Over the past several months, New Distribution has spoken with many FMCG companies. One pattern has become increasingly clear: companies are separating into three levels in the way they use AI.

  • At the first level, companies treat AI as an office productivity tool. They use it to write copy, design posters, organize meeting notes, and draft daily reports.
  • At the second level, companies begin to introduce AI into specific functions, such as customer service, knowledge bases, store inspections, and consumer insight.
  • At the third level, AI enters the operating system of the business itself. It begins to support decisions about where to invest sales spending, how to operate stores, and whether the organization needs to be redesigned—work that used to depend almost entirely on experience, meetings, and managers watching people closely.

Very few companies have reached the third level. The obstacle is not a shortage of AI tools. The industry has not yet been shown clearly how AI can enter the real work of an FMCG business.

That was the question at the heart of the AI Application Forum at the 2026 China FMCG Conference, held from May 26 to 28:

How can AI enter the real operating scenarios of the FMCG industry?

AI Has Value Only When It Enters the Business

The industry can no longer limit its discussion of AI to whether the technology is usable.

Many companies have improved efficiency by using AI to write copy, create posters, summarize meetings, and prepare reports. But the work that truly determines growth in FMCG does not happen in the office. It happens on the front line.

Are products actually selling through at the store? Is trade spending producing results? Are sales representatives carrying out the required actions? Are dealer inventory and profitability becoming risky? Does the product mix need to change?

These are the real operating questions of an FMCG company.

Brand owners now face more fragmented channels, more complex retail outlets, and greater difficulty in judging the return on commercial spending. Dealers face more products, thinner margins, and heavier management pressure. Retailers face increasingly segmented consumer needs and more frequent changes in operations.

Experience, management reporting, and month-end reviews alone can no longer support high-quality growth.

The industry therefore needs to break AI down into real business scenarios. The value of AI is not to remove human judgment. It is to help companies identify problems earlier, recognize opportunities more accurately, and adjust actions sooner.

AI can work with sales reports, inventory tables, spending records, store data, product operations, and organizational knowledge. It can turn fragmented data and scattered experience into operating capabilities that can be analyzed, monitored for risk, and reviewed after action.

This leads to three practical questions:

  • Which real problems should AI solve first in an FMCG company?
  • Where should brand owners, dealers, and retailers begin?
  • Which scenarios are ready to implement now, and which capabilities must be built first?

Only when these questions receive specific answers will AI move beyond concepts and tools and enter the operating reality of the industry.

Three Paths into the FMCG Operating System

The 2026 China FMCG Conference AI Application Forum was not designed simply to place AI experts, FMCG companies, and technology providers in the same room. It organized the subject around three paths grounded in real business work.

1. From Tool Use to the Operating System

Most early uses of AI solve isolated efficiency problems: writing a paragraph, generating an image, or organizing meeting notes. FMCG companies need AI to enter more central operating chains.

That includes judging market opportunities, identifying high-potential stores, optimizing commercial spending, supporting channel management, and making internal knowledge and experience easier for the organization to use.

The question is not whether the company has added another office tool. The question is whether AI can become a new capability across growth, sales, channels, products, and organizational coordination.

2. From Brand Owners to Dealers—and into Real Scenarios

For brand owners, AI can help the company reconsider how growth is created.

Growth used to rely heavily on channel expansion, increased spending, and the replication of model markets. In the future, companies will need more precise answers: Which regions deserve investment? Which stores have greater potential? Which spending actually produces sell-through? Which front-line actions need to be corrected immediately?

For dealers, the starting point is more concrete.

AI does not have to begin with a complex system. It can begin with the sales, inventory, spending, customer, and daily-report spreadsheets already used every day. It can help owners understand operating data and identify inventory risk, weak sell-through, changes in gross margin, and gaps in team execution.

3. From Understanding AI to Using It

One of the industry's biggest problems is that people leave an event excited by what they heard, then return to work without knowing what to do next.

The conference therefore combined the main forum with cases, scenario analysis, and role-specific practical training:

  • Zhao Bo's AI executive workshop on May 26 helped managers establish the foundations for using AI.
  • The main forums on May 27 and 28 used industry cases and scenario breakdowns to show how far FMCG applications had progressed.
  • Two advanced courses on May 28 addressed brand owners and dealers separately, translating AI into specific operating actions for each group.

Together, these three paths brought AI into the scenes where growth, channels, operations, and organizations are actually managed.

Three Courses for the Core Operators of the Industry

The forum was accompanied by a learning path that moved from basic use to hands-on work for specific roles.

Zhao Bo's AI executive workshop opened the program on May 26. It focused on the daily work of sales managers: processing documents, building personal and team AI workbenches, using AI skills, coordinating tasks, and applying AI to the analysis of operating problems.

On May 28, two advanced courses focused on the two central operating actors in FMCG: brand owners and dealers.

For Dealers: Start with the Data You Already Use

For a dealer, AI should not begin as a complicated system. It should begin as an operating tool that people can actually use.

Dealers work with large volumes of spreadsheets and records every day: sales, inventory, spending, customers, field reports, and store sell-through. Historically, most of this data has been recorded, summarized, and reported upward. Relatively little of it has been converted into operating judgment.

The dealer course therefore started with familiar work:

  • analyzing sales data to find products with abnormal movement;
  • using inventory records to detect overstock risk;
  • identifying priority stores from customer data;
  • converting field reports into management reviews;
  • helping owners and managers understand operating problems faster.

For dealers, the value of AI is practical: less judgment based only on intuition and more support from data; less repetitive preparation and more discovery of problems; less recovery after a loss and more warning before the loss occurs.

For Brand Owners: Connect AI to Growth Management

For brand owners, AI needs to enter growth management and organizational capability.

Learning to operate a tool is not enough. Brand owners must answer more systematic questions: How should high-potential stores be identified? How should spending be optimized? How should front-line execution be tracked? How should regional markets be reviewed? How should the operating condition of dealers be assessed? How can head-office strategy reach the store more accurately?

The executive course for brand owners examined how AI can enter these core operating chains.

Commercial spending, for example, has traditionally been reviewed after the money was spent. Could companies monitor it while it is being deployed and adjust in time? High-potential stores used to be selected largely through regional experience. Could data and models support that decision? Front-line execution used to depend on layers of reporting. Could AI provide more immediate diagnosis and reminders?

AI will move beyond headquarters, systems, and abstract concepts only when both brand owners and dealers can use it in their real work.

The Change Is Larger Than a New Tool

AI will not enter FMCG through one company, one department, or one tool.

It will gradually enter the growth decisions of brand owners, the operating management of dealers, the customer operations of retailers, the solutions of service providers, and the daily work of front-line salespeople.

The real change is not that a company uses several more AI tools. The industry is beginning to reconsider its entire operating logic:

  • How should growth be judged?
  • How should channels be managed?
  • How should products be operated?
  • Where should commercial spending go?
  • How should organizations coordinate?
  • How should experience be captured?
  • How can front-line actions be seen, improved, and replicated?

These questions once depended mainly on experience, manual work, and information passed through organizational layers. Increasingly, data, models, knowledge bases, and AI agents will reconnect them.

That is the larger significance of bringing AI into FMCG. From brand owners to dealers, from retailers to service providers, from operating judgment to front-line execution, the goal is to move AI out of the realm of concepts and trials and into the work where the business is actually run.

The industry needs to understand AI again—and, through it, understand the next stage of FMCG growth.