The emergence of AI has significantly improved the efficiency of person-product matching in e-commerce. "In the e-commerce field, everyone is concerned about how to maximize the efficiency of person-product-place matching. In the past, when searching for a keyword, the results given were almost 100% similar to the keyword. However, the LMA large model goes beyond that, making connections and associations. For example, if you ask it what is beautiful, it will make associations and connections. Therefore, when applied to e-commerce scenarios, the large model enables merchants to have their products discovered by users with less precise search intentions."

These words come from an interview with Hu Bao, Director of the Alibaba Mama Operation Center, during the live broadcast of "Full-scale Upgrade, Winning the Summer Increment - Alibaba Mama Full-site Promotion Upgrade Cloud Launch" on September 9th.

Online shopping has become an essential part of life, and the abundance and fragmentation of information are changing the buying habits of internet users. Consumers' purchase decision-making process is becoming increasingly shorter, and accurately capturing and converting their attention is generating new dividends within the Taobao ecosystem. After several months of beta testing, Alibaba Mama's full-site promotion was officially opened to all merchants in August.

Alibaba Mama's full-site promotion was officially launched in April to address the new changes in the e-commerce operating environment. In recent months, many small and medium-sized merchants who have not participated in the beta test have inquired about how the full-site promotion will change their daily operations. The simple answer from the editor is "more certain growth."

Data shows that during this year's 618 shopping festival, 130 merchants who were among the first to participate in the beta test achieved transactions exceeding ten million yuan through the full-site promotion. Over 1.5 million products leveraged the full-site promotion to gain traffic across the board, with an average increase of over 65% in GMV in the seven days after the investment compared to the previous period. Among them, over 300 individual products achieved transactions exceeding one million yuan. With the full-scale launch, millions of Taobao merchants will be able to directly use the full-site promotion to enhance their business value.

From the product launch in April to the full-scale rollout in August, news related to the full-site promotion has attracted the attention of Taobao merchants, resulting in a vast array of questions. During the launch event, Hu Bao, Director of the Alibaba Mama Operation Center, Qiguo, the person in charge of Alibaba Mama's full-site promotion product, and Jiechen, the person in charge of Alibaba Mama's vertical industry solutions, collectively answered questions regarding traffic, products, and ROI. The full-scale launch of the full-site promotion has also undergone several upgrades based on feedback from the market and merchants during the beta test, providing long-awaited merchants with comprehensive answers to their questions.

This article will address these questions in conjunction with the content presented at the launch event, providing merchants with centralized answers to their concerns. It will also serve as a window for the industry to understand how to leverage the full-scale launch of the full-site promotion to achieve long-term business growth.

Traffic is the lifeblood of e-commerce, and the upgrade in traffic leveraging capability achieves a "three-way convergence"

As e-commerce competition intensifies, how to sustain long-term operation and precisely reach consumers while achieving high conversion rates at low costs have become two focal points for merchants.

The earliest batch of e-commerce users has been shopping online for over two decades, and most consumers have had a decade or more of e-commerce experience, developing mature and personalized consumption preferences. In the fragmented information flow, consumers can quickly discover products they like and place orders, shortening the purchase decision-making process. Consumers' needs and consumption paths are evolving. Compared to traditional traffic logic, AI's precise matching capabilities can more quickly, accurately, and effectively match supply and demand. As consumer satisfaction improves, merchants welcome more certainty in their business, while platforms gain user loyalty and stickiness. This "three-way convergence" ecosystem fosters symbiosis and mutual benefit, ensuring success both now and in the future.

Taobao has stimulated new consumption vitality by continuously optimizing its traffic environment. Currently, Taobao's DAU has exceeded 500 million, making it the largest traffic pool in China. It can be said that not only is Taobao the largest traffic pool in China, but merchants here will also directly reach the largest and most marketing-valuable audience in the country.

Merchants' primary concern is how the full-site promotion represents a disruptive change in traffic logic compared to the past. Under the paid-free linkage mechanism, the binary structure of paid and free traffic that was once distinct is now bridged by the full-site promotion, becoming a "new lever" for merchants to leverage the overall traffic and increment within the platform.

In the past, merchants needed to use paid models to acquire incremental traffic while also planning and adjusting free traffic to make their products popular. By bridging these two, the full-site promotion not only enables merchants to obtain precise paid traffic but also guides more natural traffic through algorithm optimization.

Qiguo, the person in charge of Alibaba Mama's full-site promotion product, revealed, "The full-site promotion's paid-free linkage mechanism breaks down barriers between natural and paid traffic, fully integrating all Taobao traffic, including our core, high-quality search and recommendation resources. This activates a larger, more comprehensive traffic pool, sourced from various aspects."

It is understood that the full-site promotion covers Taobao's core, high-quality search and recommendation resources, activating a larger overall traffic pool. The full-site promotion explores traffic based on current Taobao activities, resulting in daily changes in traffic structure. For example, during the Double Seventh Festival, consumers are primarily active in the festival activity module scenario. The full-site promotion will then primarily allocate traffic to this scenario through algorithms.

