Three growth models and implementation principles of Internet finance!

There are many words that we blurt out in our daily lives, but we actually haven’t thought about their true meaning. Most arguments and wrong decisions start from unclear and inconsistent definitions.

Internet finance operations are the jobs that are closest to money, so a clear definition of user behavior is particularly important.

User behavior is composed of the five simplest elements: time (when), place (where), person (who), interaction (how), and content of interaction (what). To analyze user behavior, it is necessary to define it into various events. For example, user search is an event, which includes the time, platform, ID, search result and search content. This is a complete event and a definition of user behavior. With such events, we can connect user behaviors and observe them.

User behavior analysis refers to the statistics and analysis of relevant data after obtaining basic data on website visits, so as to discover the patterns of users' website visits and combine these patterns with online marketing strategies, so as to discover possible problems in current online marketing activities and provide a basis for further revising or reformulating online marketing strategies.

Specifically, user behavior analysis is based on the user's behavior on Internet products, as well as the time frequency of the people behind the behavior, etc., to deeply restore the user's usage scenarios and guide business growth.

A complete, multi-dimensional and accurate user portrait = user behavior data + user attribute data.

Consumer behavior in a narrow sense refers only to consumers' purchasing behavior and actual consumption of consumer materials. In a broad sense: the various actions taken by consumers to obtain, use, and dispose of consumer goods, as well as the decision-making process that precedes and determines these actions, and even a series of complex processes including the acquisition of consumer income. Consumer behavior is dynamic, involving the interaction of perception, cognition, behavior and environmental factors, as well as the transaction process.

Consumer behavior model

With the rise and development of the Internet, the model for conducting consumer research has evolved from the initial AIDMA model (Attention, Interest, Desire, Memory, Action) to the later AISAS model:

In the linear direction, there are multiple key nodes on the core path from "attention" to "sharing". Due to the "persuasive psychology slide" effect (explained below), users will be lost for various reasons in the process from the starting node to the final node, thus forming a user conversion funnel.

For Internet finance operations, it is necessary to combine the company's current goals and its own KPIs, grasp the key touchpoints of the conversion funnel, and design corresponding operation strategies accordingly.

To analyze user behavior, we need to define it as various events, and aggregate the time (when), place (where), person (who), interaction (how), and content (what) of the interaction to form a complete user task. From the perspective of task levels, from core to periphery, they can be divided into three levels: core tasks, extended tasks, and extended tasks.

For Internet finance operators, it is necessary to go deep into the system of the company's products, understand the key paths and main processes of user operations, and design operational activities based on the importance of user tasks. This will avoid the situation of brutal conflicts between operations and products to the greatest extent. Before reading on, you can stop and think about this: for investment users, what are their core tasks, extended tasks, and extended tasks?

Since different users are at different stages of the conversion funnel, there are group differences in the stages of their life cycle development on this platform. Introduction period, growth period, maturity period, dormancy period and churn period: for users at different development stages, the differences in operation strategies are often huge.

At this point, we have initially built the basic framework of our user growth strategy :

A complete P2P user portrait = user attribute data + user behavior data + transaction data + risk-return data. Internet companies are good at the first two areas, but they often focus on general Internet user analysis and lack understanding of finance and investment; financial institutions are stronger in the second two areas, and generally believe that user behavior data is only procedural information and are unwilling to collect it. It is possible that an APP has been launched for many years without any basic tracking or conversion rate analysis.

In the entire user portrait system, user behavior is the key element that connects the other three data on both the user and the platform, and it is worthy of in-depth exploration and improvement.

By analyzing the types of pages users visit and the depth of their access paths, we can help identify users’ preferences for certain investment products or operating activities, and then push more information about such products to such users, or coupons related to related products (interest rate coupons/discount coupons/full-reduction coupons, etc.). If the platform is larger, we can use it to improve the front-end architecture design to suit each individual need.

Previously, user behavior analysis agencies such as GrowingIO and Zhuge IO have launched such products. Although the accuracy needs to be further improved, it is a beneficial attempt after all. Judging from the feedback from actual use, the analysis effect of preference analysis on investment and financial management apps is better than its application in lending products.

From user introduction to user loss , the entire chain follows up to improve conversion rate and retention rate .

According to the formula:

From this, we can see that, with the average order value and gross profit margin unchanged, there are two points we can start with:

1. Increase investment frequency: Continue to optimize the operational strategy of user conversion nodes, so that users have sufficient motivation to continue to move to the next step in the conversion process of "registration-real name-card binding-transaction-reinvestment", thereby increasing the number of transactions and transaction amounts, and ultimately improving the platform user life cycle value.
2. Reduce churn rate: Improve user loyalty by releasing various tasks to guide users to continuously upgrade their levels; improve user retention through stimulating activity and recall strategies, and ultimately reduce the platform's user churn rate .

