By Eli Finkelshteyn, CEO and Co-founder of Constructor
How does personalization on an e-retailer’s site help increase loyalty?/ Why is it important for retailers to better understand their customers?
We know that consumers become repeat buyers after a personalized experience with a retailer, and there are two parts to the equation here. The first is just the nature of the market: if you’re better at showing each individual buyer what they find attractive in the moment, you’re going to be more successful every time.
But the second element is the nature of the experience—and the ability to understand how important that experience is for shoppers over time. It’s incredibly meaningful to be able to communicate to a site visitor that you understand them and that you have what they need. Fulfilling needs develops trust over time, and trust is what builds brand affinity and loyalty. This isn’t new—it’s something that brick-and-mortar stores have known for a long time, and it’s about time that online retailers are catching up.
What are some examples of the types of personalization methods e-retailers take? What are some of the common mistakes retailers make when it comes to on-site merchandising?
Ironically, a lot of “personalization” technology is one-size-fits-all because it’s not designed for ecommerce. We see product discovery solutions all the time that claim to offer personalization but still can’t understand how humans actually communicate when they come onto an ecommerce site and search for products. Someone looking for an air conditioner will get more air conditioners as recommendations, which makes sense, but then after they buy the air conditioner, they’ll keep seeing those same recommendations, as if they’re about to start an air conditioner collection. It’s because the concept of a purchase changing behavior is completely logical in the realm of commerce, but doesn’t exist outside of commerce, so one-size-fits-all software can’t properly deal with it. At the end of the day, if the platform isn’t specifically designed to evolve and learn from retail customer behavior and to drive commerce metrics, it’s not worth the investment.
Another pitfall of personalization technology is just displaying what’s globally popular by default and making ecommerce teams build a lot of rules to actually personalize the experience. With thousands of SKUs and millions of data points, the level of personalization that actually influences revenue is more than humans can handle alone. It requires deep data analysis to surface insights first, before either humans or the platform itself can act on them.
The big takeaway here is that personalization is complex because the ecommerce buying journey is complex and unique to commerce, as are the decisions that humans make through every step of it. This understanding is what separates personalization-as-novelty from personalization as a profitable business investment.
Brands, specifically clothing retailers, are competing with fast fashion. What personalization techniques of companies selling sustainably made items ensure they are maintaining customer loyalty and not losing customers to “easier” fast-fashion retailers?
From a tactical standpoint, just about anything can be an attribute that retailers can use to categorize items and help a machine-learning-based system surface attractive products to customers—including sustainability. We’re seeing a lot of retailers adding consciously created products to their catalogs, and if that’s an important feature for customers, it should be easy for them to find those products and for the AI system to figure that out.
But there’s an opportunity here that a lot of retailers aren’t taking yet, and that’s completely personalizing their site experience around customer values and lifestyles. Instead of making customers search and sort for sustainable products, what if you could ask them a few questions in a customer quiz, learn that that’s a really important value for them, and then automatically surface those products first every time they visit your site? This level of personalization is possible now.
We’re suddenly not in a supply-driven market anymore. Customers have a lot of choice, and retailers are competing for every dollar. Customer experience and personalization are key differentiators with a significant impact on revenue and loyalty. It’s important for apparel retailers, even more than others, to really make their experience feel like it speaks to each of the customers they want to shop at their stores.
How can second-hand online marketplaces, which are seeing a resurgence, compete with the search experience on larger marketplaces like Amazon?
Speaking of sustainability, second-hand marketplaces are an arena where this value is playing out in real-time. Artificial intelligence and machine learning can drive personalization in marketplaces with limited inventory (as in the case of second-hand marketplaces) by learning from and across facets, attributes, and metadata. A good product discovery strategy can help surface items that aren’t only relevant to what an individual buyer is looking for, but also attractive based on what each buyer is already clicking on (or not clicking on) and purchasing.
This is especially helpful in resale marketplaces when people might not come looking for a specific item but are browsing categories of products in general looking for a great item at a deal. Regardless of whether it’s the first or second time around, retailers can improve their conversion rates and build brand loyalty by helping their customers find what they’re looking for faster.
About the author
Eli is CEO of Constructor.io. Formerly a Data Scientist and Engineer specializing in Machine Learning, Natural Language Processing, and Distributed Computing, he spent over a decade working on tech problems mostly in ML and NLP. Currently, he works on building search services for some of the largest websites on the internet.
By harnessing real-time, store-level insights from retailers like Walmart, Zuru isn’t just playing in the global toy market – they’re redefining the rules.
Reddit is making strides in the e-commerce realm. Although still in its infancy, the platform has introduced an e-commerce product offering, signaling Reddit’s ambitions to become a more comprehensive marketplace.
71% of workers and 86% of managers want continuous learning and skill development at work
64% of workers and 73% of managers would extend their tenure by six years with improved career support and training, respectively
Deceptive pricing, as defined by the FTC, encompasses any pricing strategy that is likely to mislead consumers, influencing their behavior or decisions regarding the products or services available for purchase. In essence, all forms of advertising, including pricing information, are obligated to convey accurate information and refrain from misleading consumers.