The rise of B2B e-commerce in recent years, especially platforms centered around 'finding' (e.g., finding suppliers or products), has brought two controversial concepts to the forefront: matching (撮合) and self-operated (自营) models. Under pressure to generate revenue and profits, many platforms have gradually adopted self-operated businesses. The commodity market rebound in 2016 taught harsh lessons to entrepreneurs who lacked real trading experience. There is no definitive answer to what constitutes true self-operated business, which businesses should adopt self-operated models, and which should use matching; it requires exploration by each platform.
Matching (撮合)
Definition of Matching Also written as 撮和. According to Baidu, matching refers to in multi-party transactions, an intermediary party aggregates information from multiple parties and then matches the information to meet the information needs of all parties. For example, in multi-party transactions, a market operator or broker matches buyers and sellers according to market rules.
Matching Services Matching services refer to the process where a service company or individual, starting from people's actual lives, acts as a third party to aggregate multi-lateral information, organize and evaluate it, and then provide effective information to users in a targeted manner. This process is collectively called matching services.
This industry originated in the Western Zhou Dynasty, where intermediaries were called 'zhiren' (质人). In the Western Han Dynasty, they were called 'zangkuai' (驵侩). After the Tang Dynasty, they were called 'yaren' (牙人). In the Song Dynasty, an official system for yaren was established. During the Ming and Qing Dynasties, with further commercial development, the number of yashang (牙商) increased significantly, forming a specialized guild called yahang (牙行). In modern society, the commercial intermediary organizations that replaced yaren are mainly various exchanges, trust companies, brokers, etc.
We can see that matching plays an intermediary role, and it is an industry that has existed since ancient times. In less polite terms, we call it 'pimping' (拉皮条). After professional division of labor, matching, as a professional information service, is widely distributed across various industries.
This wave of 'finding' brought matching to the forefront, essentially promoting the matching model from traditional industries to a wide range of trading and circulation fields.
As usual, here's a mind map~
First, we must clarify that matching is indeed not a new thing. Where there is information asymmetry, matching may appear. For example, real estate agents (real estate brokers, various real estate agencies), marriage brokers (matchmakers, bingren, Baihe.com, Jiayuan.com, Zhenai.com), and intermediaries for finding connections (we can call this resource cross-border integration).
In the earliest trading companies and circulation companies, salespeople had to develop channels themselves, and transaction prices were highly opaque. This also created significant space for rent-seeking in the circulation field. In the early days, business was relatively easy, essentially because the degree of information asymmetry was high.
In the commodity sector, professional matching has also existed for a long time, initially in the form of individuals (who made fortunes early on), and later various professional matching companies emerged. In China, one of the better performers in the commodity field is Yimao (Yitong). The value brought by professional services is significant.
What Pain Points Does Matching Solve?
Efficient Trading Some transactions have extremely high time requirements, especially for commodities where price fluctuations are significant. In extreme cases, futures market liquidations can happen in an instant.
Matching solves the rapid completion of transactions. Manual is the least efficient method; manual plus systems is a step forward; electronic trading for stocks and futures is the most advanced form of matching.
Information Matching The primary function of matching is information matching. This includes transaction information and information related to supply chain services that may occur after the transaction.
In transaction information, we cannot simply understand it as price matching. There are several key points:
First, price discovery. Except for futures commodity exchanges, most commodity markets do not have real-time prices for commodity transactions. If I want to buy 20 tons of rebar produced by Baosteel at 10:00 AM, someone needs to tell me the exact price.
Second, prices often come with conditions, such as payment terms, delivery time, delivery location, etc. Additionally, there are hidden characteristics like counterparty qualifications, trading preferences, and product quality differences.
Third, transactions often involve price negotiation. For example, A is willing to buy at 6000 yuan/ton, and B is willing to sell at 6010 yuan/ton. Whether A or B is willing to compromise and whether a deal can be reached often depends heavily on the matching function. Without matching, a negotiation deadlock may occur, forcing a deal.
Fourth, during negotiation, if the matching party can provide professional judgment and consulting services, traders can make better decisions. This type of matching with built-in consulting services is of high value and requires a high level of professionalism from the matching personnel.