Developing an operational strategy is not as simple as copying competitors’ products and implementing the boss’s requirements. Or to put it another way, how to better copy the essence of competitors' products and make the boss's requirements a plannable/executable/reportable level all depends on the collection and analysis of user behavior data. Talking about operations without considering user behavior is tantamount to being a hooligan.

Review the effects of version upgrades and operational activities based on user behavior data, and make adjustments and optimizations accordingly. Regardless of whether the goal of the activity is to increase daily active users, increase GMV or single product transaction volume, it must ultimately be implemented on the user's conversion funnel node or user task.

When reviewing the effectiveness of an activity, a simple look is nothing more than “achieved” or “not achieved”, but analysis of user behavior data can answer why, what went well/badly, and how to do better next time.

Various conversion rates based on the conversion funnel, new user acquisition -activation- retention analysis based on pirate indicators, registration-investment-withdrawal data analysis based on the user task system, etc., can all help us build a framework for building the indicator system of this platform and comparative analysis of competing products, which is of significant help in formulating operational indicators, applying for resources, and reviewing results.

The construction of a growth system based on user behavior can be divided into four parts: preconditions, execution strategies, channel building, and implementation supporting principles. The following will expand on each of these.

As mentioned above, a complete Internet finance user portrait system is composed of the following four parts of data: attribute data, behavior data, transaction data, and risk-return data.

Internet financial data analysis system

• Attribute data: the most basic data of users as natural persons and social persons, and also the basis of the other three types of data
• Behavioral data: the key element that connects the user and other data on both ends of the platform, and the starting point of all operational strategies. Based on user behavior data and combined with the platform's tag system, we can also obtain derived user conversion data and user behavior preference data, which will not be elaborated in detail here.
• Transaction data: the basis for calculating platform revenue, ROI, LTV and other operating indicators, and also an important criterion for judging user value
• Risk-return data: User investment attribute data is not only the basis for differentiated operations, but also the embodiment of the platform's implementation of risk control and compliance requirements.

Based on the above data, combined with frequency, time dimension, number of users and other indicators, more indicator data can be obtained, such as:

• Time period (month) + number of investments = monthly active users (MAU)
• Last payment date + most recent investment date after payment = user churn
• Investment amount within the time period / Number of investing users within the time period = Average investment amount per person

If the user's investment behavior is combined with the most recent transaction time, transaction frequency, and transaction amount per unit time (based on the RFM model ), it will be able to provide strong support for user value judgment, recall strategy formulation, etc. after calculation and analysis.

• R (Recency): The time interval between user investments
• F (Frequency): The number of times a user invests per unit time
• M (Monetary): The amount of money invested by the user during the time period

Regarding the risk-return data, let’s expand on it a little:

For most P2P platforms, product display and recommendation are primarily based on marketing purposes, without much consideration of the user's own risk tolerance. Sometimes, even after users complete the risk assessment, they are induced to purchase products that exceed their risk tolerance, so you will see the following picture:

XX Finance provides investment advice after users complete risk assessment

(Picture from the official account @Smart Investment Advisory Alliance)

According to the direction of regulatory development in the past two years, the implementation of the suitability principle of "selling the right products to the right investors" will gradually spread from licensed financial institutions to Internet financial companies. Designing products and operational strategies based on users' risk tolerance and profit goals can, on the one hand, improve compliance levels and platform safety margins; on the other hand, it can also provide a further grasp of users' investment needs and investment capabilities. This part of the content has also been covered in previous articles. You can click "Touching the Heart's Operational Strategy - Redefining Internet Financial Users Part 1" to view it.

The matching relationship between user risk tolerance and product risk level - based on user risk tolerance (5-level classification)

The relationship between user risk tolerance and user investment goals - based on user risk tolerance (3 levels)

In 2016, Ant Financial published the " Ant Jubao Public Investor Big Data Analysis", which revealed five items that can be respectively corresponded to the four categories of data mentioned above:

Note:

1. In the actual process of building a data indicator system, the indicators will be broken down into more detailed parts. Since we are not writing a PRD here, we will not elaborate on the report fields corresponding to these four types of data.
2. The "transaction data" here mainly refers to the position data after the user has made an investment behavior; the investment behavior data such as the time, amount, number of products, etc. related to the user's first investment and reinvestment are included in the category of "behavior data".

