Product architecture for e-commerce user growth

As a growth veteran with many years of experience, let me briefly talk about the product architecture of growth.

The relationship between product stage and growth

When I first read books about growth hacking, there was always a question lingering in my mind: What is the difference between a growth product manager and a user product manager? User products are designed based on demand; growth hackers emphasize that growth is not just about attracting new users, but about managing the entire user life cycle (attracting new users - activation - retention - payment - recommendation), which is a scientific method obtained by skillfully applying AB testing. From the perspective of the product life cycle, we divide the product development stages into from 0 to 1 and from 1 to 10. User products are more like designs from 0 to 1, and growth hackers are refined iterations from 1 to 10. The two belong to two different stages of product development (growth). An excellent product manager needs to master relevant theories and skills at the same time.

To coordinate the overall product situation, an excellent product manager must first determine the growth potential of the product. Even a pig sitting on the wind can fly to the sky. Find the market during its bonus period and quickly design product features, and the product will have room for rapid development. The stage from 0 to 1 of a product is to create the "momentum" for growth. At this stage, the product must meet user value or have differentiated product design. The fastest growing areas in recent years are undoubtedly short videos and sinking markets. Their characteristics are:

1. New value. We observe the development history of short video platforms, from Kuaishou to Douyin, and then to Video Account. The emergence of new products is because some user needs are not well met; on the other hand, the express versions launched by Douyin and Kuaishou are also a supplement to the main products, using the product matrix to expand incremental space.
2. Incremental population. Pinduoduo has undoubtedly benefited the most from the incremental space and population released by WeChat mini-programs in the e-commerce field. Relying on its white-label and special-price positioning, it has quickly emerged as one of the three major e-commerce platforms. This market positioning was also copied by Taote and Jingxi, which quickly reached the incremental population in the sinking market.

The stage from 1 to 10 of a product is the “art” of growth. At this stage, you need to formulate a North Star indicator, determine a growth formula, and find the product’s Ah Moment. Below, we use the common AARRR model to break down the architecture of e-commerce user growth products in stages. It is important to note that product value must be integrated throughout the growth process.

(Illustration - AARRR model, picture from "Interpreting the correct use of the AARRR model in WeChat mini-programs")

Product Framework for User Acquisition

Where do users of e-commerce products come from? Common methods include natural traffic, promotional traffic, and product matrix. At this stage, we need to expand the scope of our product's influence, reach more potential users, and convert potential users into new users. This involves user acquisition and recommendation propagation in the AARRR model.

1. Organic traffic

The proportion of natural traffic for new products is very low, and the proportion of new users is very high; the natural traffic of mature products can exceed 50%, and the proportion of new users is very low. The platform's reputation and marketing events will affect the natural traffic of the product. A recent case is Erke. Responsible donations continue to ferment the brand's reputation on the Internet. The natural traffic of its live broadcast rooms on various platforms has skyrocketed, and its products are sold out. In the first few years after growth hacking became popular, major companies sought to create cases or marketing events that would go viral, which ultimately affected natural traffic.

Brand promotion and specific promotional nodes are closely related to natural traffic. The low period during the Chinese New Year every year and the surge in traffic during major promotions are both reflected in the increase or decrease of natural traffic. Natural traffic is mainly driven by market or marketing personnel. What product managers need to do is to be able to efficiently handle natural traffic, enable users to quickly find and use products on various platforms, and build a basic product architecture, such as h5, mini-programs on various platforms, and API docking with app stores, etc. The products in this section are all part of the infrastructure for growth products.

2. Promote traffic

All traffic that requires money to acquire users is considered promotional traffic, which involves costs and requires accurate data statistics.

Advertising: Use a fixed cost to acquire new activations and observe the subsequent conversions of these users. This is the simplest way to place ads. Its characteristics are a mature business model, the ability to acquire new users stably, the need to focus on the ROI of promotion, and the search for high-quality promotion channels. The main task of a product manager in an advertising campaign is to increase the conversion rate of the campaign, thereby improving ROI and cooperating with the market to expand the number of users acquired through the campaign. Common methods include optimizing delivery landing pages, providing more delivery landing page types, optimizing delivery materials, building a data monitoring platform, and automating docking to improve delivery efficiency. E-commerce advertising is a systematic project. Some companies cooperate with agents, while others operate directly by themselves. Product managers should pay more attention to the entire process and optimize the key nodes in the process.

