6 conversion rate analysis models to improve product conversion!

1. Why is conversion rate becoming more and more important?

1.1 The traffic dividend is gradually disappearing and the competition is becoming increasingly fierce

The difficulty of developing new platforms and new applications is dozens of times more difficult than it was ten years ago. There are three main reasons:

1. Declining growth rates;
2. Traffic is concentrated on a few large platforms such as BAT ;
3. Competition among peers is fierce.

According to the CNNIC statistical report, the Internet user growth rate was 23.4% in 2006 and 6.1% in 2015, and it has been declining continuously. While incremental traffic is decreasing, traffic concentration is significantly increasing, and large Internet platforms with BAT as the core account for 80% of the traffic in the industry. Every time a wave of Internet applications comes, there are thousands of similar platforms competing across the country. After 3-5 years of competition, only the top three platforms remain. You can imagine how fierce it is.

1.2 The significance of conversion rate to enterprises, product managers and operators

The conversion rate is the proportion of users who complete the conversion goal among the total users. Registration, ordering, login, and repeat purchase can all be quantified using the conversion rate. The conversion rate is the cornerstone of the growth of Internet platforms. High-growth platforms are basically also high-conversion platforms, and the growth level affects the company's market share and value. For employees working in the company, high-growth companies will also promote personal growth faster and obtain richer returns.

2. How to improve conversion rate

Improving conversion rate requires the participation of all employees. Executives, middle-level managers, and grassroots managers in the company must master the corresponding methods. We can jointly discuss how to improve conversion rate and achieve business growth through three levels, namely: seizing the opportunity, understanding the way, and optimizing the techniques. These six words come from the Tao Te Ching written by Lao Tzu and were named by Mr. Li Ka-shing as the motto of Cheung Kong Graduate School of Business:

• Seizing the opportunity + understanding the path = strategy
• Skill = Organizational Ability
• Business success = strategy * organizational ability = (taking advantage of opportunities + understanding the way) * good techniques

To seize the opportunity is foresight; to understand the way is true knowledge; to master the skills is actual results. Seizing the opportunity is the most important, understanding the truth is the second, and improving the skills is the third. Below we will explain the specific methods in detail from these three levels:

2.1 Methodology for improving conversion rate: Taking advantage of the situation

"Momentum" is often invisible, but it determines the direction; "taking advantage of momentum" is about going with the flow, and going with the flow will achieve twice the result with half the effort. In other words, what Lei Jun said is to become a pig on the wind; the fish pond theory, opportunity well theory, and PMF in the following methodology help us "take advantage of the situation."

(1) Fishpond Theory

Market size is one of the most important indicators that determine the size of a company. Many investment institutions only invest in companies with a market size of more than 100 billion yuan. So how do you choose a target market that suits you?

Market opportunities can generally be divided into the following three categories:

1. Everyone can see it and everyone can reach it
   For example: food recipe websites, pet owner social networking websites, event gathering websites, etc. The threshold for such opportunities is low and the competition is fierce.
2. Everyone can see it, but most people can’t reach it.
   For example: sending rockets to Mars, drugs to cure cancer, autonomous driving, etc. These opportunities have high barriers to entry and require sufficient preparation.
3. Basically, people can't see it, and can't reach it.
   For example: nursing robots, visual search engines , lifelong batteries, etc. These opportunities are relatively long-term, most people cannot see them, and they are currently difficult to achieve.

Jack Ma, founder of Alibaba Group, once said: The arrival of any opportunity will inevitably go through four stages: "invisible", "look down on", "incomprehensible", and "too late". Therefore, when choosing a target market, "invisible" and "despised" opportunities are easier to enter, and other people "cannot understand" the process of accumulation. After the accumulation is completed, it is "too late" for everyone. For example: the Internet and mobile Internet in the past, and the recently popular blockchain have all gone through a stage where most people "cannot understand". If you can "understand" them through in-depth research, you can also find the target market and customer base that suits you.

(2) Opportunity Well Theory

The target market and customer base you choose may be large enough, but there is another important factor that determines the overall value of the business opportunity: the depth of the value you provide.