Figuratively speaking, merchants used to operate in two separate ponds, but now they are directly in the ocean. With the ocean's waters available, how can merchants navigate smoothly? The full-site promotion relies on Alibaba Mama's self-developed LMA large model technology, which is based on pre-training with ultra-large-scale multimodal representations. This technology deeply integrates product texts, images, videos, and other information, and fine-tunes the model using over a billion users' behavior data spanning more than a decade, enabling precise prediction of consumer needs.

With the full-site traffic as the foundation and the LMA large model technology's support, potential products can quickly acquire traffic through the paid-free linkage mechanism, achieving deterministic growth in ROI.

As August marks a new peak in summer consumption, coinciding with the Olympics, 88VIP Member Festival, Double Seventh Festival, and other holiday nodes, the full-scale launch of the full-site promotion takes advantage of this momentum to introduce a new merchant support plan. In addition to providing billions of additional traffic, new merchants can receive up to 40% extra natural traffic boost by selecting the recommended bid starting from the preheat period.

Furthermore, a tiered benefits plan is offered to merchants, with larger investments resulting in greater rebates. Brand merchants can receive up to 3.8% rebate, high-potential merchants up to 3%, and small and medium-sized merchants can receive up to 2.5% rebate with low thresholds.

The large model helps merchants tag "good products," leveraging the current Olympic & Summer "festival promotion" bonuses

With the convergence of the Paris Olympics, summer consumption trends, 88VIP Member Festival, and Double Seventh Festival in July to September, millions of merchants are presented with new opportunities. The timely full-scale launch of the full-site promotion brings "certainty" to merchants.

In the context of the industry's relentless pursuit of new growth, no merchant can deny the allure of certainty.

According to QuestMobile's 2024 China Mobile Internet Spring Report, in March 2024, the total usage time of mobile internet users reached 203.96 billion hours, with mobile shopping ranking third in growth contribution at 10.7%. The average daily usage time per person was between 20 and 30 minutes, reaching over 40 minutes during promotional periods.

This defies expectations, as mobile shopping ranks second only to entertainment and social media in terms of daily usage time, highlighting the prosperity of China's e-commerce industry. So what do users primarily do on mobile shopping apps each day? Consumers' entire e-commerce shopping decision-making process covers "browsing, playing, searching, buying, and reviewing," with each step generating vast amounts of data.

Therefore, for products with basic data and sales, algorithms can more accurately predict their trends, and for products with good single-item conversion efficiency, the full-site promotion tags these products in the background as "platform-recommended products" or "search-recommended bestsellers." Merchants can complete the setup in just three simple steps: select products, set budgets, and establish target ROI. Even small and medium-sized merchants can leverage this "new bonus."

Under this logic, the full-site promotion offers business growth opportunities for different types of merchants.

First, for brand merchants with a diverse range of SKUs, the full-site promotion currently supports up to 200 plans (with the option to request expansion), enabling multi-tier product bursts. Second, for emerging brands new to various industries, the full-site promotion identifies potential bestsellers, driving brand growth through single-item popularity. Third, entry-level new merchants can also leverage the full-site promotion's simple operation to acquire traffic, achieving rapid cold start and breaking through zero sales.

All products can be divided into four lifecycle stages: "cold start," "growth," "burst," and "steady sales." Merchants need to focus on the product lifecycle stage based on their industry attributes.

From an industry perspective, standard product merchants and non-standard product merchants have different concerns. The full-site promotion offers distinct support for their product growth. For both standard and non-standard products, the platform provides different suggestions for the path to becoming a bestseller for potential products.

Whether standard or non-standard products, the full-site promotion leverages consumer historical behavior data and real-time market trends to shorten transaction paths and improve ROI, enabling brands to emerge faster. With AI's ability to intelligently identify market trends, it can quickly identify potentially popular products within a merchant's SKU, rapidly creating bestsellers through intelligent distribution (wherever target consumers are). Merchants can batch-launch all store products in the background, prioritizing platform-recommended and search-recommended bestsellers. With just one click, merchants can create campaigns in bulk.

As August brings together multiple holiday nodes, the market presents new "dynamics." Following the traditional model of "testing and adjusting" by e-commerce marketers would mean finding potential bestsellers only after market trends have shifted.

The full-site promotion leverages AI capabilities to quickly identify product paths to becoming bestsellers based on past exposure, click-through rates, save rates, purchase rates, and similar industry product data, helping merchants improve operational efficiency, shorten the time to become a bestseller, and quickly seize market opportunities.

With the LMA model's assistance, consumers with different interests, needs, and preferences are precisely matched with their preferred products through AI technology. The platform, as an intermediary that understands both consumer groups and product elements, tags each product genuinely needed by consumers as "platform-recommended" or "search-recommended bestseller." Merchants only need to check the ROI provided by the system and launch campaigns with one click, enabling their products to reach their target consumers.

