definition for “ Big data”. "Big data" requires new processing models to have stronger decision-making power, insight discovery and process optimization capabilities to adapt to massive, high-growth and diverse information assets.

The definition given by McKinsey Global Institute is:

A data set that is so large that it far exceeds the capabilities of traditional database software tools in terms of acquisition, storage, management, and analysis. It has four major characteristics: massive data scale, fast data flow, diverse data types, and low value density.

As a newbie in big data, I certainly cannot write such a professional and insightful definition. Even so, I still mustered up the courage to summarize two basic understandings of big data:

1. Big data is like an archaeologist that can recreate history and images from thousands of years ago. Her restoration is so realistic, it's as real as if she had experienced it herself;
2. Big data is like a prophetic scientist who can tell you in advance what will happen tomorrow. Her predictions are so accurate, just like telling you that the sun will still rise tomorrow morning.

As for the relationship between big data reproducing history and predicting the future, such as: which one is easier, which one has more application value, which one is more magical, and what are their application scopes. Although I have thought about it, unfortunately, I don’t fully understand it, so I won’t discuss it in depth. Today we will start from the small things and do a simple analysis using an example that everyone is familiar with. After much thought, I think it would be better to use big data to predict the future, because things have not happened yet, and the predicted results can be verified in the near future, which is more convincing. It’s a beautiful thought!

The next step is to choose a good example. Coincidentally, there has been a lot of attention on shared bicycles recently. Shared bicycles such as Mobike , ofo, Bluegogo, Youbike, Qibei , Xiaoming, etc. have sprung up in the streets and alleys of major cities like mushrooms after rain. Shared bicycles are no strangers to everyone. When I saw the charging standards of Xiaoming Bicycle some time ago, my eyes suddenly lit up. Taking this as an example, it couldn’t be better. Therefore, the theme of this article is: to predict the feasibility of Xiaoming Bicycle’s 1-cent bike usage strategy through big data.

First of all, we need to understand the charging method of Xiaoming Bicycle:

• Deposit : The deposit for using Xiaoming Bicycle is RMB 199 (refundable);
• Fare : The fare is charged in sections, ranging from 0.1 yuan to 1 yuan per half hour, with the initial charge being 1 yuan. The fare gradually decreases with the number of invited friends (if the total number of invitees or invitations reaches 10 or more, the fare is 0.1 yuan per half hour; if you invite one friend, the half-hour fare is discounted to 0.9 yuan, and if you invite two friends, the fare is 0.8 yuan per half hour, and so on. The cumulative discount limit is 10 people). There is no cancellation fee. The charge for using the bicycle starts from the moment the bicycle is unlocked. The charge will end only after manually locking the bicycle, opening the APP and clicking on End Ride.

To be honest, when I first saw Xiaoming Bicycle’s billing method, I thought it was quite good, because at that time I was still stuck in traditional product thinking and did not realize the existence of big data. Therefore, the first thing that comes to mind is that this billing method wants to achieve the purpose of attracting users and cheap promotion through (third-party) social attributes, small profit drive, person-to-person, and word-of-mouth strategies, and unexpectedly remain invincible in the increasingly competitive shared bicycle market. At first glance, this strategy may seem a bit old-fashioned, but it is still a wise move.

So, can reality really develop as expected? Now, let’s analyze the feasibility of this strategy from the perspective of big data, as well as its future trends and outcomes (if the analysis is inappropriate, please feel free to complain, but don’t use bricks!)

In order to explain the problem systematically, let's first set up a simple market scenario:

(1) Location : Shenzhen (Since Xiaoming Bicycle was first launched in Shanghai and Guangzhou, it would be better to use these two places as the research objects.)

(2) Population : 11.9084 million (resident population in 2016), plus non-resident population, should be no less than 20 million

(3) Applicable objects : Shenzhen is a very young city. Excluding the elderly and children, there are of course some other people, and conservatively estimated that more than half of the population is young. The market potential is huge and fully meets the requirements of big data scenario applications.

(4) User psychology : They are mainly young office workers with wide social circles and complex relationships. In recent years, the marketing model of mutual promotion through social relationships has, to a certain extent, consumed the enthusiasm of young people (this can also be regarded as a disadvantage of this strategy).

(5) Competitive environment : There are no less than 20 companies with similar products, including Mobai, ofo, Bluegogo, Youbai, Qibei, and Zhixiang Bicycle. The competition is fierce and the risk of being eliminated is relatively high.

