Summary of Key Points
This article reveals a counterintuitive phenomenon based on the practical experiences of frontline retail practitioners and industry observations: E-commerce, which once disrupted traditional brick-and-mortar businesses with its low-cost advantages, now faces operating costs (especially "traffic taxes") that are on par with or even higher than those of physical stores. The widespread adoption of AI has not reduced these costs; instead, it may have made traffic taxes even more expensive. The future for businesses lies in moving away from dependence on traffic and returning to a focus on providing personalized, human-centered services.
1. Online Costs Surpassing Offline Costs: Traffic Taxes Become the New Burden
In the past, e-commerce was considered cheaper because it eliminated the need for rent (a form of "space tax"). However, this has changed:
- Online Costs: For top-tier platforms in highly competitive categories such as clothing and food, the total cost ratio (all costs as a percentage of revenue) ranges from 35% to 38%. The majority of these costs comes not from the explicit fees charged by the platforms (which are only around 3% to 5%) but from traffic taxes. Businesses compete fiercely to gain visibility through advertising services like direct marketing and TikTok ads, which account for approximately 30% of the total cost.
- Offline Costs: For established community-based stores, the total cost ratio is between 25% and 35%, mainly consisting of rent (12% to 18%), labor (15% to 20%), and utility expenses (5% to 8%). These are fixed costs that can be predicted with ease and do not increase passively due to algorithmic factors.
In short, while online businesses may seem less capital-intensive, they actually have to pay a "traffic protection fee" to the platforms for each transaction they make. Offline stores, on the other hand, invest in building customer trust (e.g., through regular customers in community-based stores), which are not immediately lost.
2. Exponential Increase in Customer Acquisition Costs
The cost of acquiring new customers has skyrocketed over the past twenty years:
- In 2003, it was only 8 to 10 yuan; by 2013, it had risen to 100 yuan; in 2022, it reached 1302 yuan on Alibaba and 806 yuan on JD.com, an increase of 100 times!
- Businesses face a dilemma: If they invest in advertising, sales may increase, but their net profits decrease (with promotion fees accounting for over 50% on platforms like Taobao, and at least 35% on platforms like Dadao). If they don't invest, their stores won't get any visibility and won't be able to sell products. For example, Taobao's advertising tools have changed multiple times (from Direct Marketing to Gravity Cube to Wanxiang Platform), but the fundamental principle remains the same: the highest bidder gets the traffic.
3. AI Has Made Traffic Taxes More Expensive? Homogenization Leads to Fiercer Competition
Everyone thought that AI would reduce costs and increase efficiency, but it has actually increased traffic taxes:
- With everyone using AI to create content and provide customer service, the quality of content has declined, leading to greater homogenization. Platform algorithms no longer rely on content quality as a criterion for visibility and instead favor those who can afford to pay more for advertising.
- Experts predict that AI will not distribute resources more evenly but will give platforms (the tax collectors) more control over data and computing power. For example, Bridgewater Associates' Ray Dalio said, "AI allows those with the means to take a larger share of the pie." In the retail industry, this means that businesses using AI to create content only allow platforms to extract higher taxes.
4. The Way Forward: Moving Away from Traffic Dependence and Focusing on Human-Centered Services
To survive, businesses need to build barriers that AI cannot replace:
- Multi-channel Engagement: For instance, the author managed a frozen food brand and shifted from relying solely on online advertising to combining offline stores, private domain consultants, and emotional connections, which helped the company turn losses into profits within half a year.
- Enhanced Service Quality: The success of businesses like Pangdonglai and snack vending stores is not due to sudden improvements in physical stores but because of high online traffic taxes. The personal touch provided by offline stores (e.g., free services and customer relationships) has become an advantage that AI cannot replicate or tax.
- Private Domain Control: By retaining customers within their own ecosystems (e.g., through WeChat communities), businesses can avoid paying additional traffic fees to platforms.
Conclusion
E-commerce once used a "rent-free, tax-free" strategy to disrupt traditional retail, but now it has returned to the same cycle of having to pay "traffic taxes." In the next twenty years, only those businesses that maintain the human touch and provide valuable services will survive. After all, no matter how intelligent AI becomes, it cannot replace the trust and emotional connections between people.
(The entire article is written in plain language, making it easy for non-financial professionals to understand.)