The growth strategy based on user behavior relies on the establishment of three basic models: conversion funnel model , life cycle model and task hierarchy model.

in:

• The conversion funnel model analyzes the nodes of user conversion vertically. Based on this, the user's life cycle on the platform is defined and divided according to the distribution of users at different conversion nodes.
• The task hierarchical model breaks down and groups various user behaviors on the platform horizontally, dividing them into "core tasks-extended tasks-external tasks" system, and on this basis guides users to migrate and grow in tasks at different levels.
• Ultimately, by continuously optimizing user conversion rates and continuously guiding user task completion behaviors, we can achieve differentiated operations and services for users in each life cycle of the platform, and ultimately achieve rapid and sustained growth in platform users.

Regarding the conversion funnel model, Dao Shiwu has already described it in detail in previous articles, so I will not elaborate on it here.

Conversion Funnel Model (Financial Management)

Based on the above model, it is relatively easy to grasp the key conversion nodes and corresponding important indicators of financial management users. But in actual application, this only reaches the passing score of 60 points. So, how can we use the conversion funnel model in a more advanced way? Here are two directions you can refer to:

1. User branch path conversion funnel (taking communication/invitation as an example)

Conversion Funnel Model-Branch Path Conversion

Taking the "propagation" node of the conversion funnel model as the core, we can break down the process from "old users seeing the invitation prompt on the interface" to "new users accepting the invitation and completing the registration and account opening", with at least 7 conversion nodes in between. When broken down to this granularity, more operations tasks begin to emerge:

Ultimately, judging from the results, refined funnel division can, on the one hand, help operations improve efficiency and input-output ratio, and on the other hand, it is also conducive to quickly locating and solving problems during the activity. When reviewing the event after it is over, the conclusions drawn will be more reliable as to whether it was done well and how to do better next time.

2. Direction 2: Parallel conversion funnel

Conversion Funnel Model-Parallel Conversion

Note: In the above figure, for users, investment, content, daily interest rate hikes, communication and other businesses are all defined as "parallel businesses".

If we divide it according to different granularities and different business lines, we can actually break down many parallel conversion funnels within an APP. For operations, it is meaningless to break it down just for the sake of breaking it down. We need to sort out several parallel funnels that need to be focused on based on the current operational priorities and department KPIs (the number in the same period generally does not exceed 3-4, and you will not be able to take care of more than that). Generally speaking, there are several directions:

• Grasp the main user conversion process: registration/card binding/ordinary investment/fixed investment
• Grasp user active conversion: push/content/daily interest rate increase/sign-in/old user re-engagement
• Grab new users and convert them into new ones: invite friends/new users to invest for the first time

For users who enter the platform in the same period, we can use cohort analysis and other methods to analyze the usage and conversion of these users in major parallel businesses within a period of time after entering the platform (if it is a public fund, the shortest subscription and redemption time can be set to one week), and further understand the degree of overlap of users in various businesses.

Based on the above analysis, through page guidance, activity incentives and other methods, combined with the platform's user growth system, we can facilitate the transition of users between parallel businesses.

The following are some examples of parallel businesses related to "investment":

Example: Lufax

Taking Lufax’s seal system as an example, by designing the user growth path, it guides users to transition between different businesses, continuously conduct transactions of various products, and participate in various platform activities, ultimately improving user activity and retention rate.

The user life cycle is usually divided into five stages: introduction, growth, maturity, dormancy, and churn.

By refining the characteristics of each period, these five time periods can be divided into three operating intervals:

• Customer acquisition zone - the introduction period, usually referred to as "attracting customers", the main operating means is to attract new customers, and the main evaluation indicator is the retention rate;
• Appreciation zone - growth stage + mature stage, usually referred to as "receiving customers", the main operation stage is activation, and the main evaluation indicators are growth rate and conversion rate;
• Retention zone - dormant period + churn period, usually referred to as "customer retention", the main means of operation is retention, and the main evaluation indicators are retention rate and recall rate.

User life cycle model

The first problem that novices encounter is often that they don't know what criteria to use to divide the life cycle nodes. In fact, in the actual operation process, you don’t really need to make a user life cycle distribution chart and then operate accordingly. In fact, which stage of the cycle a user is in is determined by his or her behavior. Therefore, the key to operation also lies in user behavior.

For example, you may find that the subscription amount for products on the platform has barely increased recently, or has even decreased. So, you ask your BI colleagues to pull data and find that the proportion of registered but non-trading users on the platform has been on the rise in the past month, and the user reinvestment rate has also declined. Combining the user conversion funnel model above, you can draw preliminary conclusions:

• The transaction conversion rate of registered users is relatively low. On the one hand, we can conduct internal investigation to see if there are any problems with the transaction process at the system level . On the other hand, we can push novice gift packages (novice red envelopes + experience money + high-yield novice labels) to users through SMS/PUSH/in-site messages/APP homepage pop-ups to guide users to complete their first investment conversion. Here, the operation is actually to improve the transaction conversion rate for this group of introduction users.
• The reasons for the decline in user reinvestment rate may be complex and varied. You can observe it for a few more days and compare it with historical baseline data. Sometimes it is possible that the user has no money to invest because payday has not arrived yet, or there was a big promotion a while ago and a large number of users bought products with longer terms and have not yet received the money, so there is no new money for reinvestment.

