Although the current topic is less than ChatGPT, the most popular MarTech tool in recent two years is RPA. Today, let's take a look: what is the best employee RPA in the boss's mind, which is used by capital darling and more than 70% of the world's top 500 enterprises? What are the applications in marketing and operation?
In our daily work, there are a lot of repetitive work, such as:
You are responsible for e-commerce operation, and you need to spend 2 hours every day logging into multiple stores to collate daily data;
You are responsible for after-sales treatment and need to spend 3 hours every day to refund qualified after-sales orders;
You are responsible for video uploading. You need to publish the same video to different media platforms every time;
You are responsible for KOL launch, and you need to spend one hour every day to collect the broadcast and interaction data of all the content launched;
You are responsible for the live broadcast of e-commerce, and you need to spend one hour every day copying and sorting the live broadcast data of competing products;
......
These works have obvious commonalities: simple operation, clear rules, and a lot of repetition. For manual workers, this kind of work is time-consuming, labor-intensive and error prone, but RPA can perfectly solve this problem.
The full name of RPA is Robotic Process Automation. It is a group of technologies that mainly use software to execute business processes, and execute the same processes according to human execution rules and operating processes. It is affectionately called “ Digital employees ” 。
Robot: it can imitate human behavior on software, such as clicking, opening applications, entering, page sliding, etc
Process: It means that the imitated behavior is a series of specific steps in order
Automation: It means that the steps and tasks can be automatically completed without manual interference after setting the steps
The definition seems a bit abstract, but it is not mysterious: for example, macro commands in excel, game plug-in key sprites, multi play boxes, etc. all belong to the broad concept scope of RPA. Even in IOS mobile phone system, there is a very similar function, called shortcut command, which can also achieve similar functions. For example, suppose we want to capture the data of the accounts listed in the new WeChat daily list on a daily basis, how can we use RPA?
The picture is from the official website of the new list
Open RPA software according to the operation steps: open the new list website - click WeChat hot list - copy data - switch date - copy data, complete the process setting by dragging, and click Operation to automatically obtain and save data on excel.
The picture is from the screenshot of Shadow Knife RPA system
The advantages and disadvantages of RPA are clear:
The advantage is that it can help you solve some problems that cannot be solved by common systems. Because there are many systems, you have to open the computer every day, double-click the system icon on the desktop, and then click to download the chart. This is the "last mile" problem of the system. RPA is a robot assistant to solve the last mile problem ;
The disadvantage is that he cannot solve complex, non operational and non customized things.
The characteristics of RPA determine the scope of application
The biggest difference between RPA and other software is that it realizes cross system and cross platform connection in the form of non-invasive and weak coupling. Speaking of people, because it simulates human behavior, RPA almost doesn't need you to make secondary development or changes to the currently used systems, such as ERP, CRM, third-party platforms, etc., to achieve automated task execution. It can be seen that RPA is very friendly to non-technical development and routine colleagues. RPA can be used to improve efficiency as long as the business process meets the following requirements: repeatable, strong rules, stable process, and general quantity.
So at present, we mainly see that the industries or scenarios with RPA are similar. Including finance, manufacturing, medicine and retail e-commerce, these industries are already very mature and have fixed processes. They already have many fixed systems, but the employees who use the system have complex and repeated needs. The business system cannot provide process solutions for people at different levels. At this time, RPA is of great value.
Picture from Yiou Think Tank
Application of RPA in Marketing and Operation
But it does not mean that RPA is only valuable in financial and personnel management. In fact, for the daily work of operation and marketing, the application scenarios of RPA have been widely used in various industries.
1. Retail e-commerce industry
The first is a large number of repetitive retail e-commerce. In an industry where every human resource is extremely valuable, saving one human resource in any link is a cost saving. RPA has been widely used in retail, especially in e-commerce, including:
To solve the time-consuming problem of online multi-channel stores, RPA can automatically log in to the background to upload pictures, fill in the commodity properties, and finally manually check and confirm all online channels such as Tmall, Tiao Yin, and small programs. Some brands can finally save 75% of the time when goods are put on and off the shelves through PRA.
Brands need to control commodity prices in multiple channels to avoid confusion in the price system. Therefore, there is a job in e-commerce called shop tour, which is to collect the delivered price of goods of various specifications on the commodity settlement page by simulating the order placement to check whether the price is broken. For brands with a large number of SKUs, it is unrealistic to do it manually every day. It can only be done once a long period or only monitor the main commodities. But with RPA, we can realize fast and automatic SKU price inspection. For some emerging dairy products, RPA has been used to check the price of 3000+sku on 6 platforms in one day, and 6% of the products are below the price.
Timely interception of problematic orders can effectively reduce cargo damage and logistics costs. Usually, operators need to manually refresh the order to be intercepted in the background query before conducting logistics interception. It takes about 4-8 minutes to process such an order, and it is difficult to cope with the traffic surge scenario. For example, a kitchen appliance brand successfully intercepted 20000+orders and reduced 300w+cargo damage during the promotion period of the 11th National Congress of the Communist Party of China by using the interception application developed by RPA.