Fifth, the trading environment is not a blank slate; there is considerable room for rent-seeking. If these interests can be communicated through an appropriate matching channel, the likelihood of reaching a deal increases significantly.
Therefore, in B2B transactions, even for standard SKUs, there are many things that need to be communicated and confirmed. In the current stage, human communication is often more efficient. Some things even require human communication.
Matching is a business. The more refined it is, the more timely the information provided, the wider the coverage, and the more stakeholders bound, the greater the service revenue from matching.
The Significance of Matching
Customer Acquisition Through matching, platforms can reach a large number of customers, whether upstream or downstream, traders or end-users. Most users in the industry who need to complete transactions can be contacted through matching. Even if a deal is not reached, matching can play a significant role in user coverage and activation.
Data Value The data value here includes both the transactions completed and the potential logistics and supply chain finance referral value after the transaction.
Of course, if a platform has sufficient transaction coverage and obtains enough real transaction information, data analysis can be used to make many market judgments.
The value of data often lies in its proper use. No matter how much raw data there is, without analysis and processing, it has little value.
For data value, having as much real market transaction information coverage as possible is more valuable than the matching itself. However, in a competitive environment, obtaining this real transaction information requires clever methods.
Information Service Fees Charging for matching services cannot support a trading platform's revenue. But from a profit perspective, truly efficient matching can cover the personnel costs of matching. However, to directly monetize this point, the role of the internet is small, and the value is relatively low.
On the other hand, free matching has a lower barrier. From an internet perspective, charging matching fees is a last resort. Once free matching becomes paid, many long-tail customers will be lost.
Categories Suitable for Matching
Commodities with high price volatility and market speculation are most suitable for matching, but such products are often difficult to convert into supply chain services. (Contradiction)
In general, high-frequency transactions, standard products, low-margin distribution, large order values, and transactions between strangers are typical characteristics suitable for matching.
But there are also many other variants of intermediaries. Nongtanjun recently encountered a very interesting model, but I'll keep it a secret and share it after exploring its actual effects.
Ways to Improve Matching Efficiency
Team Collaboration Increasing personnel brings diminishing marginal returns to matching efficiency. From an internet perspective, linear growth is not enough; exponential growth is what matches the explosive nature of internet efficiency.
But is it possible that in the early stages of industrial internet, at a certain point, without sufficient market penetration, matching efficiency itself is difficult to achieve exponential growth? Platforms may have other exponential growth points.
Improving Efficiency From the perspective of improving efficiency, small team collaboration is the most efficient. Besides team morale, incentive systems, and suitable personnel,
appropriate grouping, such as by region or category, focusing on finding buyers or sellers, etc., all need to be tried to know what works. Targeted analysis of user trading behavior and habits is also necessary.
CRM software is very helpful. Incorporating individual information into system information exchange, supplemented by system prompts and judgments, can already achieve semi-automated standardized operations.
Full automation of matching in B2B is very difficult. The closer to C-end trading habits, the more likely online transactions can be realized. Fully automated matching for commodities—try it and you'll know~.
Objective View of Matching
- Matching services are of great value in improving transaction efficiency and providing matching for asymmetric information.
- There is no necessary relationship between the internet and matching, but matching is not necessarily unsuitable for internet applications.
- Professional matching is a business and must follow basic business rules. Improving efficiency and reducing costs are key considerations for matching.
- Not every industry is suitable for matching; analyze specific situations.
- Don't blindly believe in matching, nor fear it. For B2B, doing matching is actually competing with offline matching companies. Whether it's worth it depends on discovering points beyond matching itself; only then is matching valuable.
Self-operated (自营)
Definition of Self-operated Baidu's basic explanation is producers directly operate their own products. This is a literal interpretation and does not match the actual trading environment.
Those who have experience in trading and supply chain circulation know that the definition of self-operated should be the company as the main entity conducts trade and buying/selling of goods, bearing all risks (including price fluctuations and supply chain circulation) and enjoying corresponding returns or losses.
