User behavior can be summarized by 5W2H:
Who (who), what (what behavior), when (when), where (where), why (for what purpose), how (by what means), how much (how long and how much).
User behavior analysis is to find out the laws of users using products through the statistics and analysis of these data, and combine these laws with the marketing strategy, product function and operation strategy of the website to find out the possible problems in marketing, products and operation. Solving these problems can optimize the user experience, realize more accurate operation and marketing, and make products grow better.
Second, why do you need user behavior analysis?
In the era of PC internet, the annual growth rate of netizens reached 50%, and you can get a lot of traffic by building a website casually; In the early days of the mobile Internet, APP also experienced a wave of traffic dividends, and the cost of acquiring a customer was less than 1 yuan; In recent years, with the decline of traffic growth dividend, the competition is becoming more and more fierce. There are hundreds of competitors in every field, the cost of obtaining customers has soared to an unbearable level, and business growth has become slower or even retrogressive.
Photo: The competition in the Internet industry is becoming more and more fierce.
In such a high-cost, high-competitive environment, if enterprises can't use data analysis to do refined operations, they will have a huge waste of resources, which will inevitably lead to high operating costs and lack of competitiveness. For the Internet platform, the traditional data analysis mainly focuses on the data of the results, but lacks the analysis of the user's behavior process that produces the results, so the value of data analysis is relatively limited, which is why many enterprises feel that they have done sufficient data analysis in recent years, but it has not had much effect.
By analyzing the 5W2H of user behavior, we can know where the user comes from, what operations he has performed, why he lost, where he lost, and so on. Thereby improving the user experience and platform conversion rate, and allowing enterprises to achieve business growth with refined operations.
Third, how to collect user behavior data?
User behavior analysis is so important, why do Internet companies seldom do it well? The main reasons are incomplete data collection and imperfect analysis model.
1. How to collect user behavior data efficiently?
The traditional data analysis is too extensive, and the application value of the analysis results is low, because the data is not fine enough and the analysis model is not perfect. In order to do a good job of analysis, we must first have rich data, so we should start with data collection. Traditional user behavior data collection methods are inefficient. For example, when we get some user behavior data, we need to add monitoring codes to the corresponding buttons, links or pages to find out how many people clicked this button and clicked this page. This method is called "burying points". Burying points requires a lot of manpower and energy, and the process is complex, which leads to high manpower and material costs.
In the era of mobile Internet, embedding points have become a more painful thing, because each embedding point needs to be published to the app store, and the review cycle of Apple App Store is even more difficult, which greatly reduces the timeliness of data acquisition. Because data analysis is an extremely important link in business development, even if the cost of manpower and material resources is too high, this work cannot be saved.
Therefore, we also see that there are some excellent user behavior analysis tools at home and abroad, which realize the function of collecting data without buried points. For example, there is Mixpanel in foreign countries, and domestic geeks can collect data without burying points on the WEB, H5, Android and iOS. Through the collection without buried points, the perfection and timeliness of data can be greatly enhanced.
2. How to accurately collect user behavior data?
For some core business data, we want to ensure the accuracy of 100%, so we can supplement it by burying points in the back end, so that we can experience the efficiency and convenience brought by no burying points and ensure the accuracy of core business data. Digital geek supports data integration in four ways: no buried point, pre-buried point, post-buried point and digital geek BI importing data.
Fourth, how to do a good job in user behavior analysis?
First of all, we should make clear the business objectives, deeply understand the business processes, find out the key data nodes that need to be monitored according to the objectives, do a good job in collecting and sorting out the basic data, and have enough data and scientific models to support the analysis results more effectively.
The previous generation of user behavior analysis tools (more accurately, website statistics or APP statistics), the main function is still limited to the analysis of browsing behavior, rather than the analysis of users' deep interaction behavior, so the analysis value is relatively limited. At present, the impression of most Internet practitioners on user behavior analysis is still at this stage.
