--- title: "5 Seconds to Influence Purchase Decisions: Building an AI-Driven Model for Retail Terminals" description: "At the FMCG2019 China FMCG Conference hosted by New Distribution, founder Zhao Bo opened with a discussion on digital supply chain applications and practices in the FMCG industry, highlighting deep distribution as an effective solution for many FMCG brands. This aligns with Pinlan's advocacy for an AI-driven retail terminal model, which uses AI and data to manage terminals and influence both consumers and store owners." author: "李剑" publisher: "New Distribution" email: "zhaobo258@gmail.com" telephone: "+8615854817671" published: "2019-08-30" categories: "Distribution & Channels" language: "en" canonical: "https://xinjignxiao.com/en/articles/5-seconds-to-influence-purchase-decisions-building-an-ai-driven-model-fo-3e873476/" markdown: "https://xinjignxiao.com/en/articles/5-seconds-to-influence-purchase-decisions-building-an-ai-driven-model-fo-3e873476.md" original_source: "https://mp.weixin.qq.com/s/O6phQVWdCP1T_lcO9Hbqhg" translation: "https://xinjignxiao.com/zh/articles/5%E7%A7%92%E9%92%9F%E5%BD%B1%E5%93%8D%E8%B4%AD%E4%B9%B0%E5%86%B3%E7%AD%96-%E6%9E%84%E5%BB%BA%E9%9B%B6%E5%94%AE%E7%BB%88%E7%AB%AFai%E9%A9%B1%E5%8A%A8%E6%A8%A1%E5%9E%8B-3e873476.md" attribution: "New Distribution — https://xinjignxiao.com/en/articles/5-seconds-to-influence-purchase-decisions-building-an-ai-driven-model-fo-3e873476/" citation: "李剑. “5 Seconds to Influence Purchase Decisions: Building an AI-Driven Model for Retail Terminals.” New Distribution, 2019-08-30. https://xinjignxiao.com/en/articles/5-seconds-to-influence-purchase-decisions-building-an-ai-driven-model-fo-3e873476/" usage_policy: "https://xinjignxiao.com/ai-policy.txt" --- # 5 Seconds to Influence Purchase Decisions: Building an AI-Driven Model for Retail Terminals > At the FMCG2019 China FMCG Conference hosted by New Distribution, founder Zhao Bo opened with a discussion on digital supply chain applications and practices in the FMCG industry, highlighting deep distribution as an effective solution for many FMCG brands. This aligns with Pinlan's advocacy for an AI-driven retail terminal model, which uses AI and data to manage terminals and influence both consumers and store owners. At the FMCG2019 China FMCG Conference hosted by New Distribution, founder Zhao Bo opened with a discussion on digital supply chain applications and practices in the FMCG industry, highlighting deep distribution as an effective solution for many FMCG brands. Managing highly dispersed small stores and achieving refined terminal layout... These ideas align perfectly with Pinlan's advocacy for an AI-driven retail terminal model. Using AI to drive sales growth and leveraging data and AI capabilities for terminal management is the core of this terminal-driven model, because both consumers in front of terminal shelves and store owners managing small shops are people who can be influenced. Since 2016, two major events have occurred in the Western world: businessman Trump was elected the 45th President of the United States, and the UK voted to leave the European Union. A few days ago, the American documentary "The Great Hack" was released, which, through the use of consumer data in the 2018 Facebook privacy scandal, revealed the data company behind the "Trump victory" and "Brexit"—Cambridge Analytica—and its key figures, all presented in visual form. In the film, Cambridge Analytica proposed a "behavior change" model: "Behavioral psychology + Big Data + Precision targeting = Behavior change." This model was proven successful by the two major events mentioned above. The film also detailed a case study from 2013, where in the general election in Trinidad and Tobago, Cambridge Analytica helped the Indian party win. Through a series of "Do So!" behavior change campaigns, the election, which had been evenly matched, lost its balance, and the Indian party achieved a decisive victory, entering Congress. These facts show that consumers can be influenced. But how can we influence consumers? Deloitte, a renowned consulting firm, pointed out in its article "Crossing the Digital Divide: How to Respond to Chinese Consumers' Digital Demands" that the interaction between consumers and smartphones varies across different scenarios and shopping stages, as shown in the figure below: In the demand generation stage in stores, it is the only scenario where consumers are not influenced by their phones. However, when we carefully consider the in-store shopping scenario, the time consumers spend making purchase decisions in front of shelves is very short. According to statistics, a person spends an average of 15 minutes in a store, 15 seconds in front of a product category, and 75% of people make purchase decisions within 5 seconds. 40% of people will switch to another product if they cannot find the product they want on the shelf. To better capture consumers and encourage them to buy our products, the management of promotional content and display in terminal stores must achieve the following: Creating a "5-second win" power in terminal displays to influence purchase decisions is the core demand for winning the shelf consumer battle. The battle for terminal consumers is like an election; only products with the ability to "win on the spot" can gain consumer favor. This requires those managing terminal stores to have refined management and standardized execution capabilities. Some may say, "I just need to sell goods to the store owner; display is not my management focus." But the profit principle tells us that whether it's the brand, distributor, or small store owner, only by selling more goods can they obtain more "absolute profit." Efficiently responding to consumers enables selling more goods, improving capital turnover, and increasing total profit. Earlier, we mentioned that consumer purchase behavior and small store owners' purchase behavior can be influenced. Brand owners, distributors, retailers, and small store owners all need to build an effective growth-driven model at consumer touchpoints: "Behavioral psychology + Big Data + Precision targeting = Behavior change." From a data perspective, e-commerce