CPS Alliance: From Taobao to JD Alliance, the most traditional distribution is a stable way to acquire users and sales. In the PC alliance era, 30% of the traffic of an e-commerce brand I once worked for came from the sales alliance. With the advent of the mobile Internet era, major social e-commerce companies have made rapid progress, walking on the edge of policy control. A major feature of the CPS alliance distribution system is that it has a relatively independent and complete product structure. Building a distribution product system, from rules, tools, data to commission settlement, is the basic work of a product manager. The next step is to provide fine management of distribution users, use product mechanisms to guide the expansion of the scale of distribution users, and promote more active users. Distribution platforms, influencers, and ordinary users are common distribution partners in the alliance. Online education platforms use secondary distribution to flood WeChat Moments with courses, which is another eye-catching work around distribution in recent years.

Other promotions: Online traffic costs are becoming increasingly expensive, and there are a large number of potential users offline. OTA platforms represented by Qunar once converted offline users to online and acquired a large number of incremental users. The characteristics of offline users need to be combined with business characteristics to find suitable scenarios. The disadvantages are that the cost of user education is relatively high, the quality of users is easily low, and the promotion cost is high. In addition, factory pre-installation, cross-industry cooperation, specific channel cooperation, etc. are still effective promotion methods, which can be selected according to the characteristics of each platform.

3. Product Matrix

Social fission: Using social relationships on WeChat to promote products has become a mature model since Pinduoduo’s Kan Yidao. Routines will only become outdated, but they will never become ineffective. By combining its own business and constantly introducing new routines, social fission can continue. After designing a fission strategy, you need to pay attention to relevant indicators such as sharing rate and fission rate. Regarding the growth product architecture of WeChat Mini Programs, I will write a separate article to explain it in detail later.

Platform dividends: Just as WeChat mini-programs have released huge incremental space, new platform products such as Baidu mini-programs and quick apps will release certain dividends. After the common APP, PC, and H5, WeChat Mini Programs have become the standard configuration of major manufacturers. In addition, Baidu Mini Programs, Quick Apps, etc. are enriching a company's product line. Product managers need to pay attention to the dynamics of these new platforms, promote rapid launch, follow up on platform policies, and introduce new users to the platform. Drucker once said that it is important to look for opportunities rather than just solve immediate problems. There is a window period for the dividends of these new platforms. If you seize the dividends, you can achieve rapid growth. In addition to WeChat mini-programs, another feature of platform dividends is that the ceiling is relatively fixed.

New traffic entrance: When Douyin and Kuaishou are developing rapidly, the small stores of Douyin and Kuaishou have become the new traffic entrance. Cross-flow of different businesses within the group, etc. Growth product managers need to constantly pay attention to industry changes, find new traffic entrances, and build new product connections.

Multi-product matrix: Some education companies or game companies, taking advantage of the attributes of vertical user aggregation, have developed more than 10 products, performed ASO optimization in the app store, and used vertical positioning for promotion to acquire more users. There are always various unique ways to adopt user acquisition strategies, and as long as they are effective, they can continuously generate positive returns.

(Illustration - User life cycle, picture from Product Manager from 0 to 1)

Product framework to improve conversion

How do users convert transactions? We need to stratify users according to their types and needs and build efficient conversion paths to improve users' order conversion rates. This involves user activation and revenue generation in the AARRR model. The average order value that generates revenue is affected by the platform's overall strategies such as free shipping threshold and discount threshold, and will only be discussed briefly here.

1. New User Guide

The core of new user guidance is to let users experience the core value of the product. For e-commerce platforms, the indicator definition of user activation should be the completion of the first order. For some tool platforms, new user guidance includes an introduction to functions to reduce the user's learning cost for product use. The core of the e-commerce platform’s new user guidance is the new user zone, which guides users to quickly convert and complete transactions.