I think to do a good job in user behavior analysis, we should master the following analysis models:
1. User behavior is tracked all the way, supporting AARRR model.
Dave McClure, an investor in 500 Startups, put forward an analysis model to analyze the "piracy indicators" obtained by users at different stages, which has been widely used in Silicon Valley.
AARRR is the abbreviation of five words: acquisition, activation, retention, income and reference, which correspond to five important links in the user life cycle. First of all, we should analyze user behavior based on the user's complete life cycle.
1). Get users
In marketing promotion, what channel brings the highest traffic and what is the ROI of the channel? The conversion rate of different advertising content is the data analyzed in this step.
Source channel is the first step to acquire customers. Combining automatic system identification with customized channels, the retention and transformation effects of various source channels are analyzed. The website access sources, App download channels and search keywords of various search engines can be easily statistically analyzed through the data analysis platform. Through multi-dimensional analysis of UTM promotion parameters, cross-analysis of promotion channels, event names, display media, advertising content, keywords, landing pages, identification of high-quality channels and inferior channels, fine tracking, and improvement of channel ROI.
Through the channel quality model, formulate corresponding customer acquisition and promotion strategies:
Figure: Channel quality model
The channel shown in the above figure is an example, and the channel quality will also change dynamically. The first quadrant channel has high quality and large flow, so we should continue to maintain the strategy and intensity of channel delivery; The quality of the second quadrant channel is relatively high, but the flow rate is relatively small. It is necessary to increase the channel delivery and continuously pay attention to the changes in channel quality; The quality of the third quadrant channel is poor and the flow is small, so we should carefully adjust and gradually optimize this channel; The channel quality in the fourth quadrant is poor, but the traffic is large. We should analyze the channel data to make more accurate delivery and improve the channel quality.
2). Activate the user
Activating users is the most critical first step to achieve business goals. If a large number of users use your products every day, but there is no strong connection between users and you, you will not be able to carry out subsequent operations.
3). User retention rate
At present, the key factor of a product's success is not virus mechanism or a large amount of marketing funds, but user retention rate. It is very important to develop products that attract users back. Facebook has a reservation rule of "40–20–10". Numbers represent daily retention rate, weekly retention rate and monthly retention rate. If you want the DAU of the product to exceed 1 10,000, the daily retention rate should be greater than 40%, the weekly retention rate and the monthly retention rate should be greater than 20% and 10% respectively.
Reservation is one of the important links in AARRR model. Only by doing a good job of retention can we ensure that new users will not be lost in vain after registration. Like a basket that keeps leaking. If the cracks at the bottom are not repaired, it is difficult to achieve sustained growth by just pouring water into them.
4). Earn income
Realizing revenue is the foundation of every platform's survival, so it is very important to find a business model that suits you. According to different business models, there are different revenue methods: media platforms are realized by advertisements, games are paid by users, and e-commerce is paid by commissions or sellers. In the field of enterprise services, LTV: CAC is greater than 3, which can effectively and healthily grow.
5). Virus spread
Through the optimization analysis of the first four stages of the model, from unstable users, active users to the final loyal users, we can retain and transform customers to the greatest extent and cultivate them into loyal users of the enterprise. Through social word-of-mouth communication, enterprises can bring efficient benefits.
In today's high customer acquisition cost, socialization can bring a better user base and lower customer acquisition cost to enterprises.
2. Transformation analysis model
Conversion rate is the core of sustainable operation, so I also spent a lot of space to explain it in detail. The common tool of transformation analysis is transformation funnel, or funnel for short. New users are constantly losing in the registration process, eventually forming a funnel shape. In the process of user behavior data analysis, we not only look at the final conversion rate, but also care about the conversion rate of each step of the conversion.
1). How to build a funnel scientifically?
In the past, we will build a funnel through the experience of products and operations, but we have no confidence in whether this funnel is representative and how important it is to improve the overall conversion rate. At this time, you can understand the mainstream path of users through user traffic analysis.