Note two points: First, the company is the main entity for trade flow, meaning the platform (company) must handle accounts and issue invoices. Second, it must bear risks and enjoy returns.
Mind map here,
Purpose of Self-operated
Transaction intermediary. Some might say, 'Are you kidding? Isn't transaction intermediary what matching does?'
Actually, B2B platforms have distorted a concept by calling platform invoicing 'self-operated.' Strictly speaking, what's missing here is bearing market price risk and possible supply chain services.
Theoretically, self-operated in B2B e-commerce platforms should be a 'back-to-back' transaction. In this business, the platform should not aim to profit from price differences as its core goal (if the purpose is only to earn the spread, then it's not a trading platform but an internet+ trade distributor, which is another path).
If not just to earn the spread, what is the purpose of the platform doing this 'self-operated'?
- It solves the guarantee attribute of transactions between strangers. A doesn't know B, but both A and B recognize the platform's credit, so they trade through the platform. Here, self-operated only serves as a pass-through.
Through the platform's intermediary and credit attributes, it provides the basis for transactions. Although most transactions are strong relationships between acquaintances, solving these weak-relationship transactions, accumulating users, and building platform credibility is a process that needs persistence.
- Doing 'self-operated' here is for customer acquisition and expanding sales channels. Through pass-through trading, customers are obtained, and it's possible to establish user sales channels.
Another purpose of customer acquisition is to better provide integrated services (three flows), truly transforming and driving logistics and supply chain finance after transactions. First, use 'self-operated' to attract traffic, then convert. Whether the conversion rate is high is one of the core points we need to focus on in self-operated.
- Through 'self-operated' trade, transaction flow is obtained. From a trade completeness perspective, matching has a lower grasp of trade. Generally, pass-through trading is higher in both authenticity and control. However, from the perspective of VAT invoicing, whether flat-in-flat-out complies with tax regulations needs to find the right point.
3.1 What can you do with trade flow? 3.1.1 You can obtain bank credit for the trading entity. This is why trading companies have always been doing flow. It's forced by China's banking system; without transaction flow, no credit can be granted. Some commercial banks are already ahead in credit systems, but it will take time to change. Bank credit loans have relatively low capital costs, so this is a channel that must be valued. 3.1.2 With transaction flow, you can do a certain level of data accumulation and analysis. You can analyze user habits from a trade perspective and also guide transaction data to supply chain finance risk control models.
Although data completeness, i.e., customer penetration, is an extremely difficult pain point and problem, if the goal is clear, gradual efforts can be made to convert.
- Channels (Taiwanese call it 'pipeline,' which is more apt) Building channels is building infrastructure. For true channel distribution, platforms should avoid market speculation as much as possible.
Forward price speculation and providing credit terms to customers both carry risks, but theoretically, the value of providing credit terms should be far greater than speculation.
Pure speculation—Nongtanjun has been in the industry for over a decade and has seen countless trading tycoons stir up storms in the market, only to end up liquidated, leaving a mess. The market doesn't believe in tears; speculation is gambling. If you win ten times and lose once, the wins will be lost.
On the other hand, the smoother the channel, especially for products with strong distribution attributes, occupying channels through trade distribution and occupying channels through distribution are complementary development priorities.
The better the channel, the more attention must be paid to channel service efficiency. Earn channel profits by improving efficiency, not simply by exploiting information asymmetry. Some types of products may have high channel spreads, but with the trend of information transparency, these spreads will eventually diminish significantly.
How to truly build channels well, improve your own efficiency, partially benefit customers, and extend customer LTV (lifetime value) is what should be done solidly.
Suitable Categories
Low-frequency transactions, small and dispersed amounts This corresponds to long-tail customers. The value of long-tail customers in channels is completely different from that of head customers.
Long-tail customers are the core source of self-operated distribution profits. Efficiently allowing long-tail customers to enjoy the services that head customers receive, even if not fully, should be where the internet plays its most significant role.
Opaque prices, high gross margins Price opacity exists widely in all aspects of the supply chain. The closer to the consumer end and the end of the supply chain, the higher the opacity. For self-operated, a vertical environment with certain distribution profits is relatively ideal.