Opportunity value = X (number of people affected) * Y (depth of value provided)

If your target customer base is small but the value you provide is deep enough, the overall value will be greatly improved. As a latecomer, don’t deliberately pursue a large market. Instead, focus on providing a core value and the most profound value.

(3) PMF

How to measure whether your product has found PMF?

"The life cycle of a startup can be divided into two parts: before finding product-market fit, and after finding product-market fit" - Marc Anderson (a famous serial entrepreneur , venture capitalist, and Silicon Valley guru who has read countless projects). Product -market fit (PMF) refers to whether your product has enough value, but this is not determined by you. The key lies in whether it is recognized by the market and customers. "99% of Internet companies fail simply because they fail to find the right product." We can evaluate whether a product has found PMF through these questions and user data indicators.

Determine through questions or questionnaires:

• Would you recommend our products to your friends?
• How disappointed would you be if you could no longer use this product?
• How many users are leaving your product, and how fast?

Judging by user data standards, user-level product standards:

• Use more than 3 days per week
• New daily active users (DAU) exceeded 100
• 30% of new users retained on the next day
• Reach 100,000 users

SaaS Product Standards:

• 5% paid conversion rate
• LTV/CAC>3, that is, user lifetime value/user acquisition cost>3
• Monthly churn rate <2%
• Monthly sales volume reaches 100,000
• Payback time for user acquisition costs < 12 months

How to motivate users, attract users and quickly achieve PMF?

Simon Sinek said in his book " Star With Why" that "people don't buy what you do, they buy why you do it. You just need to do what you do to prove your beliefs." They identify with your "WHY", but in fact they are also doing it to satisfy their own beliefs. Apple's iPhone sells well not only because they made a phone that is easier to use and better looking, but also because people agree with their philosophy of challenging the status quo of existing products and "Think Different".

By thinking about WHY, you can give your product a personalized attribute and form a natural filter for users, which can resonate with the user's spirit and allow users to identify with your product or service from the bottom of their hearts, occupy a high position in their minds, and establish lasting user loyalty .

2.2 Methodology for improving conversion rate: Mingdao

Take advantage of the situation and be realistic, understand the truth and seek the truth, and only by combining the realistic and the virtual can you get things done. Tao means rules. Those who understand the way determine the ideas and seek the path. User decision-making models, behavioral motivation models, and North Star indicators can help us "see the way" in the pursuit of improving conversion rates.

(1) User conversion model

The AARRR conversion funnel is highly consistent with the consumer decision-making model in marketing : customer acquisition corresponds to user needs, and we understand how users find us; activation corresponds to the process of users collecting information, during which users complete their first activation; retention corresponds to the evaluation plan, and users will stay for a good experience; revenue corresponds to the decision to purchase, and when the user completes the purchase, the platform also earns revenue; if the user has a continued good experience after the purchase, it will lead to the re-dissemination of the product.

(2) Fogg's Behavior Model

So how do users complete the above-mentioned activation, retention, and transaction conversion behaviors? We need to understand what are the reasons behind the behavior? The behavioral model (B=MAT) proposed by Fogg of Stanford University shows that three elements must converge at the same time for behavior to occur: motivation, ability and trigger. When the behavior does not occur, at least one of these three elements is missing. When motivation is high, the behavioral difficulty people can accept is also relatively high. When motivation is low, the behavioral difficulty is correspondingly low. Therefore, when designing a trigger mechanism for a product, it is necessary to consider the influence of these factors. A good product has a low enough behavioral threshold.

Summary: The consumer decision-making model in marketing helps us to clearly define the goals for each stage; the motivational behavior model (FBM) guides us to continuously improve our motivation; and the product realization funnel (AARRR) helps us know what to do at each step. They are not isolated, and their ultimate goal is to make users buy. The three models combined are why, how, and what.

(3) North Star Indicator

① AHa moment

AHa moment is also called epiphany moment. In Internet products, AHa moment refers to the key conversion behavior that affects user retention . The AHa moment of Facebook, Twitter and other companies is as follows:

These quantitative indicators cannot perfectly represent all of your users. They simply represent the majority of users. After a certain number of user behaviors, the moment when they truly realize the value of the product, some users enter the state quickly, while others enter slowly. (Just like some people are slow to warm up while others are quick to ignite), the AHa moment should be a specific, quantifiable behavior/experience that occurs relatively early in the conversion funnel.