Figure: User traffic analysis
User traffic analysis is intuitive, but it requires analysts to have certain experience and judgment ability. In order to solve this problem, digital geeks have developed an intelligent path analysis function, which can analyze the mainstream path of user conversion with one click only by selecting the conversion target. Reduce the efficiency of creating funnels to a few seconds.
Figure: Intelligent Transformation Analysis
2). Funnel comparative analysis method
It is not enough to use ordinary funnel for transformation analysis. It is necessary to analyze the detailed factors that affect the transformation, and segmentation is very important. For example, the conversion funnel can optimize the channels by comparing the channels of user sources and mastering the conversion differences of different channels; According to the comparison of user devices, we can understand the conversion differences of users of different devices (for example, the conversion rate of a higher-priced product from ordering to payment is significantly higher for users using iphone than for users using android).
Figure: Comparative analysis of funnels
3). Analysis method combining funnel and user flow direction.
The general conversion funnel only has the mainstream, and there is no detailed information about the inflow and outflow of each step. When we analyze the user registration conversion, if we can know where the users who have not converted to the next step have gone, we can plan the conversion path of users more effectively. For example, in the transformation path shown below, 88% of users who did not enter the second step left directly, while 65,438+00% users chose to log in directly as registered users, and only 2% users bypassed the login page and entered the homepage of the website; And 100% of the users who did not switch from the second step to the third step left. This is a typical closed landing page. You only need to optimize the conversion rate in the third step to improve the overall conversion rate.
4). Micro-transformation behavior analysis method
Many behavioral analysis products can only analyze the transformation at the functional level and the event level, but there are serious shortcomings in the analysis of user interaction details. For example, the funnel above, we found that the last step is the key to the transformation, but the last step is to register the form, so it is very important to analyze the detailed behavior of filling in the form. This behavior is called micro-change.
For example, the time to fill in the form, the loss of users who filled in the form but did not submit it, and the blank rate of the form fields.
Figure: Form Fill Transformation Funnel
Figure: Time to fill in the form
Through the above-mentioned micro-conversion analysis, the conversion rate of users from the beginning to the successful registration is 85%, while the traffic is only 8%. It can be concluded that the biggest leakage point that affects the transformation is the filling rate, so how to improve the filling rate is the core of our promotion of registration transformation. Effective content and accurate channels are the core factors that affect the filling. We have talked about the channel factors in customer acquisition analysis, which leads to the fourth tool of our micro-transformation analysis: user attention analysis.
5). User attention analysis method
User interaction with page content, such as clicking, browsing, staying time on page elements, scrolling, etc. , all represent the user's attention to the information to be displayed in the product and whether it can attract the user's attention.
Business data can be visualized, so how can behavior data be visualized? Several geeks translate the above behaviors into five kinds of heat maps, namely, split screen contact heat map, link click heat map, page click heat map, browsing heat map and paying attention to heat map. Through the cross analysis of five kinds of heat maps, users can effectively analyze the most concerned content.
Figure: attention heat map
Only by mastering the interactive behavior analysis of micro-conversion can the conversion rate be improved more effectively. And all analysis tools that can't effectively improve the platform conversion rate are wasting human and time resources of enterprises, which is also the fundamental reason why many enterprises don't benefit from user behavior analysis.
3. Refined operation mode
The previous operation can only be aimed at all users, if we want to do accurate operation behavior for some target customers.
Figure: User group portrait
For example, if we want to accurately market registered users who use iphone in a certain area but have not been active for three days or have not formed a transaction transformation, we need the cooperation of operators, product personnel and technicians to retrieve data and formulate operating rules, which involves a lot of manpower and time investment. The new generation of user behavior analysis can adopt user grouping, user portrait, customized user active and retention behavior, accurately locate users, and thus realize refined operation.
Figure: Creating a User Group
4. Qualitative analysis model
User experience is the top priority of enterprises. In product design, user research, R&D, operation, marketing, customer service and many other aspects, it is necessary to master the real user experience process. However, how to optimize the user experience has been controversial internally, mainly because it is difficult to describe it concretely and vividly. It is very important to optimize the experience of your product if you can reproduce the specific scene when users use your product through behavior analysis.
