Ecommerce Personalization for the Mid-Market Retailer Part I
Part 1 of a 2 part series on ecommerce personalization.
At Gorilla, we see ecommerce is quickly shifting away from one-to-many conversations where retailers can submit massive communications to all customers and expect the best results. Customers today expect personal one-to-one conversations between consumers and retailers like they receive in brick-and-mortar stores. Your business and customers are unique. Your shopping experience should be too. One of our driving mantras is to create authentic shopping experiences that connect brands to the right customers. Ecommerce personalization can be a powerful tool to help retailers speak more relevantly to their customers.
Until recently, ecommerce personalization was only accessible to big retailers who could create their own personalization algorithms in their R&D facilities. What ultimately drove Amazon to the top of the internet retail hill was its ability to recommend products that users would find interesting in dynamic ways across their site. If you searched for a book on traveling to Italy, you might also see other books on Italian cuisine or lightweight travel backpacks. Ecommerce personalization is quickly becoming accessible to mid-market retailers as companies learn to scale complex algorithms into usable elements that make sense for the mid-market, both from an implementation and financial perspective. If you are unfamiliar with personalization, here is a quick overview of what it is and how it can impact your business’ bottom line.
What Is Ecommerce Personalization?
Ecommerce Personalization is a one-to-one interaction or one-to-one many interactions (site versioning for different visitor segments). A one-to-one personalization is custom web pages delivered to individuals based on explicit or inferred inputs. A one-to-many personalization is a finite set of web pages delivered to customers based on how these customers map to predetermined segments. Generalization – or one-to-all – is a single clickstream path or set of items appearing to all customers, regardless of their previously exhibited behavior or intent. 52% of consumers who experienced personalization like it according to a recent Forrester Report.
Personalization matters because shoppers value recommendations. Shoppers are often persuaded by recommendations because it helps them discover products and solutions they might not be familiar with otherwise. 77% of customers say they find recommendations in general extremely useful and 1/3 of consumers report having purchased products based on recommendations they were shown.
Personalization tools are able to drive key metrics such as revenue, conversion, average transaction value, time on site or margins, providing measurable ROI for marketers who have to fight for their marketing budgets.
How Personalization Works
Ecommerce Personalization is driven by shopper inputs, the engine algorithm and the resulting outputs.
• Shopper Inputs are data points implicitly gathered about the customer during their browsing session by observing customer behavior and then evaluated to extract commonalities, associations and cause-effect relationships. Common inputs are the time spent on site, keyword searches, customer reviews, location ID, product attributes, merchant-driven rules, clicks, sales and margin.
• The Engine Algorithm is a separate formula for determining which recommendations are most appropriate in a specific scenario. Common approaches include collaborative filtering, Bayesian reasoning, choice modeling, and simple data mining.
• Resulting Outputs are how marketing efforts are displayed onsite. Simple outputs can be cross-sells, upsells or mboxes (targeted marketing info) appearing on product details pages. Whereas a more complex result will come from landing pages or home pages changing, dependent upon the creation of different experiences retailers create for each cluster of shoppers.
But this is only the beginning of a topic that deserves some deep exploration. In part II of this post, we’ll go into detail about the functional requirements for personalization, the challenges common to retailers and the potential ROI for merchants who get it right.
