Product operation: How to increase the payment rate?

Recently, I occasionally discuss some issues with my friends for our own entertainment; one of my friends asked me this question: How to increase the payment rate?

I asked him whether this was referring to a specific business or a general one, and he said it was in a general sense. I asked him what he thought about it, and he said he hadn't thought it through yet, but his general idea would be to consider it from the perspective of new users paying and old users paying.

I just talked about my thoughts, and then these two days I happened to have some ideas and inspirations. I felt that this was not a problem, but a type of problem, so I did some deductive thinking based on that problem.

This article mainly includes two parts. One part is about thinking about improving the payment rate, and the other part is the deduction based on this, expanding it to a more general boundary.

1. How to improve the payment rate

First, let me talk about the temporary answer I gave that day:

This is how I understand it: payment rate = paid UV/overall UV - what we need to do is to increase paid UV, or improve the quality of overall UV. The latter is actually a question of channel and user matching, and the former is the object of discussion.

The corresponding payment conversion path for users using the product is non-payment - first payment - continuous payment - loss.

Then the corresponding problem of increasing paid UV becomes:

• Why do users pay? How to increase their motivation to pay and lower the threshold for payment?
• Is there any way to pass on the benefits to users and allow them to complete their first paid conversion?
• Why users should continue to pay, the benefits of additional payments, and the penalties for not paying;
• Why do users churn, how to extend LT, how to warn of churn, and how to recall them after churn.

Later I thought about it again. The payment rate is actually fundamentally related to the characteristics of the business model itself, and the definition of payment may also be different. For example, the payment rate values ​​of e-commerce and value-added service products are definitely different.

After careful consideration, the differences in payment between different business models are mainly determined by two points: one is the product decision cycle, and the other is the user life cycle.

The product decision cycle refers to how long it takes for users to learn about a product for the first time and finally decide to take action, as well as the complexity of the decision. This is affected by many factors, such as the degree of standardization of the product itself, the price, the cost of trial and error, and the decision-making scenario.

For example, buying a car or a house is a typical thing with a long decision cycle; which takeout to order or which clothes to buy are things with a relatively short decision cycle.

Generally speaking, the higher the degree of standardization of a product, the lower the price, the lower the cost of trial and error, and the easier it is to make a decision.

On the contrary, it will be difficult to make a decision. At this time, there will be many means to assist decision-making, such as free product trials, long-term follow-up to establish a trusting relationship, etc.

The user life cycle mainly refers to how long the relationship between the user and you is. Is it one-time, low-frequency, or medium-high frequency? This depends on the business model itself. For different businesses, the user life cycle is definitely different.

For example, you may only buy a house once or twice, but you will buy clothes every now and then, and you may even order takeout every day.

The length of the user life cycle directly determines the formulation of subsequent strategies.

One-time business mainly focuses on customer acquisition cost, average order value, and conversion rate. After all, it is a one-time deal. Low-frequency business is to make users think of you in specific scenarios, and the value of a single user can be greater than the customer acquisition cost.

Medium and high-frequency businesses mainly focus on customer acquisition costs, stickiness, repurchase rate, ARPPU, LT (life cycle) and LTV (life cycle value).

Different business models are naturally different, and the characteristics of payment rates are also different, so we can only come up with such a basic assumption:

• The big premise is related to the business model itself, and mainly depends on the product decision cycle and user life cycle;
• A small premise, the conversion steps mentioned before, paid conversion of new users, and continued payment from old users.

To be specific, we can only analyze specific problems specifically.

Second, deduce

This is just one question, but can it be extended into a series of questions, such as how to improve XX, for example, how to increase the turnover? How to increase usage time?

So I started thinking about whether there is a universal way of thinking that can help solve similar problems.

My initial answer was to quantify the indicators first, then break them down according to different dimensions, then find the related influencing factors, come up with some guesses and hypotheses, and finally confirm or falsify them.

Let's look at each one separately:

1. Quantitative indicators

The main thing here is to define the problem clearly. If it can be quantified, then quantify it. If it cannot be quantified, then see if you can find a way to quantify it. After all, things that cannot be quantified are difficult to evaluate and improve.

For example, regarding the payment issue mentioned above, assuming there are 1 million DAUs and 10,000 paying users every day, is that a lot or a few paying users? And what kind of users are considered paying users?

Therefore, the first step is to find appropriate quantitative indicators and clearly define the relevant calculation standards.