Based on the above market scenario, in order to roughly calculate the average charge of Xiaoming Bicycle, a simple mathematical model is established as follows:

• Sample size : about 1,000 people (a larger sample size would be more accurate.)
• Study period : half an hour (e.g. 17:00 ~ 17:30 in a day)
• Frequency of car use : once per person (this is considered an ideal situation, that is, each person uses the car once)
• Discount strength : 0.1*999 = 99.9 (yuan). Each user can be invited at most once, so the discount amount is 0.1 yuan. The first user enters actively, so there is no discount (which can be ignored under big data).
• Ideal charge : 0.9 yuan/half hour (from the above analysis, it should be easy to come to this conclusion).

Conclusion 1: Judging from this model, the company's actual profit is definitely more than 0.9 yuan per half hour. Compared with other shared bicycles that charge 0.5 yuan per half hour, the profit margin is much larger, and the strategy seems to be good.

In order to make everyone understand better, the above ideal model is converted into a data model, which is shown in Figure 1 in the form of a data table:

Figure 1

From the table data in Figure 1, we can see that the proportion of users who charge 0.1 to 0.5 yuan for a half-hour ride is extremely small (even in the ideal extreme case, this part of users accounts for about 16.7%, while in an ideal case, it is less than 2%, and the actual situation is even less). Why do we need to pay attention to this group of users? Obviously, this has something to do with the competitor's charge of 0.5 yuan per half hour. Looking back at the mathematical model under ideal conditions, it is obvious that it is quite different from the actual situation. Therefore, the conclusion drawn from it is also problematic, so the model needs to be further refined. Before refining, the data model is converted into a chart for more intuitive display, as shown in Figure 2:

Figure 2

What aspects of the mathematical model under the previous ideal situation need to be refined? Let’s consider this in several steps:

1. In the absence of external competition, regarding the frequency of car use, is it really possible to have everyone use it at the same frequency?

Obviously that is impossible. If we exclude the influence of other factors, it is not difficult to understand that the lower the charges, the stronger the user's willingness to use the car, that is, the higher the frequency of use. The relationship between the user's car usage cost and car usage frequency is roughly shown in Figure 3:

Figure 3

Based on the analysis of changes in vehicle usage frequency, the following data can be roughly obtained, as shown in Figure 4:

Figure 4

As can be seen from Figure 4, considering the change in frequency of car use, the average charge per half hour has dropped significantly (reaching 10%), but compared with the charge of 0.5 yuan per half hour, the data is still quite good. One thing that needs to be explained is that the number of users and frequency of car use in Figure 4 are only proportional values, not specific estimates. This ratio value is closely related to factors such as socio-economic level, user psychology, and competitive environment. Since the competitive environment has not yet been considered, a more detailed explanation is provided later.

2. What will be the conclusion when considering competition?

This is a long and complicated competitive process, so I will not make a specific mathematical model analysis. Here is a simple analysis:

1. Users with a car usage cost of 0.1 to 0.5 yuan per half hour can settle down and are likely to be active users, but their proportion is extremely small and there is basically no profit margin (it is very likely to be a loss-making business!);
2. Among users whose car usage cost is 0.6 to 1.0 yuan per half hour, a small portion will be converted into users with a cost of 0.1 to 0.5 yuan per half hour, and the result is the same as the first point; the vast majority of the rest will gradually become inactive, low-frequency, dormant, and zombie users over time, and will eventually be poached by competitors (0.5 yuan per half hour).
3. Through the first two points and combined with the above analysis, the feasibility, trend and outcome of the 10 cent car strategy are self-evident. What needs to be emphasized is that the 10 cents bike-using strategy does have an impact on Xiaoming Bike (personal opinion: the impact is relatively large. With the intervention of Alipay and WeChat Pay in the future, shared bikes can be used at will without a deposit, not just Xiaoming Bike. It will have a huge impact on the deposit pool.), but it is not the same as Xiaoming Bike (there are many other factors that affect the development of Xiaoming Bike).