So, how do you quantify the key conversion behaviors of your product?

Take the Internet financial management platform as an example: maximizing the intersection of behavior and retention is the key conversion behavior.

② North Star Indicator

“North Star Metric”, also known as “ OMTM ”, is one metric that matters. It is called the North Star Indicator because once this indicator is established, it will shine high in the sky like the North Star, guiding all employees of the company to move in the same direction. The North Star indicator is an output indicator and a hysteresis indicator. It can only represent one dimension of the business and is not responsible for the mutual sacrifice of other indicators.

How to choose a North Star Metric:

• What kind of indicator is everyone in the company working towards?
• All strategic planning is aimed at improving that indicator?
• This metric must be related to the value you provide
• Does everyone understand the meaning of this indicator?
• Don't give up or change this indicator easily, you must stick to it for a period of time

The healthy operation of an enterprise is affected by indicators in multiple dimensions. Each dimension corresponds to a key indicator. At least each product should include the following three categories: Breadth of retention, Depth of engagement, and Monetization.

③ From AHa moment to North Star indicator

The AHa moment key conversion behavior is the input indicator of the North Star Metric. If users experience the A-HA moment, then it will definitely improve your North Star Metric. Push users to experience the A-HA moment as soon as possible and make them addicted to your product.

2.3 Methodology for Improving Conversion Rate: Optimization Technique

"Skill" is ability, which is a combination of knowledge, methods, strategies and experience. Many companies have advantages in momentum and morality, but ultimately fail at the level of skill. "Skill" is the process and strategy that can solve practical problems, and is a skill that can improve effectiveness and efficiency. Funnel analysis, micro-conversion analysis and other 6 secrets to increase conversion rate help us "optimize our skills".

(1) Fundamental analysis method

① Segment analysis

Segmentation analysis is the origin of all analysis methods. Because the information value of indicator data under a single dimension is very low, segmentation helps us solve almost all problems. For example, the conversion funnel actually breaks down the conversion process into steps. The so-called drill-down means continuously decomposing according to certain dimensions when analyzing changes in indicators. For example, by regional dimension, from regions to provinces, from provinces to cities, and from provinces and cities to districts. The so-called roll-up means the opposite. As dimensions are drilled down and rolled up, data will be continuously segmented and aggregated. In this process, we can often find the root cause of the problem.

The analysis and evaluation of traffic channels also require a lot of cross-dimensional segmentation methods. For example, if we cross-analyze the quantity and quality of channels, we can find high-quality channels. The first quadrant channel has high quality and large traffic, so the channel delivery strategy and delivery intensity should continue to be maintained; the second quadrant channel has relatively high quality but relatively small traffic. We should increase the channel investment and continue to pay attention to changes in channel quality; in the third quadrant, the channel quality is poor and the traffic is small, so we should carefully adjust and gradually optimize this channel; in the fourth quadrant, the channel quality is relatively poor, but the traffic is large, so we should analyze the channel data to make more accurate investment and improve the channel quality.

② Comparative analysis method

The comparative analysis method is to compare two interrelated indicator data, quantitatively display and explain the relative values ​​of the research object such as size, level, speed, etc., and discover and identify problems of the business at different stages by comparing indicators under the same dimension.

Comparative analysis method:

1) Comparison types include absolute number comparison and relative number comparison. The two types of data need to be combined for comparison.

2) The comparison criteria are divided into the following four categories:

• Time standard: year-on-year, month-on-month, and fixed base ratio. Through these three methods, you can analyze information such as business growth level and speed.
• Space standards: divided into three categories: comparison with similar spaces, comparison with leading spaces, and comparison with expanding spaces
• Empirical or theoretical criteria: e.g., Engel coefficient, activity level;
• Plan completion criteria: For example: KPI.

3) Principle of comparative analysis:

• The connotation and extension of the indicators are comparable
• The time frames of the indicators are comparable
• The calculation methods of indicators are comparable;
• The overall properties are comparable.