When users are at different stages of development, user value will change accordingly. Therefore, corresponding operational goals and strategies need to be designed for different stages. The relevant content has been described in detail in previous articles and will not be expanded here.

Mutual Finance User Task Hierarchy System (Financial Management Side)

For users on the financial management side, all behaviors on a platform can be assessed and analyzed within the framework of "core tasks-extended tasks-external tasks". This is the "user task hierarchy model" that Dao Shiwu will introduce in this section.

After careful observation, you will find a very interesting phenomenon. "Model I - User Conversion Funnel Model" is actually a platform-centered user conversion perspective, while "Model III - User Task Hierarchy Model" is a user-centered demand satisfaction perspective. Both models have the same transformation nodes, but Model I is flat while Model III is weighted.

1. Task hierarchical structure:

Note: The reason for placing "Becoming a Platform Investor" in the extended task area is that for users, the task of "Becoming a Platform Investor" is actually a prerequisite for the task of "Making Money". It is indeed very important, but it is not the first issue that users care about, so the weight of this task is moved to the extended task.

2. Market opportunities in the user task hierarchical model:

• Core mission : Provide opportunities for hundreds of millions of users. In the past, financial institutions such as banks, securities firms , insurance companies and funds in the financial system were responsible for meeting the core mission requirements of Internet finance, and later it was Ant Financial, which launched Yu'e Bao. You will find that this type of company provides basic services of "save money - make money - withdraw money" to a large number of users. They don't have much special features in user operations , but they choose the right time and task model, and have the highest probability of success.

The ability of these companies to cope with reshuffle risks is "strong"

• Expansion mission : Provide opportunities for a user base of tens of millions. Those that meet the requirements of extended tasks often show improvements in efficiency. It includes two types of companies: one is Internet giants such as Tencent and JD.com , which start from the expansion task of becoming a platform user (commonly known as "user import"), and are essentially engaged in traffic business; the other is P2P players such as Ping An, Jiufu, and Paipaidai, whose models are: ① Ping An Lufax: through strong financial product integration and supply capabilities to complete the "investment-reinvestment" task ② Jiufu/Wukong Finance and Paipaidai: seize the P2P time window, complete the "obtain income-increase income" task, and simplify and lower the entry threshold to complete the "becoming a platform investor" task

The ability of these companies to cope with reshuffle risks is "medium"

• Extension mission : Provide opportunities to reach millions of users. Most of the companies that are characterized by meeting the needs of extended tasks are leading Internet-based P2P companies. They cannot find a breakthrough in their core tasks, so they focus on expanding their tasks and making great efforts in external tasks, and they do a great job in customer acquisition methods, operational methods, subsidy intensity, differentiated asset acquisition and packaging.

The ability of such companies to cope with reshuffle risks is "weak"

PS: If you overdo the external business and go astray, the ultimate outcome is pyramid selling and Ponzi schemes. Many people are obsessed with the so-called user fission method. If it is separated from the core tasks of users, it is definitely not a healthy user growth model - it can neither bring in legal income nor sustain growth in scale.

3. Task hierarchical relationship:

• Core tasks : The first thing users care about is making money, and the main process of making money is "investing capital (deposit) → obtaining income (value-added) → taking profits (withdrawal)". Whether the platform is safe, whether the platform benefits are attractive, whether the money can be withdrawn in time when needed, and whether the main line operation process is convenient are the issues that users are most concerned about in the core task area. They directly determine whether "extended tasks" and "external tasks" exist.
• Extension tasks : Among the extension tasks, the three first-level tasks of "reinvestment", "increase income", and "early profit-taking" correspond to the three second-level tasks of "investing principal", "obtaining income", and "taking profits" in the core tasks. The former is a further expansion and optimization of the latter.
• Extended tasks : Extended tasks are attached to core tasks and extended tasks. When operational activities mainly fall in the extended task area, the needs can basically be met at the operations manager level. If the operational activities fall into the extended tasks or core task areas, they often require more in-depth cooperation from the product manager . Especially for operational activities in the core mission areas (experience money, additional income, etc.), it is often necessary to coordinate multiple departments horizontally, and these activities are often the most likely to fall into traps.

