The AI Edge: How Top Retailers Are Using AI To Win

Not just a fancy tech tool; AI’s a real game-changer that's streamlining our work and supercharging our innovation - driving efficiency, precision, and scalability.

The panel session at eTail Connect Asia on effectively managing the involvement of AI and Gen AI in revolutionising e-commerce to boost customer engagement, enabling informed business decisions and maintaining a competitive edge sparked further discussion. In this blogpost, the speakers who participated in that session share their insights on leveraging AI to enhance data accuracy, streamline content creation, and deliver personalized customer experiences. From analyzing extensive datasets to generating tailored recommendations, AI is at the core of their efforts to provide high-quality, relevant, and timely content.

As we navigate the challenges and opportunities presented by AI, explore the future of retail without compromising on customer data privacy and trust.

Transforming Workflows with AI

One of the speakers mentioned that their company implement AI based on the type of task rather than by department. Whenever they see an opportunity to boost efficiency, precision, and scalability, they encourage AI-driven solutions, freeing their team to focus on higher-level, strategic work.

Here are some areas AI has made a significant impact on the way they work:

Data Analysis and Validation: AI plays a vital role in processing large volumes of information, ensuring data accuracy, and streamlining content summarisation. For instance, their media team uses AI to quickly analyze extensive datasets, allowing them to identify trends and refine messaging for higher effectiveness across advertising platforms.

Content Adaptation and Personalisation: In content creation, AI enables us to scale adaptations efficiently. Starting with a master copy, they use AI to customise content for various formats and audience segments, ensuring brand consistency while meeting the unique needs of each client touchpoint. This allows them to deliver high-quality, personalized content at scale.

Strategic Refinement and Innovation: AI acts as a valuable tool for refining their ideas and strategies, helping them identify gaps or inconsistencies in their approach. By prompting questions that challenge and strengthen our proposals, AI supports them in making more innovative and reliable decisions, ultimately enhancing the quality and effectiveness of their work.

On the other hand, another speaker mentioned that they generated most of their content using AI and generate it in 1/10th the time it used to take them the last time. They use it for copy, image and video across our marketing campaigns, on the app and website, in our advertising and content such as product descriptions.

Furthermore, they use it to offer a far flexible search for consumers based on intent and using natural language. It is also used to generate personalised recommendations for each customer based on their buying history, cohort behaviour and demographics.

Lastly, for customer service, they use AI to analyse and create dashboards for decision making.

Building Customer Trust with AI

Finding the Right Balance between Innovation, Experimentation, and Data Security

To maintain quality standards, there is a need to establish key benchmarks for accuracy and brand alignment early in the training phase, ensuring the AI understands critical standards from the outset. Additionally, iterative feedback must be provided and common errors have to be categorised, enabling us to refine outputs continually and manually validate them against our accuracy criteria. This approach allows us to harness AI’s capabilities while preserving the quality and integrity of our work.

For example, while AI aids in generating content and ideas, all final campaign materials undergo human review to prevent factual inaccuracies and off-brand messaging.

Another speaker mentioned that they have a robust internal governance process and all ideas and use cases for new technologies using Gen AI are vetted. They have built and bought many technologies that can be Coke Chat GPT.

All output is screened by human supervisors for accuracy, relevance and acceptability. Logs are maintained on all errors and tools are corrected for past errors systematically.

Cultivating Long-Lasting Relationships: How to Master the Power of Predictive AI

AI-driven suggestions need to be genuine and aligned to customers’ interests, as overly profit-driven recommendations can erode trust.

To foster transparency, AI-driven product recommendations can be clearly labelled. In addition, they could provide a brief explanation of why each item is suggested. This approach helps customers understand how AI works and builds comfort with its recommendations. Additionally, offering users the option to indicate that they would not like to receive similar product suggestions anymore, empowers them to personalise their experience, reinforcing that AI is there to enhance their preferences, not dictate them.