③ Cluster analysis

Cluster analysis is to identify things based on their different attributes, grouping things with similar attributes into one category, so that things in the same category have a high degree of similarity. Cluster analysis has simple and intuitive features. Application of cluster analysis in website analysis: user grouping, user tagging method; source clustering mainly includes channels, keywords , etc.; page clustering, similar/related page grouping method, for example: in page analysis, there are often pages with ? parameters, such as: information details page, product page, store page, etc., all belong to the same category of pages

3. Six secrets to improve conversion rate

3.1 Conversion rate improvement rules

Improving conversion rate is an ongoing and long-term process. At a certain stage, indicators can be improved through data analysis and optimization methods. However, after a period of time, due to various reasons such as changes in the environment and changes in user habits, some methods and measures need to be adjusted in order to maintain a high conversion rate. Therefore, we need to master the closed loop of improving conversion rate and make continuous improvements.

The process of improving conversion rate is like the process of an airplane taking off. The driving force is the user-perceived benefit (motivation) minus the perceived cost (behavior difficulty), and a good user experience (trigger mechanism) can increase the conversion rate faster. All of our analytical methods are designed to optimize this equation: improving experience and lowering barriers.

3.2 Funnel Analysis

A commonly used tool for conversion analysis is the conversion funnel, or funnel for short. New users continue to be lost during the registration process, eventually forming a funnel-like shape. In the process of analyzing user behavior data, 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 construct a funnel scientifically

In the past, we would build a funnel based on our product and operational experience, but we were unsure whether this funnel was representative or how much effect optimizing this funnel would have on improving the overall conversion rate. At this time, we can understand the mainstream path of users through user flow analysis.

Figure: User flow analysis

User flow analysis is very intuitive, but it requires analysts to have certain experience and judgment. In order to solve this problem, Shujike has developed an intelligent path analysis function. After selecting the conversion target, you can analyze the mainstream path of user conversion with one click. The efficiency of creating funnels is reduced to seconds.

Figure: Intelligent conversion analysis

(2) Funnel comparison analysis method

It is not enough to use ordinary funnels for conversion analysis. It is necessary to analyze the detailed factors that affect conversion. The ability to perform segmentation and comparative analysis is very critical. For example, by comparing conversion funnels by user source channels, we can understand the conversion differences between different channels for channel optimization; and by comparing by user device, we can understand the conversion differences among users of different devices (for example, for a higher-priced product, the conversion rate from ordering to payment is significantly higher for iPhone users than for Android users).

Figure: Funnel Comparison Analysis

(3) Combining funnel and user flow analysis

The general conversion funnel only has the main process, but no detailed information on the flow in and out of each step. When we analyze user registration conversions, if we know where the users who did not convert to the next step went, we can plan the user's conversion path more effectively. For example, in the conversion path shown in the figure below, 88% of users who did not enter the second step left directly, while 10% of registered users chose to log in directly, and only 2% of users bypassed the landing page and went to the homepage of the website; and 100% of users who did not convert from the second step to the third step left. This is a typical closed landing page, so you only need to optimize the conversion rate of the third step to improve the overall conversion rate.

3.3 Micro-conversion analysis

Many behavioral analysis products can only analyze conversions at the functional and event levels, but they are seriously lacking in the analysis of user interaction details. For example, in the funnel in the figure above, we analyzed that the last step is the key to influencing conversions, but the last step is the registration form, so the detailed behavior analysis of filling out the form is crucial. We call this behavior micro-conversion.

For example: the time spent filling out the form, which fields users who filled out but did not submit the form were lost, the blank rate of form fields, and other form filling behaviors.

Figure: Form filling conversion funnel

Figure: Form filling time

Through the micro-conversion analysis of the above form filling, the user's conversion rate from the beginning of filling to successful registration is as high as 85%, while the flow to filling is only 8%. It can be concluded that the biggest leakage point affecting conversion is the filling rate, so how to improve the filling rate is the core of our efforts to improve registration conversion. Effective content and precise channels are the core factors that influence completion. We have already discussed channel factors in customer acquisition analysis, which leads to the fourth tool of our micro-conversion analysis: user attention analysis.

3.4 Heat map analysis method

The user's interaction with the page content, such as clicking, browsing, length of time spent on page elements, scrolling, etc., all represent the user's attention to the information the product wants to display and whether it can attract the user's attention.