4. Task hierarchical role coordination:

• The core task is the main area of ​​concern for product managers. The core function design and user experience design of APP usually fall into this area.
• External tasks are often the focus of operations managers. Various operations activities such as attracting new customers, promoting activation, and retaining existing customers fall into this area.
• The extended task area is the boundary between product and operation. For product managers, extended tasks are the direction of continuous optimization of user experience. For operation managers, the subtasks in extended tasks are good carriers of operation activities.

If you observe carefully, you will find that a senior product manager or operations manager is often able to think in the direction of "core task → extended task → extended task", and at the same time can fully base themselves on their respective business goals and KPIs, and then design plans and arrange priorities.

On the other hand, sometimes when operations colleagues put forward an operations requirement to product colleagues, they will be surprised: "Hey, why did the product colleague get so angry on the spot?" In fact, if you understand the user task hierarchy model, you will know that the requirement you put forward this time is likely to make the product colleague feel that the user's core task process is disturbed. Based on the insight into user task stratification, operations can have a more reasonable grasp of the weights of various functions in the product, and have a common basis for discussion when communicating with product and development.

The following takes the "Task Center" of JD Finance APP as an example to disassemble and analyze the relevant tasks:

JD Finance’s Task System Analysis

From the above figure we can see:

• The most critical point of Internet finance is transaction, so the operational incentives superimposed on core tasks and extended tasks are often relatively large.
• Inviting friends to use JD Finance APP can indirectly increase transaction volume, so a moderate level of incentive is given.
• Playing the gold coin game every day is a simple activity to promote user activity, which is far from the core task and does not meet the core needs of users. Therefore, it is given a relatively weak operational incentive.

However, it should be noted that different types of companies have different focuses on their business goals when they are at different stages of development. Therefore, the design of operational activity plans and supporting incentive measures will have their own personalized aspects. The model cannot be applied directly, and specific analysis is required for specific issues.

Mutual Finance User Growth Framework

For Internet finance products, user conversion and growth are achieved at two levels:

• Financial aspects

The increase in user investment amounts and the complexity of investment products represents an increase in user risk tolerance and return targets.

• Internet Level

The improvement of user maturity within the platform is manifested in the continuous migration and growth to the next link of the funnel based on the growth system of the main conversion funnel.

From a financial perspective, the user's family development stage in the financial life cycle is a background factor that affects the user's financial growth; at the same time, the user's risk tolerance, return goals, income level, investment experience, etc. are all key factors that affect the user's financial growth. However, since the financial life cycle involves a large amount of personalized offline data and is too closely related to investment, protection, and asset allocation, we will not go into detail in this article and will discuss it in another article in the future.

From the Internet level, the node of the user's development life cycle on the platform is a background factor that affects the user's growth at the Internet level; at the same time, the user's node in the main conversion funnel, activity status, retention status, etc. are all key factors that affect the user's growth at the Internet level.

Overall, the user's growth process is intertwined at the two levels of finance and the Internet, and is ultimately reflected in the user's various investment behaviors on the platform.

Investment growth system for financial management users

Previously, Ant Fortune (formerly "Ant Jubao") had made a very vivid stratification of financial management users, divided into 7 levels from the most basic bank deposits (kindergarten) to the most advanced asset allocation (sixth grade). According to this standard, we will find that the current situation of most fund companies is to teach third and fourth grade classes to first graders; and a number of domestic smart investment advisory companies are to teach sixth grade classes to kindergarten children. These situations actually ignore the stratification of Internet finance users and the process of user growth, which is reflected in the number of users and management fee income, and the return effect will naturally not be very good.

For Internet finance platforms, they need to help and guide users to achieve growth and progress based on their own product resources and user stratification and in combination with corresponding operational strategies. In this regard, I have always felt that JD Finance's "Xiaobai Fund" has done a good job (I haven't seen the transaction data, and JD's friends are welcome to add to it^_^):

JD Finance-Xiaobai Fund

When users click to enter "Newbie Fund", they can see "Daily Earn", "Monthly Earn" and "Expert Zone" from left to right, which correspond to money market funds, bond funds and hybrid/stock funds respectively (there was also "Weekly Earn" based on short-term financial management funds), helping users to outline the growth path of "First Year (Money Market Fund) → Second Year (Bond Fund) → Fourth Year (Hybrid Fund)". Users can also obtain income rewards ranging from 3% to 4% by learning about fund product knowledge.

For novice users, money market funds and bond funds are within the risk range they can bear and they can obtain additional income subsidies, so they will naturally be motivated to participate in the investment and growth process.

In fact, for most financial management apps, if you do the following two things, this article will be worth reading:

• Divide user growth and advancement, and provide graded products and services
• First, we should serve the lower grade students well, and focus on user experience and operation strategy, supplemented by investor guidance and education.