A speaker believes this is a non-issue. All algorithms have used AI since they begun. When it comes to paid search or advertising or influenced posts etc, they follow a calculated approach to get the right products in front of the right consumer for various reasons.

ML (which is AI by the way) has always been used in search. In fact, the right implementation of AI in search transfers the power of search to the user and this allows for far higher accuracy and access!

AI Tools and Data Foundations: Addressing Challenges

Integrating campaign data across platforms like Meta, LinkedIn, and Google Ads presented a challenge due to varied data formats and metrics. AKIN addressed this by standardising data collection methods, enabling smoother AI analysis.

Start with the use case and application of AI then work back to the data requirements to enable the technology to do what you need. Don’t start with data, start with the use case.

A speaker stated that they built a common data warehouse and APIs to access information on top of that. This entailed defining data structures that enable capturing of all relevant data, creating tools to capture this data and store it and an ability to access it when needed.

Customer Response to AI in Content Marketing

Generative AI, which generates human-like text and images, are at the forefront of the content revolution. It is directing retailers to leverage its advanced algorithms for content creation. With AI-powered tools, brands can now deliver a steady stream of captivating visuals and personalised content for their target audience, turning them into customers.

Coca-Cola is pushing the boundaries of brand marketing by integrating AI and tying this back to the spirit of 'giving'. They are reimagining iconography to refreshen its brand image of spreading liveliness and joy in the form of consumer-created holiday cards. This mood of merriment ties in nicely with what Coca-Cola stands for and its campaign is a strong example of leveraging symbolism and the figure of Santa Claus to accentuate its core values. They are bringing creativity to another level, while harnessing AI-powered tools to share “Real Magic.”

The integration of AI into marketing and content production processes will help retailers stay agile to market trends and consumer feedback, ensuring that their content marketing campaigns remain relevant and impactful.

One speaker mentioned that the average quality of content delivered through AI plus human supervision significantly improves over a period through iteration. As a result, customers get faster, more relevant information and spend less time getting things done. This shows up in conversion of visitors to customers and in customer satisfaction through surveys and NPS.

Prioritizing AI Use Cases

At AKIN, they prioritise AI use cases by focusing on the areas where AI has the greatest impact on accuracy efficiency, and consistency.

For accuracy, they look at the quality and volume of data available for training AI. Reliable data ensures that AI can deliver precise outputs, particularly for tasks requiring high accuracy, like content validation and maintaining brand tone. Use cases where they're confident in the data’s relevance and AI’s ability to learn nuanced client language are given priority.

Another speaker claims that efficiency guides the company to apply AI to resource-intensive tasks. AI-driven automation in data analysis or content adaptation reduces production time and frees their team for high-level work. They prioritise tasks where AI can significantly improve speed and productivity, offering substantial returns on time saved.

Consistency is essential for long-term projects that demand steady quality and tone. Training AI on specific brand guidelines allows one to maintain uniform outputs over time, especially for recurring content needs. Scalable AI solutions that enhance consistency across multiple projects and clients are highly valued.

As with any other tool begin with the low hanging fruit- essentially this means use cases which involve less technical knowledge and have high impact either in terms of costs or productivity. It is less important to begin with a specific function as opposed to something that can be implemented quickly and demonstrated rapidly. This builds confidence and quick wins which help to scale it to more complex use cases.

Effectiveness of GenAI Content

At AKIN, they typically leverage GenAI to adapt ad copies rather than create them entirely from scratch. They begin by crafting a master copy, while they train the AI tool on the client’s brand voice, tone, and key messaging. Once the AI is finely tuned to capture these specific nuances, it is used to adapt the copy across different formats and audience segments. This approach ensures consistency across our campaigns while allowing us to scale content quickly and efficiently.

In-House AI vs. Partnering with Providers

Partner with companies and brands that reflect the values of your business. Build where you need to own and retain the IP or where there’s not a provider that you believe will support your needs over the term you expect to need the technology.