Intense platform competition, overly monopolistic goods channels, and insufficient weak-relationship trade markets are all enemies of B2B platforms doing self-operated.
Commodities generally have high volatility, so theoretically they are not suitable for self-operated trade. Pure 'back-to-back' opportunities are too rare. For products with low price volatility and heavy channel dependence, the platform's need to handle price risk is much lower.
However, if the platform has the ability to do medium- and long-term trade, aggregate medium- and long-term procurement and sales, combine futures hedging to lock in related risks, and strictly execute, it is possible to explore a suitable path for commodities.
Related products For example, if I sell product A, and in this industry, customers also use B, C, and D, then product expansion is natural. For the platform, there is almost no new customer acquisition cost; the challenge is the ability to obtain channel resources.
Inventory sharing, centralized distribution, and intensive logistics services can greatly improve service efficiency and reduce logistics costs.
Key Points for Self-operated
Clear Goals First, goals must be mentioned. Whether B2B is a platform or a channel must be clear from the start.
If it's a channel, build the channel solidly. Use various methods—ground promotion, internet tools, matching traffic, leveraging people in the channel, SAAS—use all worthwhile methods to build the vertical enterprise funnel and truly build the enterprise user funnel.
If it's a platform, try to avoid self-operated trade as much as possible. Don't abandon the core foundation of matching just because matching encounters difficulties, revenue difficulties, or data difficulties.
Even if you do it, try to avoid doing it on the same products as matching.
Because:
Competitive Exclusivity Cooperation and win-win are not empty words.
In corporate competition, if a platform acts as both a service provider and a trader, it is severely exclusive from a customer acquisition perspective.
Imagine I am a trader. The platform does both matching and self-operated for the products I deal in. If I let the platform know my customers, won't they steal them in no time? How can I not resist the platform? Not only will I not trade through the platform, but I also won't use its logistics or financing. Business is a zero-sum game; there isn't that much incremental growth. If you put the platform against users, the business is hard to do.
If I am a supplier, I might have handed over small, dispersed channels that are not cost-effective for me to distribute myself to the platform. But if the platform engages in channel crossing, white-labeling, and eating into multiple margins, it earns profits but steals channels. If suppliers and manufacturers all resist the platform, can the platform build its own factory to produce the products?
If I am a customer, I originally procured products through acquaintances, who gave me certain benefits over time. Even if I am a legitimate large company without benefits, I still have choices in supplier service attitude and efficiency. For procurement, having two or three alternative procurement channels is normal.
When the platform comes in, claiming to do 'sunshine' procurement, ERP integration, and one-stop service, but its actual service efficiency and 'quality' are not as good as the original suppliers, guess whether I will use the platform's channel often?
Tiering and Layering Customers need to be tiered and layered. The traditional ABCD classification in sales is not fully applicable in B2B platforms. We need to distinguish the different needs of long-tail, head, and middle-tier users and provide differentiated services accordingly.
This sounds vague and is hard to standardize because each industry has different characteristics. Refined operations also place high demands on corporate governance.
Most B2B e-commerce has strong traditional genes. How to truly use internet tools and operational models is a key point that needs real thought and exploration.
Self-operated and Matching
For internet B2B e-commerce, whether matching or self-operated, the core point should revolve around improving efficiency.
Use whatever is efficient: if manual is efficient, use manual; if systems can improve efficiency, use systems; if ground promotion is efficient, do it; if self-operated is efficient, do it. If you really want to say which is better, self-operated or matching, there is no result.
Essentially, B2B e-commerce is a business, and since it's a business, it cannot violate business rules. It must reflect the characteristics and efficiency of the internet while conforming to the rules of the business environment.
Doing B2B well is not easy. Besides managing the company's organizational efficiency, it must also improve the efficiency the entire company brings to the industry. Moreover, a reasonable model needs not only incremental growth but also growth in the incremental rate. A linear growth project is not without value, but it doesn't have as much internet value.
Source: 思维浓汤 (ideasoup)
-END-