Financial management end user transaction behavior growth system

From the perspective of the composition of the growth system of user transaction behavior on the financial management side, it mainly includes the following elements:

• In the vertical direction, based on the user conversion funnel model, set the conversion path and conversion goals, from "complete registration" to "complete new investment"
• Horizontally, at the main nodes of the conversion funnel, users are encouraged to convert and grow from the previous node to the next node. From "downloaded but not registered" to "not invested N days after remittance & no balance in the account", each has its own conversion goal.
• In terms of supporting conditions, according to the characteristics of users in the corresponding life cycle, set up a complete set of operation means, namely " trigger conversion conditions (such as within M days after download) - trigger conversion methods (such as homepage mask) - trigger conversion incentives (such as tiered cash coupons) ", to ensure the implementation of user conversion and growth goals.
• Key points to emphasize: At the conversion nodes that facilitate users to "complete their first investment" and silent/lost users to "complete new investment", manual telephone follow-up can be appropriately introduced, with the main content being "platform confidence building + inquiries and answers to reasons for non-conversion + preferential incentives". As long as the cost is controllable, the manual method will make users feel warmer and the conversion effect is generally good.

Below is Lufax’s “January Return Gift” campaign, which targets users who have not invested for N days and have balance in their accounts. The main purpose of the campaign is to encourage dormant users to “complete new investments” by sending investment voucher text messages.

LuX Institute-January Return Gift

From the above figure we can see:

• There is a risk in the first step of user contact. The sent text message may be disabled by the security software on the user's mobile phone, causing the user to not see the message. A relatively safe approach is to use the method of "SMS + email/push/phone call" to ensure that users can definitely see it.
• The wording of the text message can easily make people mistakenly think that it is a scam message. The user has clearly not done anything, so why would he "get the qualification to draw a lottery" out of thin air?
• On the basis of being able to accurately send text messages to dormant users, the best way is to ensure that all users who receive the text messages can win the prize, otherwise it will affect the user experience. The guiding ideology of this type of activity should be [being certain of getting something, but not sure how much]: being certain of getting something can ensure that users have the motivation to participate; being uncertain of how much can provide users with the fun of "taking a gamble" . However, this is also a problem for large companies, including many banks. Due to the large scale of users, in order to control marketing costs, they have to resort to the magic weapon of "limited quantity, first come first served".
• Try not to be too restrictive in the products that apply. There must be some reason why users enter a dormant state, so the incentives given to these users must be more attractive. It is best to give them a universal coupon with no threshold, or at least put it on star products or some newly launched products.

In short, the conversion and growth of users at the Internet level is actually a relatively complex process, which includes both the conversion of users in the main process and the conversion of users in the branch process. At every point of conversion, we must consider clearly the user's interests and risks/difficulties, and prepare products and operational methods in advance, ultimately promoting the continuous conversion and growth of users.

At the end of this section, let’s take a look at another case study – Orange Finance’s novice task growth task. As mentioned above, the user's growth process is intertwined at the two levels of finance and the Internet, and the two are interdependent and mutually reinforcing.

Breakdown of Orange Finance's newbie tasks and growth tasks

From the above figure we can see:

• From the perspective of growth at the Internet level, the newbie task of Orange Finance is to focus on guiding users to complete the main process transformation of " channel introduction → registration → first investment → dissemination". After completing such an in-depth operation process, users will definitely have a better understanding and trust of the platform (investing money & investing in connections), so the cost of users leaving the platform becomes higher
• From the perspective of financial growth, Zhou Zhou Sheng is a short-term high-yield product (redeemable after 7 days, with an annualized maximum of 8.39%), and the term of Ai Ding Cun ranges from 1 month (annualized yield of 5.5%) to 12 months (annualized yield of 8.29%). In order to pursue high returns, users tend to prefer products with longer investment terms.

After buying a fixed-term product, only VIP users have the privilege of early redemption, while ordinary users cannot redeem in advance (to become a VIP user, the investment amount on the platform must reach a certain scale), which further improves the user retention rate on the platform.

• From the perspective of incentive tendencies, it can be seen that there are obvious differences between Juzi Finance and JD Finance. In the above example "JD Financial Task System Dismantling", the "Invite friends to use the JD Financial APP" is given a medium-level incentive of 50 gold coins, and in Orange Financial Management, it is the highest intensity incentive of 400 oranges. Because JD Finance is backed by JD.com, with large users and high business complexity, it focuses on guiding users to convert and be active in various business systems; Orange Financial is an entrepreneurial platform with relatively low user penetration rate, so it focuses on expanding the scale of users and encompassing more users before seeking conversion.