The second step is to find a suitable formula for quantitative expression. After quantifying it into a formula, we will know which related indicators will affect this indicator.

For example, profit = revenue - cost - this is a basic formula. If you want to increase profits, all you need to do is increase revenue and reduce costs.

For example, the payment rate mentioned above = paid UV/total UV, so what we need to do is to increase the numerator, and improve the quality of users when the denominator remains the same.

The next step is to continuously refine this basic formula.

2. Dimensional decomposition

One way of thinking is to split the data from the whole to the part in spatial order, for example, income = income from business 1 + income from business 2 + income from business 3…

Another way of thinking is to organize the whole process in chronological order, that is, the natural flow of things happening, to see which links are before, during, and after the event; for example, the revenue formula based on the conversion path, revenue = traffic * click-through rate * purchase conversion rate * purchase success rate * average order value.

If you don’t have a clear idea, just list some dimensions for analysis first, and then check them out as you go, such as the common gender, age, region, etc.

The purpose of the decomposition is to integrate as many related influencing factors as possible into the basic formula of this quantitative indicator.

Formula before disassembly:

Profit = Revenue - Cost

After initial disassembly:

Profit = traffic * click-through rate * purchase conversion rate * purchase success rate * customer unit price - (labor cost + marketing cost + software and hardware cost)

Then just keep disassembling and improving it.

The more detailed the analysis, the deeper the understanding of the business, and accordingly, the stronger the control over the business.

3. Find the influencing factors

In terms of broad classification, the influencing factors can be divided into the following categories:

• Those that contribute positively to the goal, the so-called driving force;
• Those that contribute negatively to the goal, the so-called resistance;
• It doesn't have much impact.

What needs to be done is to clarify the classification of the current influencing factors, and then enhance the driving force and reduce resistance. According to the 80/20 rule, give priority to factors with high contribution to the goal, high degree of influence, and large impact range.

Finally, consider the difficulty and time required to comprehensively evaluate the cost-effectiveness and priority.

4. Make a hypothesis and confirm or disprove it

Finally, based on the above assumptions, we make some corresponding adjustment strategies to confirm or falsify our ideas, which is what we usually call MVP and iteration.

The above part may still be a bit abstract, so let’s end with a case study. Let’s look at it as a whole. Your friend plans to open a Taobao store and wants to discuss with you whether this can make money and whether you want to give it a try.

So you slowly tell him that we are going to open a Taobao store, and the ultimate goal is to make money, that is:

Profit = Revenue - Cost

From the time a user sees our products or stores to the time they finally complete the purchase conversion, they will generally go through the pre-purchase, purchase, and post-purchase stages.

Specifically, it is:

See the store or its products - become interested in buying - enter the store or the product details page - browse the products - decide to place an order - place an order - successful purchase - receive the items - continue to buy or leave.

In order to support the above-mentioned user behaviors, we need to do the following: product selection, procurement, buying advertisements to acquire users, optimizing stores and detail pages, logistics and delivery, customer service consultation, and maintaining user relationships.

Based on the basic revenue formula + various factors affecting the user conversion path, we can get an upgraded formula like this:

Profit = exposure * click-through rate * order conversion rate * payment success rate * average customer unit price * repurchase rate * user life cycle - (purchase cost + marketing cost + labor cost + inventory cost + other expenses)

The next step is to see in which areas you have significant competitive advantages if you want to open a Taobao store, what driving forces you can increase or what resistance you can reduce, and finally look at the target group and product selection.

Your friend exclaimed, “Oh my god, this is too complicated.”

Let’s do a simple calculation. As long as every order is profitable on average, this will be true, that is, the revenue of a single order > the cost of a single order.

You can calculate how much your monthly expenses are, and then see how many transactions you need to complete to reach the break-even point, or spend money in the early stages as a traffic-generating tool, and then increase profits through profitable products.

Finally, to summarize briefly, for this kind of improvement of XX, you can first quantify the indicators and derive the basic formula; then continuously improve the basic formula by breaking it down in different dimensions; finally, find the key influencing factors, and then verify and iterate based on some of the conjectures obtained.

It should be noted that this quantitative decomposition method is suitable for mature business models, but may not be applicable to businesses that are still in the exploratory stage. However, some ways of thinking can still be reused.

Author: Wang Jiachen

Source: Product Manager from 0 to 1