Special area for new users: The discounts for new users will be gathered in the special area for new users. By lowering the average order value (1 yuan for purchase) or offering a full refund for the first order, the decision-making cost for new users can be reduced, allowing users to quickly place orders to experience the product. Some high-frequency platforms even start to guide users to repurchase in the new user area, offering preferential subsidy prompts for the second and third orders, allowing users to quickly pass the novice period. The standard for measuring whether the new user area is effective is to see how many of the platform's new user conversions are completed through the new user area. If a large number of new users fail to complete their first transaction through the New User Zone, the New User Zone needs to be optimized.

Registration and login: As basic functions, registration and login need to provide convenient services for users. Some e-commerce platforms with a long history need to be compatible with the habits of old users and should not blindly pursue the simplification of functions.

2. User Path

For old users, the most important point is to optimize the user's shopping path, that is, to design product functions around the user's movement lines.

Search recommendation: On mainstream e-commerce platforms, more than half of users find their desired products through search or recommendation and complete the purchase every day. Users with clear purchasing intentions will search directly. As e-commerce recommendation capabilities become stronger, users with unclear intentions often browse products recommended by the system. Search recommendations have become the main user purchasing path on e-commerce platforms. This also reflects the reality that users are attracted by products on other platforms and then make purchases on e-commerce platforms, and that e-commerce platforms are being turned into tools.

Home page: The home page is the starting point of the APP user path. Its role is to meet the browsing needs of more users through the setting of special columns while satisfying the user's main path, and to allocate traffic to different businesses. In addition to search, which is an active intention of the user, other columns or modules can adjust the distribution of traffic through rules. In addition to general modules such as search, recommendation, promotion, and classification, it is a common problem that the characteristic modules of products cannot be explored, resulting in similar functions of home page columns.

Product details page: The product details page has two functions: allowing users to quickly make purchasing decisions on target products and allowing users to purchase more products. The modules of business detail pages are becoming more and more standardized, and the differences among various platforms are similar. What they compete on is the refined operational capabilities and the product design capabilities combined with their own businesses.

Shopping cart: Shopping cart, checkout, and order are essential links in the purchase process. Complete functions, clear processes, and accuracy are the most important requirements.

3. Shopping guide

Around the user's path, there are a series of products that promote users to make consumption decisions and act as shopping guides, such as flash sales, limited-time sales, rankings, live broadcasts, short videos, content seeding, etc. These products act like lubricants to improve product conversion efficiency.

(Illustration - User growth conversion path, picture from "Tang Yihang: How to grow from the product dimension")

Product framework for improving retention

The core goal of user retention for e-commerce products is to promote repeat purchases. If the platform can provide high-quality goods or services, it can enable users to build brand loyalty and generate continuous purchasing behavior. Retention is a reflection of the comprehensive strength of a platform. No platform is perfect. What product managers need to do is to highlight the core advantages of the platform through product design and attract users who match the platform attributes to continue using the product. Study the curve of the relationship between users' purchasing behavior and retention. The turning point where the retention curve gradually becomes flat after users have purchased a certain number of times is the purchase target we want to guide users to achieve.

Membership: E-commerce membership has gradually entered the paid era, and free membership is like chicken ribs. By increasing the cost for users to leave through payment, and by offering discounts and concessions, we can cultivate users' habit of continuous use of the product. If they do not become members, they will be lost to other platforms.

Community: As a retention tool, e-commerce platforms all hope to retain users through the community.

Marketing: Traditional sign-in still plays a role. Users can accumulate platform discounts through cumulative visit behavior, attracting a fixed number of price-sensitive users to participate. E-commerce mini-games are more like an upgraded version of sign-in. The platform gives benefits through users' continuous interactive behavior, and users kill time through fun games and get virtual or physical rewards.

Push: Push and SMS together constitute user reach. Common e-commerce reaches can be divided into order-related information reach, user asset-related information reach, promotional information reach, and user behavior information reach. Due to the abuse of Push, the general effectiveness is decreasing and requires refined maintenance.

Community: Related to it is private domain traffic, which cooperates with operations. The product builds basic functions and marketing tools, and operations implement refined sop management.

Conclusion

Looking back at the growth formula of e-commerce: GMV = number of active users * order conversion rate * average order value. User acquisition and retention affect the number of active users, and improving conversion affects the order conversion rate and average order value. In the chain of growth, each step affects the final result. To achieve good growth, you should build a complete growth product architecture.

Author: E-commerce Product Notes

Source: E-commerce Product Notes