When we usually do competitive product research and reference methods, such background factors are often ignored, and the result of directly copying other people's methods is often "70% depends on luck and three% depends on financial resources". This situation should be avoided as much as possible.

In summary, in the process of continuously completing tasks and gaining growth, users have gained more rights and psychological satisfaction, and the platform has also gained user activity and loyalty. Here we can see that a good operation strategy design can enable both users and the platform to achieve a win-win situation. The construction of the user growth system is interspersed with various routines of "interest", "honor", "emotion" and "security", and the peeking into human weaknesses is vividly reflected here.

On the basis of fully grasping user behavior, in order to maximize the effectiveness of the user growth model, we also need to have supporting guiding principles and measures:

• Everything starts from reality. Any method or model must be used based on the company's industry attributes and current development stage. This is a basic principle. The deliciousness of his is my arsenic, especially the methods of his friends cannot be copied simply and roughly. The most typical example is the difference in incentives for the old and new recruitment of JD Finance mentioned above, which reflects this principle.
• Use PCDA to verify and expand the effectiveness of the model. Through the continuous cycle of the planning stage (Plan)-Execution stage (Do)-Check stage (Check)-Action stage (Action), combined with operational goals and data review, and cooperate with ABtesting to form an effective operational method system suitable for this platform.
• The refinement of indicator dismantling and operational activity design. Clarify your key goals, and then continuously decompose the goals through MECE (Mutually Exclusive Collectively Exhaustive) method; finally, disassemble them to the minimum granularity and design the operation plan based on this. For example, the "user branch path conversion funnel (taking communication/invitation as an example) mentioned above, "Daoshi Wu emphasizes "indicating self-interest factors in the old user interface, and manifesting altruistic factors in the invited user interface, and ultimately maximizing the conversion rate of communication/invitation". It is reflected in the Latte Smart Investing activity page below: for old users, it is highlighted that "each friend gives you 50 yuan"; for new users, it is highlighted that "up to 5% of the incentive bonus" is highlighted.

Latte Smart Investment-old new cases

• Xiaomi plus rifles are sometimes more effective than long gun cannons. To put it completely, this sentence should be "The millet and rifle (what you hold today) are sometimes more effective than the long gun cannon (who may have one day in the future)." If the platform has not been established for a long time, the data system is not complete, and the complete user life cycle cannot be obtained. At this time, if you want to quickly improve the transaction conversion rate, you can directly find data classmates and pull up the user data of "opening an account but not trading" and "first investment has not been re-invested" in the past month, send a batch of novice gift packages to new users and push it with text messages and follow up the phone number, and send a batch of (high-yield novices logo + targeted cash coupons) to the first investor users to see the effect directly. In the grassroots stage, everything is in trouble. At this time, simple and rough are more effective than following the steps
• Finally, remember to calculate ROI clearly—in the industry’s low point, money still has to be used on the blade

At this point, the whole article is coming to an end.

Daoshiwu sorted out the entire set of underlying thinking frameworks behind the growth logic based on user behavior. Because of Elon Musk , the "first principle" has been very popular in the past two years. I also tried to start from the lowest needs and behavior patterns of Internet finance users and make a brief deduction of the complete set of logic, see the figure below:

The underlying framework based on user behavior growth logic

The direct goal of users participating in financial business is to obtain profits, and ultimately to consume. According to the statement of "Financial Science", "a basic creed of finance is that the ultimate function of the financial system is to meet people's consumption preferences, including all basic necessities such as food, clothing and shelter." To be honest, it means making money is to spend money better. At this point, platforms like Alibaba and JD have perfectly realized the closed loop of users' "investment-consumption".

The process of users obtaining profits is the process of investing funds into the financial management platform and transferring funds after achieving the profit target. Users' operations on any financial management platform can eventually be abstracted into the behavioral system of "invest funds → obtain returns → transfer funds".

Among all the disciplines at present, Daoshiwu believes that economics is the most reasonable framework and tool to explain user needs and behaviors. According to Mr. Zhang Wuchang's view, economic science can ultimately be summarized into three most basic axioms: the law of demand, the concept of cost and the meaning of competition. These three points are actually the starting point of all operational strategies.

