In a constant struggle to evade the viability challenge of competing solely on commoditised products and prices in highly competitive markets, retailers are looking to increasingly present their customers with solution ideas and inspirations to clear customer missions.  For example, some of my recent analysis has explored Coles and Woolworths’ respective shoppable recipes and meal inspiration platforms (https://www.coles.com.au/recipes-inspiration) as both brands look to support customers in solving the feeding the family mission.  Similarly, Ikea USA has an Ideas & Inspiration platform on its website (https://www.ikea.com/us/en/ideas/), where customers can see Ikea shoppable curated solutions (see first image below) for transforming different rooms across the house (i.e., a workspace).  Interestingly, Ikea has taken this further by showcasing shoppable room solutions created by customers, leveraging data from customer posts on its Instagram account (see second image below).

Having your customers become your online solution-inspired salesforce is a powerful ploy to grow social license and legitimacy in highly competitive markets.  It marks a substantial progression beyond the historical online trading focus of product category menus, display pages and descriptions, reviews, and price points!  Notably, the data stock produced from this content opens a raft of AI possibilities beyond what is capable from product purchases and transaction data, including:

#1 The ability to send customers personalised solutions to shopping missions instead of product recommendations.

As customers engage with personalised solutions and inspirational content, an array of learning can occur on the types of solutions a particular customer likes based on their browsing activity and engagement with the content.  For example, customer-curated Ikea shoppable room solutions using Instagram post data allow customers to ‘love’ particular solution profiles.  As data stock accumulates on what solutions customers love, collaborative filtering algorithms can match users to ‘profiles’ they are most likely to desire based on the resemblance of their tastes across the platform with akin users.  It’s similar technology to what has powered dating platforms for many years!

Beyond collaborative filtering, content-based filtering algorithms can match customers browsing individual product pages with all product-tagged profiles within the ideas and inspiration platform of solutions containing similar product attributes. Suddenly, a customer browsing a particular product becomes aware of the possibilities with Ikea solving a higher-order need for them to design a room or create a space, deepening the relationship beyond a shallow product transaction.

#2 A retailer can construct more insightful customer segmentation to inform macro-level decisions concerning offline assets (e.g., store merchandising strategy).

Gathering data concerning how customers like to organise and shape products into broader solutions can be leveraged with k-means clustering techniques to produce far deeper insights on how to merchandise products in-store for the most strategically significant target customer segments.  For example, Ikea’s library of shoppable workspace or study room inspiration may reveal that many customers under 35 upload or love and engage with workspace profiles showcasing a gaming chair, given their desires to work and game in one space.  If such a segment is strategically significant for Ikea to target, this insight may lead to macro changes in Ikea’s store showrooms to merchandise workspaces with gaming chairs.  It also helps customers discover the width of your offer in a unified omnichannel experience to solve their missions, lowering the probability of cherry-picking products from multiple retailers to construct a solution for creating a workspace.

Where to Next?

In the arms race for customer literacy supremacy, over the coming years, I expect exponential growth in the trend towards developing more inspirational and solution-centric online platforms as retailers come to terms with the ability of such tools to accelerate customer learning at scale without any dependence on lag-based customer transaction data.  Additionally, growth in this trend will be fuelled by the power and efficiency of such online content in becoming the new sales force of a brand, especially when curated by actual customers who possess a far greater social license and legitimacy in the eyes of the customer than any brand representative can claim.  As the potential dawns on retailers, we will likely see more aggressive market activity in acquiring such inspiration and solutions content from customers.  For example, loyalty points or gift-voucher incentives to shoppable inspiration and solutions content contributors.  Could they offer cash incentives as they endeavour to somewhat disintermediate the Instagrams of the world as the first point of shopping inspiration for many customers?  A strong case can exist on the ROI of such a pursuit, given the money retailers are already shelling out on social shopping!  Inspire your customers before they are inspired elsewhere and by another brand!

Notice that none of this involves GEN AI.

There is so much hype amongst executives around GEN AI that we must remember the broader capabilities of the entire AI application stack and the power of integrating various applications to work in unison within a coordinated system designed to strengthen competitive positioning.  Sure, GEN AI applications can take some costs out of your business through automating content creation across your website operations or providing automated customer service assistance using Chatbots, but just how much are such endeavours materially lowering customer search costs or raising switching costs to leave your brand relative to your competitors? Businesses ultimately exist to coordinate resources and know-how to solve customer problems.

Any robust AI strategy should be ambidextrous in exploring the potential of AI to augment the value proposition to solve customer problems better while also continually unlocking automation capabilities to reduce costs.  The collaborative filtering, content-based filtering, and k-means clustering techniques I refer to in this article belong to the clustering family of algorithms within the machine-learning sphere of AI without any reliance on Generative AI applications.

We must maintain sight of the power of coordinated machine-learning techniques amidst the current craze and the unquenchable thirst for GEN AI applications by many executive decision-makers!