The following is a brief analysis of the impact of socioeconomic level, user psychology, and competitive environment on the 10-cent car strategy:

1. Socioeconomic level : Let me give you two examples. First, comparing China and India, it is obvious that the 10 cent car strategy will have a greater impact in India. Judging from the economic level alone, the shortcomings of this strategy will be more obvious and exposed more quickly; secondly, comparing China's urban and rural areas, it is not difficult to understand that the shortcomings of this strategy will be more prominent and exposed more quickly in rural areas (just speaking of the facts).
2. User psychology : mainly refers to consumption concepts, user customs, psychological state, cultural cultivation and other aspects. This is closely related to the region and it is not convenient to give examples.
3. Competitive environment : There is a very important competitive factor in the above analysis, that is, there is an external competitor that charges 0.5 yuan per half hour. When this factor changes, the analysis conditions are destroyed, and the results will naturally change accordingly. The original shortcomings may turn into advantages, and the original strengths may turn into disadvantages. Recently, the entry of Alibaba and Tencent has broken the original competitive balance. Using a bike without a deposit is a big shock to the companies related to shared bicycles. After using Alipay and WePay to ride a bike several times, I began to think about returning the deposit and uninstalling the app. If this competitive environment does not change, deposit refunds, app uninstalls, user loss, etc. are likely to be major evolution trends. This means that it won’t be long before the huge deposit pools of major shared bicycle companies will disappear. Under the circumstance of operating with little profit or even at a loss, they will eventually be unable to escape the fate of being merged and acquired. It is not a good thing for ordinary users either. Obviously, the next step after the merger and acquisition is the increase in car use costs. In the future, it is likely that reservations will also be charged.

Actually, the article should basically end here, but I would like to digress a little.

First topic

A small questionnaire survey on the operation mode of mutual promotion among friends:

• Number of respondents: 20
• Question 1: Are there any good shared bicycles? Can you recommend two?
• Question 2: Do you know Xiaoming Bicycle?

The statistical results are shown in Figure 5:

Figure 5

Although the questionnaire design is relatively simple and the sample size is small, it still has certain reference value. It can be seen from the data in the table that the mutual promotion operation method among friends is far less effective than imagined.

Second topic: About Xiaoming Bicycle - Temporary parking function

The following is Xiaoming Bicycle's description of the end of riding and temporary parking functions:

After using the vehicle, you need to lock it. When locking it, a pop-up window will prompt two options: "Temporary Parking" or "End Ride" (If you need to temporarily park and lock the vehicle during the ride, you can choose to click "Temporary Parking". After the vehicle is locked, other users cannot unlock it and ride it. Please note that the time during this process is also included in the billing range. To unlock it, click "Temporary Parking" again to ride again; if the vehicle has been used, you can choose "End Ride" to lock it. At this time, the interface will display the total amount, mileage, time, calories consumed and other data of the journey)

Compared with Mobike and ofo, the temporary parking function in Xiaoming Bicycle seems to be an improvement and optimization of the end-of-ride function. In fact, there is not much improvement in user experience . Sometimes, for users who are used to locking the bike being equivalent to ending the ride (such as using Alipay or WeChat payment for riding), the experience is even worse, which can easily cause unnecessary losses to users.

So can this function be done better? The answer is yes. By the way, my basic view on demand optimization is that, generally speaking, I do not approve of optimization plans that sacrifice the experience of the original functions, especially those that are closely related to the user's interests.

The following is an optimization solution for reference only:

1. When the bike is locked, the charge is calculated, but the payment is not made. If the user does not click "Temporary Parking" on the APP within 3 minutes, the charge will be calculated based on the charge when the bike was locked after 3 minutes, which is equivalent to ending the ride.
2. The charge will be made when the bike is locked, but no settlement will be made. If another user uses the bike within 3 minutes (it should not affect other people's use) or the user chooses another Xiaoming bike, the charge for the previous user will be settled based on the charge when the bike was locked when the next user starts using the bike. (These two points can actually be combined. When the bike is locked, the charge is calculated, but no settlement is made. If the user does not click "temporary parking" on the APP within 3 minutes, or another user uses the bike, or the user chooses another Xiaoming bike, the charge will be settled after 3 minutes based on the charge when the bike was locked, which is equivalent to ending the ride.)
3. When the vehicle is locked, charges will be made but no settlement will be made. If within 3 minutes, the user clicks "temporary parking" on the APP (if "temporary parking" is used, other users cannot unlock and ride the vehicle after it is locked), the charges when the vehicle was locked will be invalidated and the charges will continue from the first use of the vehicle.

The above is all the content of this time. If there are any inappropriate points, you are welcome to comment. I hope Xiaoming Bicycle will get better and better.

Mobile application product promotion service: APP promotion service Qinggua Media advertising

The author of this article @阿良 is compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting! Site Map