• The law of demand : The core idea is "the role of price on supply and demand is certain". Here, "price" can be the rate of return of financial products given by the financial management platform, or it can be the user experience of the Internet, subsidies and incentives provided by operations. Which platform will the user's needs be directed to, whether to stay or leave after a certain platform, are affected by the "price" provided by the platform.
• Cost : For users, from downloading the APP to investing and sharing, every node of the entire conversion funnel, whether it is choosing YES or NO, it means cost. In Internet finance operations, these four types of costs need to be considered: sunk costs, opportunity costs, marginal costs and accounting costs, which are the basis for users to make behavioral decisions. In fact, the operation strategy of many platforms is to continuously increase the user's sunk costs (money, time and emotion) and continuously increase the expenditure of user's psychological accounts , so that users can stay because the cost of leaving is too high.
• Competition : According to Baidu Encyclopedia , competition is the psychological needs and behavioral activities that individuals or groups try to overcome or overwhelm the other party. That is, the behavior of each participant in sacrificing the interests of others and obtaining personal interests to the greatest extent, with the purpose of pursuing attractive goals. Competition is an adversarial behavior that each party of an individual or group strives to surpass the other. If the same operation activity is introduced (such as asking friends to help bargain, inviting friends to get rebates, voting for babies, etc.), if the ranking mechanism is introduced, it will often greatly increase the user's willingness to participate and the time for continuous participation.

Persuading Psychology Slide

According to the theory of persuasion psychology, the following four elements need to be considered for the persuasion and behavior of users:

• The first is gravity, which represents the user's initial motivation to do something. For Internet finance users, it is to earn profits through investment
• From an angle, it is the motivation and needs that operations dig out from users, and exists based on gravity. For example, the initial motivation of users is to make money by investing. Operations design various activities to design activities such as "invite friends to add income" and "learning financial management knowledge to send red envelopes" to cultivate users' motivation to invite friends to join the platform and the motivation to learn financial management knowledge.
• Promotion is the incentive provided by operations to users, with the purpose of guiding users to complete specific behaviors and promoting continuous transformation and growth of users.
• Friction is the resistance of users to complete specific behaviors on the platform: sometimes it is objective, such as the APP is unstable, or the success rate of card binding is low, or the platform product yield is low; sometimes it is subjective, such as the main color of the platform's UI is green (this was the first few versions of XX Investment), which makes investing users feel uncomfortable, or the platform name is difficult to pronounce, etc., for different reasons.

In the persuasion psychology slide model, the "gravity" representing the user's initial motivation corresponds to the "demand" of the three axioms of economics. The "friction" representing the user's completion of specific behavioral resistance is also corresponding to the "cost" of the three axioms of economics. They are the embodiment of the economics axioms in operational strategies.

Foger's behavioral model

The BJ Fogg's behavior model believes that to facilitate a user's behavior, the following three elements are required:

• Motivation, according to the definition of Baidu Encyclopedia, is the internal psychological process or internal motivation of individual activities guided, stimulated and maintained by a goal or object, and is the basis of most human behavior. In organizational behavior, motivation mainly refers to the psychological process that stimulates people's behavior. Through inspiration and encouragement, people can generate an internal driving force to move towards the desired goal. It can be clearly seen here that the user's motivation is the internal driving force for behavior, a psychological mechanism with high user autonomy, and it often takes effect through "stimulation and encouragement". In Internet finance business, the lowest motivation for users is to get profits, and the direction of efforts of each platform is to allow users to invest on their own platform, continue to retain, and bring more users to invest.
• Ability is the quality of a user to complete a certain operation, or the level of completing a certain behavior. In Internet finance business, the user's ability to act is generally reflected in whether he owns a mobile phone, whether he has a few seconds of operation time, or whether he has a certain amount of investment funds, and the threshold is very low
• Triggers, here refer to the incentives provided by operations to users to facilitate users to complete certain behaviors.

In the Foger behavior model, there is a correspondence between the "gravity" (initial motivation) and "angle" (motivation excavated from the user) of the persuasion psychology slide model. "ability" corresponds to "friction" (the resistance of the user to complete a specific behavior on the platform), and the "trigger" corresponds to "push" (the motivation provided by operations to users) - this means that every element of the persuasion psychology slide is ultimately reflected in the corresponding operational key points in the Foger behavior model. The Foger behavior model is the basic framework of all operational strategies.

The above article is based on the January return ceremony of the Institute of Lu X as an example. Let’s take a look at the practical application of the Foger behavior model:

• Motivation: The initial motivation of users is to earn profits through investment. The motivation mined by operations is to earn additional subsidies and rewards provided by the platform.
• Capability: Users only need to own a smartphone, and almost all Internet users can meet this condition
• Trigger: The incentive provided by LuX operates here is to get a lottery opportunity, and users have the opportunity to get investment coupons ranging from 5 yuan to 50 yuan.

When users complete specific behaviors according to the path set by operations, the user's transformation and growth path has taken a new step forward.

Finally, the core guiding ideology of the three articles "The Touching Operation Strategy (1-3)" is summarized in three useless sentences:

The above is the only way to grow Internet finance users.