In the precise distribution of full-site traffic, merchants experience business growth, consumers satisfy their needs, and the platform further strengthens its ecological competitiveness.

Faster ROI delivery: Full-site promotion helps merchants enhance the certainty of traffic acquisition

The article began by highlighting changes in consumer behavior, with shorter decision-making chains indicating improved supply-demand matching accuracy. In the past, merchants' e-commerce marketers tested which products could become bestsellers using various tools and models, but this approach involved significant trial-and-error and time costs. Moreover, paid and free traffic paths were separate, requiring separate management teams.

This mobile shopping ecosystem could potentially "disturb" consumers. When merchants "blindly test" products, some consumers may see irrelevant products upon opening product pages, leading to dissatisfaction with the shopping experience. The certainty of the full-site promotion lies in precisely capturing consumer intent and matching the most suitable products (price, service, quality).

After beta testing, merchants participating in the full-site promotion reported greater certainty in traffic acquisition, specifically faster ROI delivery.

Currently, with the full-scale upgrade, the system recommends three tiers of ROI for tagged products. Following the platform's recommended ROI tier results in optimal efficiency, effectiveness, and certainty in operations. Compared to the industry's two-week delivery cycle, the full-site promotion shortens ROI delivery from 15 days to 7 days, leading the industry in delivery speed.

Time is precious for merchants, especially in the new consumption environment where trends change weekly or monthly. The 7-day ROI timeline is significant, enabling merchants to make the fastest operational adjustments.

What is the specific value of faster ROI delivery? Merchants can start with a small budget and gradually find trending products suitable for their categories before exploring more traffic. By identifying the ROI that meets their profit margins, they can quickly create bestsellers and establish a winning strategy.

To address new customers' initial concerns about trying the full-site promotion, a new benefit—a compensation mechanism (applicable to products tagged by the system)—was introduced in August.

The compensation mechanism benefit is a significant upgrade for merchants in the full-site promotion. Hu Bao introduced, "If the results do not meet expectations within seven days after the investment, we provide a compensation mechanism for eligible cases." Merchants can initiate cost protection claims and receive compensation for excess costs within that period, with a maximum guarantee of 15 days.

Simply put, by protecting merchants from multiple angles, they can achieve their operational goals fastest by following the platform's recommended ROI for campaigns.

When merchants select tagged products for the full-site promotion, they have three options for target ROI ratios, each accompanied by estimated growth data for leveraged transaction amounts. Merchants can also customize their ROI ratio. Based on beta test results, selecting the platform's recommended three ROI ranges results in more stable traffic acquisition. Adopting the platform's recommended ROI can yield up to 40% extra traffic boost.

The system-recommended ROI, based on product historical spending and customer acquisition cost, represents the minimum budget directly related to the final leveraged transaction amount growth.

Xiaoxiandun, a leading brand in the ready-to-eat bird's nest market, is also one of the beta merchants. According to the brand, to run a campaign on Alimama, merchants only need to select target products, set daily budgets, and target ROAS, eliminating the need for frequent bid and audience adjustments. This saves brands a lot of time and effort. More importantly, the ROI after the campaign exceeds the industry average by 300%, making it a powerful tool for merchants to reduce costs and increase efficiency.

Now that merchants have a basic understanding of the full-funnel marketing, some may wonder what types of products will be tagged by the system and how untagged products can enjoy the incremental benefits of full-funnel marketing. Firstly, best-selling products from leading merchants rarely get tagged. For example, Maoren Underwear, with a scale of tens of billions, has thousands of SKUs, and the AI algorithm of full-funnel marketing selects potential products more precisely in the second tier. Secondly, products with insufficient data or competitive advantages may also face this issue.

Therefore, firstly, merchants should confirm that products with recommended tags from the platform are potential bestsellers, indicating their popularity among consumers. Furthermore, if a product is not recommended by the system, merchants can still optimize it through the "Optimization Suggestions" feature provided by the platform's marketing assistant.

In summary, full-funnel marketing enables competitive products to precisely reach consumers with genuine needs in the most effective way.

Conclusion

Currently, Taobao and Tmall have over 1 billion users and millions of merchants. With over 20 years of accumulated consumer behavior data, AI e-commerce will unleash a new era of ecosystem benefits. The efficient and precise matching demonstrated by full-funnel marketing represents the future direction of AI e-commerce.

Throughout the press conference, we learned about the "pay-for-free" traffic logic, product success logic, and faster ROI delivery logic. All of these are the best evidence of the "user-first" and "technology-driven" strategies.

In short, based on the LMA model, full-funnel marketing truly prioritizes user experience. By leveraging data intelligence, it tags the most competitive products on the platform, ensuring the fastest ROI delivery and maximizing both ecosystem experience and business efficiency.

From a merchant's perspective, full-funnel marketing simplifies the marketing process, allowing merchants to focus on improving product quality and service experience. With the support of full-funnel marketing, every product that impresses consumers can quickly become a bestseller.