data is easy to obtain, but terminal shelf data has always been outside the data and digital system. Traditionally, enterprise informatization capabilities have focused on transaction data such as inventory, purchasing, and sales, and financial data. The "perceptual" data of displays has been difficult to obtain and digitize. With the development of AI technology, it is now possible to obtain shelf display data, promotional posters, and price tags through computer vision technology. But how can we obtain this data in a cost-effective, labor-saving, and time-efficient manner? Pinlan, in collaboration with Tencent's WeChat Work department, has developed and launched a flexible digital AI model for data acquisition, terminal display, and promotional execution management. Leveraging WeChat's super-connectivity for "people, goods, and places," this model can quickly connect the data relationship between "goods" and "people" through product recognition. By using the WeChat mini-program to take photos of shelves, special displays, posters, and price tags with a phone camera, it automatically generates the "perceptual" data of terminal stores needed by enterprises. This is an AI-driven terminal digital system that conforms to deep distribution management. Built on Pinlan's AI item recognition platform "Pinshi," and with the help of the mini-program "AI Store Inspection," it expands the scope of shelf data collectors: anyone with WeChat, including brand sales representatives, distributor sales representatives, shopping guides, third-party shopping guides, supervisors, store owners, and shelf stockers, as well as crowdsourced workers taking orders via mobile, can achieve AI digital store inspections. The precise SKU recognition of the Pinshi platform can automatically generate data for terminal shelves, end caps, pallets, and special displays, and automatically review and issue rewards. AI capabilities are also reflected in promotional management and task assignment. Some enterprises use this system to combine shopping guide fan acquisition with later promotional management. Shopping guides add fans on-site, and later, through automated operations management via WeChat mini-programs, achieve AI-driven activation marketing. AI item recognition becomes a "trigger" for the following activities: precisely pushing personalized promotions for each store and each shelf to consumers based on products on the shelves, truly achieving a closed loop from promotional execution, fan conversion, to repurchase activation. Use promotional management content on posters, shelf stickers, and product packaging. Directly use shelf product recognition to make "instant" decisions on promotions. There are so many scenarios that it feels very cumbersome and complex. To enable consumer goods companies and distributors that are just starting their digitalization to use it quickly and effectively, Pinlan has summarized a simple and practical methodology for quickly building a deep distribution AI terminal-driven model based on experience with clients. Here it is for sharing: This model is built on modern "OKR" goal management, with market-driven as the core, helping enterprises quickly establish a precise growth model for terminal stores that includes data collection, AI intelligent recognition analysis, and AI automated execution review. Quantify management goals, precise to each store, so that organizational personnel know what to do and how to do it. By coordinating with brand and corporate market strategies, dynamically analyze promotional management, task assignment, and shelf execution effects, building dynamic insight capabilities for terminal stores. Help manage all store terminals, forming promotional management that adapts to each store and shelf, "responding to circumstances." Treat consumers as part of digitalization, and through the combination of products and automated marketing, achieve "conveying feelings and intentions" to convert purchase decisions. The foundation of digitalization is the details of precise execution management. In simple terms, it means making all actions and manageable touchpoints clear, accurate, quantifiable, and standardized. Then, through standardized transparent execution management, every unit and person in the execution chain can know how to do it, when to do it, and do it correctly and well. Enterprises already using SFA, CRM, etc., can integrate these functions into existing systems through software development kits such as SDKs and APIs. Enterprises without such management software can quickly deploy and implement management through AI Store Inspection on WeChat. AI Store Inspection charges no software fees. Enabling customers to use it and achieve real performance growth is the common mission of Pinlan and its clients. Based on the number of users, Pinlan charges a fixed annual service fee and works with clients on the project's operations management for one year. Best practices will be provided in the form of tool model packages, regularly upgraded for free to users, allowing customers to continuously iterate and upgrade their AI operations management. 5 seconds is the time consumers make purchase decisions at the shelf. Scan the QR code now to download the solution details, build your own digital AI terminal-driven model, and win the battle for terminal consumers. About the author: Max Li, Co-founder/COO of Pinlan, omnichannel marketing expert with 18 years of marketing management experience in global Fortune 500 companies; founded multiple brands and won the Global Commercial Space Award; certified marketing management instructor by Wilson Learning. --- ## Citation metadata - Publisher: New Distribution - Author: 李剑 - Published: 2019-08-30 - Canonical: https://xinjignxiao.com/en/articles/5-seconds-to-influence-purchase-decisions-building-an-ai-driven-model-fo-3e873476/ - Original source: https://mp.weixin.qq.com/s/O6phQVWdCP1T_lcO9Hbqhg ## Copyright and AI use This article is sourced from New Distribution. Search, quotation, summarization, and model training are permitted, but every use must credit New Distribution and retain the canonical source URL. Contact: zhaobo258@gmail.com · +86 158 5481 7671