Business data can be visualized, but how can behavioral data be visualized? Shujike converts the above behaviors into five types of heat maps: split-screen reach rate heat map, link click map, page click map, browsing heat map, and attention heat map. Through cross-analysis of the five heat maps, we can effectively analyze the content that users are most concerned about.

Figure: Attention heat map

Only by mastering the interactive behavior analysis of micro-conversions can we improve the conversion rate more effectively. Any analysis tools that cannot effectively improve the platform conversion rate are wasting the company's human and time resources, which is also the fundamental reason why many companies have not benefited from user behavior analysis .

3.5 Qualitative analysis

User experience is a top priority for any enterprise. In many aspects, including product design, user research, R&D, operations, marketing, customer service, etc., it is necessary to understand the user's real experience process. However, how to optimize user experience has always been a topic of internal controversy, mainly because it is difficult to describe it in a specific and vivid way through quantitative data analysis. When identifying abnormal user behavior through behavioral analysis, being able to reproduce the specific scenarios in which users use your product is crucial to optimizing the product experience.

When I was at Taobao , the user experience department would optimize the experience by inviting users to the company for interviews and usability experiments. However, this method requires a relatively high investment of time and money, and the samples may not be representative. In order to solve this problem, Shujike developed a user behavior screen recording tool. It saves costs by eliminating the need to invite users to the company for on-site recording. It intuitively and efficiently restores the user's real operations in the form of video, allowing all positions in the company to grasp first-hand information on user experience, helping product development to improve user experience.

Figure: User behavior screen recording and playback interface

3.6 A/B Testing

（1）What is A/B testing?

A/B testing is a method of scientifically optimizing products through data analysis. It develops two or more plans for the same optimization goal, randomly selects two groups of users, and lets one group of users use plan A and the other group of users use plan B. The click-through rate , conversion rate, active retention and other indicators of different plans are counted and compared to find the best product decision plan. In the lean startup philosophy, don't make one big, comprehensive thing, but instead keep making small, precise things that can be quickly verified. Quick verification, how to verify? The main method is A/B testing.

It should be noted that A/B testing is not a simple comparative test. 99% of domestic companies mistakenly believe that it is just a comparative test. Through simple proportional indicators, they select a group of methods with better performance and then find that this group of methods has led to a decline in the overall indicators. The reason is that the A/B testing method is incorrect, and no statistical method is used to scientifically interpret the test results from the random distribution of traffic.

（2）The value of A/B testing

• Avoid risks: "Post-hoc" product verification. If it fails to meet expectations, rollback will lead to high development costs and high risk of customer churn.
• Scientific decision-making: Most product managers rely on intuition to make decisions, but the reality is that what we think is not necessarily what users think; even the best PMs cannot run more than half of the A/B tests;
• Low cost and high efficiency: In the traditional development process, the launch needs to be scheduled, and the development iteration efficiency is low. AB testing does not require version release and can directly and quickly verify the solution.

(3) Application scenarios of A/B testing

• Can the new page increase dwell time and improve the conversion rate of key behaviors?
• Can adjusting the button style bring more clicks and improve conversion rates?
• Is the new process smoother and a better user experience than the old one?
• Can the new algorithm effectively improve product conversion rate?

(4) How to conduct A/B testing

• Develop a clear test plan (time, quantity, goals, success criteria)
• Define measurable conversion criteria
• Find test elements, publish test plans, and allocate traffic
• Track data performance, adjust test elements, and find the best solution
• Continuous Improvement

Mistakes not to make:

• No clear test plan
• Too little traffic, inconsistent distribution, too long
• Lack of monitoring
• No evaluation criteria

Summarize

At the macro level (strategy and planning), the fish pond theory and opportunity well theory in the growth methodology are used to help plan the company's target market and customer groups, clarify the depth of the core value provided by the company, use AARRR, consumer decision-making model, and behavioral motivation model to complete PMF (product-market fit), and find the Aha moment (key retention behavior) and quickly convert it into a North Star indicator;.

At the micro level (practical methodology), 6 conversion rate analysis models are used to improve product conversion rate and user experience; if everyone can master the growth methodology and the secret to improving conversion, they will definitely be able to achieve rapid business growth.

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

Product promotion services: APP promotion services, advertising platform, Longyou Games