Preface

The text and pictures of the article are from the Internet, only for learning and communication, and do not have any commercial use. The copyright belongs to the original author. If you have any questions, please contact us in time for handling.

Author: Zhu Xiaowu's play data

PS: if you need Python learning materials, you can click the link below to get them by yourself

http://note.youdao.com/noteshare?id=3054cce4add8a909e784ad934f956cef

Some time ago, president Luo and President Wang became their own start-up companies and became dishonest people. Little five plans to go to IT orange to see their companies.

IT oranges were found out of a database of dead companies by accident, and the "dead" data of famous companies between 2000 and 2019 were counted.

Xiaowu uses python to crawl down the data of dead companies to observe the death history of startups in the past decade.

get data

F12, Network view asynchronous request XHR, turn page.

The url that returns json format data was found successfully,

https://www.itjuzi.com/api/closure?com_prov=&fund_status=&sort=&page=1

 

Readers who don't know JSON can see [python playing with JSON data], part of the crawler code

def main():
    data = pd.DataFrame(columns=['com_name','born','close','live_time','total_money','cat_name','com_prov','closure_type'])
    for i in range(1,2): #Setup crawl N page
        url= 'https://www.itjuzi.com/api/closure?com_prov=&fund_status=&sort=&page='+ str(i)
        html = requests.get(url=url, headers=headers).content
        doc = json.loads(html.decode('utf-8'))['data']['info']
        for j in range(10): #10 dead companies per page
            data = data.append({'com_name':doc[j]['com_name'],'born':doc[j]['born'],'cat_name':doc[j]['cat_name'],
                    'closure_type':doc[j]['closure_type'],'close':doc[j]['com_change_close_date'],'com_prov':doc[j]['com_prov'],
                    'live_time':doc[j]['live_time'],'total_money':doc[j]['total_money']},ignore_index=True)
            time.sleep(random.random())
    return data

 

Data of 6271 dead companies were obtained successfully.

Count 10 years of life and death

As of November 24, 2019, there are nearly 6271 companies marked as "closed" in the IT orange database. We select 5765 companies in the last ten years (2010-2019) to see the demise of startups in this decade.

It is often said that 1998 is the first year of China's Internet, and 2010 is the first year of mobile Internet.

No wonder the mobile Internet in 2010 is so busy. Wechat, Xiaomi, meituan, iqiyi, etc. were established in succession in this year.

Baidu became the biggest beneficiary after Google exited China, Taobao became a new growth point for Alibaba, and Tencent announced that QQ had more than 100 million people online at the same time. Since then, Baidu, Alibaba and Tencent have officially become the "three giants" - BAT. In addition, Netease's online games, Sina's Weibo, Sohu's video and input methods have also begun to attack, and the competition for mobile Internet has officially begun.

Since 2010, the trend of the number of companies born and died over the years is shown in the figure below.

2013 and 2014 are the birth tides of the company. Three years later, they correspond to a wave of death tides in 2016 and 2017. In 2017, more than 2000 companies closed down.

In this decade, many "tuyeres" have ups and downs. Online car hailing, group buying, live broadcasting, gene testing, bike sharing, short video, bitcoin, VR|AR, unmanned shelf, artificial intelligence, live delivery

At every tuyere, there are hundreds of "pigs" standing, trying to take advantage of them.

In these famous battlefields, such as the battle of hundred regiments, the battle of vertical e-commerce, the battle of take out, the battle of taking a taxi, and the battle of cycling, there are so many kinds of bridges that the people who eat melons can not see. There is a fight between the eldest and the second, and the third is gone; there is a merger between the second and the third, and they continue to compete with the eldest; there is also a merger between the first and the second, which leaves other families far behind

There are also live answers like the Chongding conference. When they are held, all the families fight against each other fiercely, but they don't make it to the lunar new year since the 2018 lunar new year.

Behind the demise of tuyere is the money burned by countless start-ups. At the beginning, every company believed that it could burn down its rivals, but burned itself out, but could not get financing any more.

There are many reasons for the demise of start-up companies. Apart from the core factor of fierce industry competition, the most important one is the lack of business model. The entrepreneur's lack of understanding, thinking and preparation on how to maintain stability and how to make profits is not enough to support his enthusiasm at the beginning of the tuyere after the industry is stable.

In addition, the "false air outlet" and "false demand" have also fascinated many startups. Sharing bicycles and chargers derived from the "sharing economy" are popular, but it's not necessary to share phones, toilet paper and basketball.

Interestingly, the database of the dead company also adds a list of the most popular companies. The first one is indeed the famous "fast broadcast".

It's true that some companies are dead. He's still alive;

Some companies are alive. He's dead.

Code

 1 import requests
 2 import json
 3 import pandas as pd
 4 import time
 5 import random
 6 from fake_useragent import UserAgent
 7 ua = UserAgent()
 8 headers = {'User-Agent':ua.random} #Disguise request head
 9 ​
10 def main():
11     data = pd.DataFrame(columns=['com_name','born','close','live_time','total_money','cat_name','com_prov','closure_type'])
12     for i in range(1,2): #Setup crawl N page
13         url= 'https://www.itjuzi.com/api/closure?com_prov=&fund_status=&sort=&page='+ str(i)
14         html = requests.get(url=url, headers=headers).content
15         doc = json.loads(html.decode('utf-8'))['data']['info']
16         for j in range(10): #10 dead companies per page
17             data = data.append({'com_name':doc[j]['com_name'],'born':doc[j]['born'],'cat_name':doc[j]['cat_name'],
18                     'closure_type':doc[j]['closure_type'],'close':doc[j]['com_change_close_date'],'com_prov':doc[j]['com_prov'],
19                     'live_time':doc[j]['live_time'],'total_money':doc[j]['total_money']},ignore_index=True)
20             time.sleep(random.random())
21     return data
22 ​
23 if __name__ == "__main__":
24     final_result = main()
25     #final_result.to_csv("final_result.csv", index_label="index_label",encoding='utf-8-sig')

 

​