7 generative AI strategies for in-store management without replacing humans
Marketing and e-commerce specialists said generative AI can be used as a chatbot and trainer for store managers, and businesses that do not have the tech will fall behind.
When worries about job displacement hit employees in this new era of generative AI, marketing and e-commerce experts reveal ways to leverage the technology’s features for in-store management. If executed correctly, it may improve customer experience and satisfaction.
So, experts suggest using this technology as an ally instead of completely replacing the human touch.
Industry leaders, such as Intrepid Singapore CEO Fan-Ru Meng and AnyMind Group Managing Director for Product Development Ryuji Takemoto, support integrating generative AI into businesses whilst preserving the essential role of human intuition and expertise.
“Tech is a tool to improve the quality of our services alongside a human expert. This is assurance that we wanted to offer, to our team, and each individual, that ChatGPT, automation, or any generative AI is not there to replace human skills,” Meng told the Singapore Business Review.
Varun Sharma, senior vice president for Asia Pacific and Japan at Emplifi, called generative AI a “people enhancer” because these tools cannot thrive without humans’ input of factual data sets.
Change and the unknown may frighten some due to job displacement, but it will be a boon for brands worldwide, said Chris Vincent, chief international officer at Pattern, a marketing firm.
How to use AI
The first strategy to use generative AI is to train store managers or chat managers to improve their chat services with customers, which will also enhance their experience, according to Meng.
Before AI was invented, store managers needed a trainer who would provide a dedicated session with a team. This will be removed because AI can allow store managers to ask hypothetical questions and in turn, will be given the answers to those queries.
“You can also configure that into different tongues and manners from different countries,” said Meng. “In that way, when it comes to the text-generated or text interactions, you will be able to help agents from any part of the world to… respond to the real customer in the type of manner that the customer is coming from.”
There may be different types of interactions or tones, though, when it comes to engaging with customers about electronics or fashion, he added.
Pattern’s Vincent said generative AI can also be trained in multiple languages and allows firms to provide chat support to customers worldwide. Generative AI can also learn from previous interactions, continuously improving its responses and problem-solving abilities, he said.
The second strategy is that Open AI’s invention may also be used to create a question-and-answer format. Takemoto said it could be used to create prompts in customer chats to produce certain actions such as advertising with relevant product recommendations and return orders sent to logistics companies.
The third strategy is for generative AI to be configured into a chatbot to directly answer simple questions from customers.
“What we see in ChatGPT that is different from the current chatbot is that it can process a larger data set right and give you more comprehensive [answers] and create more of a better reasoning to the answer,” said Meng.
For Emplifi’s Sharma, AI-powered chatbots can improve customer satisfaction because it cuts response time and will address concerns urgently. If a brand is not doing this, Sharma said it will fall behind competitors who are actively investing in these tools.
The fourth strategy, Takemoto said, is to create product reviews and user-generated content on social media. In this way, the business will understand the differences from competitors and use them in the marketing plan.
Takemoto said that store managers and chat managers may also use the technology to automatically generate content for advertisements and landing pages.
Sharma shared that their company has an AI tech solution that generates ready-to-publish social media copy for brands. Social media teams can place their instructions on the tool and can select customisations such as tone, emojis, hashtags, or questions.
The fifth strategy is to leverage text and image data, analyze, and explore other data utilisation methods for influencer marketing to drive more potential customers to make purchases, said Takemoto.
The sixth strategy is to use generative AI to create product descriptions to decrease manual inputs from humans, who can focus more on other essential aspects of their work, Takemoto said.
Emplifi’s Sharma cited as an example using AI for personalisation through emails that show products that are appropriate for the target audience or landing pages that offer customers content that is aligned with their needs, interests, and ideals.
Pattern’s Vincent said generative AI can create thousands of product listings with a push of a button. This is more efficient compared to having a large teamwork for a week to produce the content.
“But generative AI could produce it in an hour with the proper framework around it. And that’s just scratching the surface,” he said.
The seventh strategy, according to Vincent, is to use AI models to predict customer demand and optimise stock levels. This provides big help in managing the supply chain and inventory. It would ensure that popular products are always available whilst avoiding overstocking.
No escaping cybersecurity issues
AI technology also has its negative effects such as privacy concerns being raised in news articles all over the internet.
To manage this issue, Sharma advised brands to train their AI models with strict guidelines, feeding them with data that is legally obtained and compliant with general data protection laws.
A 2022 Gartner research showed that by 2025, businesses that use AI across the marketing function will shift 75% of their staff’s operations from production to more strategic activities.
This means amidst all the controversies surrounding generative AI, Sharma said use cases for AI across marketing functions will continue to evolve.
On the side of the consumers, Euromonitor International senior analyst, Quan Yao Peh, said they must be aware of trade-offs in data sharing with brands.
Relying on AI should also need a market specialist to ensure accuracy in the data being collected.
Vincent advised that specialists should be critical on two points: One, making sure the information going into the tool, such as the fact sheet, is accurate; and two, making sure the model is quality tested and refined before leveraging for mass production of assets.
As ChatGPT continues to gain ground in the marketing space, Meng said the one thing businesses will focus on in the future is to stay competitive by “harnessing the potential of ChatGPT and generative AI.”
He also said the challenge is to integrate generative AI with human expertise. “If we would just integrate those, for the sake of integrating but not improving quality, not improving productivity, or doing this with less efficiency,” he said.
Forward-looking, Takemoto stressed the potential for using generative AI in creating base marketing content, such as text and visual content.
“In addition, tasks that require digesting large amounts of data and inputs [from both the business and end-user] and forming an output [action or recommendation], would work well in the various forms of marketing today,” he added.
For Vincent, generative AI, which has about 1,800 versions in the marketplace, can be seen as a “major boon” in the e-commerce platform as it can maximise productivity. “Those that don’t will fall behind. It’s that simple,” he concluded.