RPA can be used to monitor the price of competitors and grasp their latest commodity price information in real time. As a reference for the pricing of commodity activities in your shop, ensure your optimal pricing.
2. Operation post in other industries
Similar to e-commerce, RPA can actually be used to improve efficiency for various operating departments involved in a lot of repetitive work.
Similar to the new scenes on the product, there are also multiple media platforms for uploading videos/articles. This work is very simple as long as login - copy - paste - upload, but it is also very boring and needs to be repeated every time for each platform. RPA can easily solve this problem.
One of the most widely used scenarios of RPA is data collection. Although the collection efficiency is lower than that of crawler software, there is no need to worry about being intercepted by anti crawler software because it simulates human click behavior. Furthermore, the collected data can also be automatically generated into reports for visualization. Know that a domestic cosmetics brand has used RPA to realize automatic collection and processing of a large amount of data and automatic data warehousing.
RPA can also be easily competent for the simple and repeated operation of searching mobile phone numbers to add users as enterprise and WeChat friends.
RPA can realize the automatic marketing function of MA part - sending promotion information in batches. For example, a FMCG brand used RPA to set up a mass distribution application for Taobao stores, and mass chat to push coupons and trial products in batches during the daily promotion activities.
RPA+expands application scenarios
The core technology threshold of RPA software is not high, and the incoming manufacturers have gone from volume price volume service to volume product itself; More extension of product operation and product functions including RPA.
Picture from Yiou Think Tank
The first is to roll up the product operation - process reengineering, that is, how to make RPA easier to use 。 Although RPA claims that it can complete the design of automated processes by dragging and dropping icons, it is actually not easy for non-technical personnel (otherwise, RPA manufacturers will not have so many self-developed courses to teach users how to use their own software).
RPA+machine learning can automate process design 。 Just deploy the RPA platform to computers, mobile phones and other terminals, and the system will capture the user's behavior track and business data in a silent state. Employees only need to operate according to their daily working methods. With the help of machine learning technology, the RPA system will identify automated business processes from a large number of behavior tracks.
RPA+ChatGPT can realize automated process of dialogue generation 。 Recently, the domestic RPA manufacturer has released the latest product introduction (testing phase). The system can automatically generate the function of automatic process by publishing requirements in the chat box. The final effect is unknown, but it is a good direction.
The second is to roll up the scope of use of the product RPA+AI, making the scope of application of RPA wider 。
RPA+OCR can realize the extraction of paper data and open up online and offline processes 。 It is most widely used in financial and automatic audit scenarios. For example, with the help of the OCR system, the offline data will be online to realize the automatic flow of business processes.
RPA+intelligent customer service really solves user problems 。 In the after-sales scenario, ordinary intelligent customer service can only answer users' questions, but problems such as return and replacement after-sales still need manual intervention, and RPA can really solve users' problems. If the user has after-sales demand after purchasing the TV, the intelligent customer service confirms that the master is required to come to the door for maintenance through the set questions. At this time, the RPA system will automatically arrange personnel to come to the door and call the corresponding parts from the nearest warehouse.
RPA+NLP/knowledge map improves business understanding And make RPA become half an expert from an assembly line worker. For example, RPA added to the knowledge map can automatically correct the content of the listing annual report and identify the content that violates the laws/policies.
What RPA manufacturers are there in China
Due to the low threshold of RPA core technology and wide application scenarios, there are many RPA players in China, as shown in the following figure:
The picture is from the first new sound
Typical representatives of each module:
Original RPA manufacturers: that is, pure RPA players focus on the RPA track at the beginning of their creation. For example, the representative Shadow Knife RPA is an automated office manufacturer focusing on the e-commerce industry;
Technology giant incubators: For example, Alibaba Cloud RPA used to be called "code stack", which was originally developed to meet internal needs, and began to provide external services as demand increased;
Old brand manufacturers in T0B software industry: such as UFIDA, Kingdee, etc;
AI/NLP and other technology field entry manufacturers: BTL Technology started from the AI field and then entered the RPA track;
Firms in the field of financial technology: banks are the main application field of RPA, and also go from procurement to independent research and development to commercialization.
Any tool has a threshold for use. No matter how many advantages it has, RPA is no exception. At present, if you really want to make good use of RPA, you still need to invest a lot of time in learning, such as the most common and simple data acquisition. In actual operation, you will still encounter various small problems, which ultimately lead to the inability of the entire process to be executed.
The biggest attraction of RPA to enterprises is to release human efficiency - to save money. If the business scale is not large enough, the cost of purchasing the system plus the cost of training staff is much higher than the cost of directly deploying staff, and the enterprise will not consider accessing RPA.
In the long run, is it the best solution to automate all processes for RPA? Its potential harm is to solidify the wrong process and ignore the optimization process itself. This can be achieved not only by a few mouse clicks.
Author: Song Xing's View of Numbers
This article is authorized to be released and does not represent the position of APP dry goods stores. If it is reproduced, please contact the original author.
The original link is as follows: https://mp.weixin.qq.com/s/0TMl1fYMzmORb_VUf4ic1Q