September 14, 2022
An opinion piece by the CTO of Red Ant, a retail technology solutions company
In today’s on-demand economy, customers expect to engage with retailers for a range of queries – such as delivery tracking, order changes, returns or complaints − at the click of a button and on the channel of their choice. Interactions must be seamless and personalized as customers pivot from smartphone to online and chatbot to human advisor.
But success is about more than being reactive. The real revenue opportunity lies in reaching out to engage customers wherever they are, using dedicated technology such as AI and data analytics to engage their interests, make a sale and assess post-purchase experience to cross- and up-sell further.
There is a clear role for AI and analytics in optimizing omnichannel customer journeys. But while many brands understand they need to embrace technology for success, 25% of retailers are also confused about which technologies to adopt, according to a recent Red Ant survey.
Opportunity and hesitation
Retail tech team leaders need to ensure they are making the most of what they already have – their data – as well as proven digital technologies for optimum business agility and performance. But there is a clear need to understand what data is needed, what makes data good, and how to gather it. It is not about just having as much data as possible.
With AI-driven insights, retailers can truly understand their customers and what makes them tick to boost customer loyalty and increase revenue. Data-rich retailers can use AI to improve search and recommendations for customers, enabling them to find the exact items they want as quickly as possible.
This is the secret to hyper-personalization − which consumers are increasingly expecting − and to support sustainable retail strategies that are snowballing in priority for retail leaders. Using AI, retailers can ensure brand consistency across all channels and create immersive experiences for employees, buyers and partners.
These smoother, better and differentiated experiences will encourage more frequent visits, increase loyalty and can even drive customers in-store rather than online, if this is the goal. Other key applications for AI in retail include price and promotional optimization, in-store or on-shelf availability, social media monitoring or sentiment analysis, demand forecasting and even fraud/threat detection, allowing them to efficiently manage in-store activity and make changes that are relevant to both the customer and the business.
Although retailers are keen to buy into AI, reasons behind their hesitation include the following: AI is expensive, retailers need highly skilled professionals, and they need to understand their data in terms of quality rather than quantity. Some may also be stuck in a purchase cycle and others may have spent four or five years getting their data into shape to be accessible for that system.
The majority of the time, they are not holding back because they are worried AI cannot deliver proven ROI. It is more that they do not understand how it fits into their business and how to deploy it. The idea of a trial system is not really possible in a small sample size – it requires a large quantity of data and a long period of time to evaluate the real success of a project.
How to optimize customer service journeys
So how can retailers get started with AI to optimize customer service journeys and transform their retail experience? AI can help with data quality. It uses mechanical tools to spot relationships and patterns in data that a human could not and build a business process to make sense of this pattern.
There are two main areas for deploying this in a retail environment:
Retailers can offer good product and service recommendations to customers based on data and AI developing feedback mechanisms to allow these predictions. For instance, if a retailer notices a customer always buys perfume at a quantity of 50 ml, it can use this information to recommend another brand.
Traditionally, AI in retail has acted as a product recommendation engine – but this is highly reactive and it is more effective to use it to drive the customer either in-store or online and encourage a purchase. You have to train a system on whether a suggestion is good or bad and what that looks like by putting in feedback mechanisms; specialists in data transformation can help with this.
Next best action
AI can also be used to surface a ‘next best action’ – identify customers, their propensity to respond positively to outreach, and also how best to engage them. A retailer can use AI to get the customer to respond to something, for instance going into the store or inviting them to an event. AI can recognize the habits and behaviors of a customer to prompt them into action using a set of rules that determine how they are likely to react. It learns from tactics of the best store associates and translates this to guide other store associates. This can be applied to marketing campaigns, too.
Training the system
AI can leverage performance and data from the retailers’ best salespeople. The key is to ensure the data system has good feedback mechanisms and enough volume of ‘trial and error’ interactions. Giving AI feedback and telling it if it is right or wrong is paramount to its success and the reliability of its recommendations.
As an example, a retailer may decide to reach out to customers after day 3, week 2, week 4 and day 100 of their purchase — and provide a list of people for store associates to contact. The customer dashboard gives insight into the next actions to take, for instance reaching out to a customer that has not made any purchases for a length of time.
If the AI system recognizes that this customer usually makes a purchase at a certain time of year, it can help with making a context-specific set of actions for engagement. If this is carried out in combination with another human being − a store associate − this comes across as a personalized interaction to the customer. The email system is the vehicle for engagement, and AI is the driving technology behind making it happen.
How to measure success
You can tell AI is successful at a basic level if the outcome results in a purchase. A retailer can track the effectiveness of their AI with high levels of detail, enhanced by the quality of its feedback mechanisms. If the AI is given feedback on whether the purchase was successful, it can then help the system to learn – it is a cyclical process. Telling AI that it made a good recommendation is the secret to making intelligent decisions that will impact customer loyalty and the bottom line.
Bringing all the insights into one place on a customer analytics dashboard gives retailers the power to monitor and analyze activity online and on the shop floor with smart, actionable insights that drive store performance. These can be further customized for different user levels, so store associates, managers and the head office can easily access retail analytics that are most relevant to their roles and responsibilities.
Despite the vast potential of AI technology, implementing it in today’s complex retail environment remains an intimidating undertaking. However, it is a necessary task to ensure retailers can compete and survive in the increasingly challenging digital and economic landscape and build resilience under uncertainty.
To get closer to customers and stay competitive, retailers must adopt an omnichannel strategy that harnesses data from all customer channels, both in-store and online. They also need a strategy to use and train AI to interpret the data around each recommendation, use this to engage with customers according to their preferences, with relevant and consistent messaging using data-empowered store associates to guide them through the process.