In the past, when I was in Taobao, the user experience department would optimize the experience by inviting users to the company for interviews and doing usability experiments. But this method needs a lot of time and expense, and the sample may not be representative. In order to solve this problem, several geeks have developed a user behavior recording tool, which saves the cost without inviting users to the company to record on site, and restores the real operation of users intuitively and efficiently in the form of video, so that all positions in the enterprise can grasp the first-hand information of user experience and help product development to enhance user experience.
Figure: User behavior recording and playing interface
Summary: Analyze the whole life cycle of users through AAARRR model; Improve the product conversion rate through the conversion rate analysis model; Improve operational efficiency through refined operation; Optimize user experience through qualitative analysis; If the above four aspects are done well, business growth can be realized through user behavior analysis.
5. What is the future direction of user behavior analysis?
Many people ask me, there are already several companies that do user behavior analysis, why do you want to start a geek? I think the goal of data analysis is to optimize the operation efficiency by applying the analysis results, but the main analysis tools at home and abroad only stay at the analysis level, and there is still a lot of room for efficient application. Therefore, mathematicians should make new breakthroughs in application besides being more professional and effective in analysis. The problems reflected by the data analysis results are mainly divided into two categories: operation (including marketing) and products. Therefore, it is necessary to provide targeted solutions to these two types of problems.
1. Operation automation
As we said before, the user behavior analysis system can achieve refined operation, but the specific application needs to manually formulate the operation and marketing strategy, which can only be applied through products and research and development, and when the strategy changes, it needs to re-develop the corresponding tools, which also takes up a lot of time and affects the efficiency of operation and marketing. Several geeks have developed membership marketing systems and automated operation tools. Operators and marketers make rules directly. The system automatically pushes accurate activity information to qualified users according to the rules, which directly improves the work efficiency of operators. Operators can shift their focus to planning instead of wasting too much time in repeated execution. Automated operation can save a lot of operating costs for enterprises.
Figure: Creating an Automated Action Rule
2. Scientific decision-making of products and operations (marketing)
User behavior data analysis is often carried out after the behavior occurs, while products and operations are made through experience. Once the decision is wrong, it will lead to irreparable results. Therefore, if we can pass the user's shunt A/B test before going online, verify the product and operation scheme in a small range, and select the best scheme for release, the scientific decision-making can be greatly improved.
Google optimizes its products and operations by running tens of thousands of A/B tests every year, bringing the company $654.38+00 billion in revenue.
The method of A/B testing is very effective, but it is not widely used by domestic Internet companies, which is mainly related to the complexity of applying A/B testing.
Several geeks have complete A/B testing tools. Business people can use visual test editing tools on websites and applications to create and run tests. By automatically interpreting the test report, the threshold of A/B testing is greatly lowered.
Figure: Website-side visual editing test tool
3. Analysis automation
User behavior analysis is professional, so we should not only master different analysis methods, but also be familiar with the business, so as to give valuable analysis results in combination with the business. If you can directly load the SDK like 360 security guards, you can automatically diagnose and analyze and give a solution. This is the direction of future data analysis. Several geeks have also made positive attempts in this regard and achieved initial results. At present, they have the functions of automatic data warning and automatic reporting.
User behavior analysis is a science. Being good at obtaining data, analyzing data and applying data is the basic skill for everyone to do a good job. Every enterprise should strengthen the application of user behavior analysis big data, find out the rules from the data, and use data to drive enterprise growth.
Digital geek is a new generation of user behavior analysis platform in China and an essential big data analysis tool for growing hackers. It supports APP data analysis and web analysis, and has created six original conversion rate analysis models. It is the first data analysis product in the field of user behavior analysis using quantitative analysis and qualitative analysis methods. Based on the user behavior analysis system, it provides two intelligent data application solutions, namely, member marketing system and A/B testing tool, so that enterprises can quickly realize data-driven growth.
This article was originally written by digital geek CEO Xie. Welcome to reprint. Please keep the full text and author information.