Please join RichRelevance on October 29 for a live webinar on personalization for today’s omnichannel consumer. Customers expect a seamless experience across all available channels, and this webinar will help you connect the dots between mobile, tablet, online and in-store touchpoints.
Join Doug Bryan, an analytics veteran, and Cristina Hernandez, a Senior Sales Engineer, as they share the fully interactive Personalization Maturity Model to help analyze your needs and lead a discussion on how you can create an optimal shopping experience, resulting in greater customer loyalty and increased sales.
Today, we’re happy to expand our product recommendations capabilities with Advanced Merchandising. Now, you can present product ensemble recommendations to inspire consumer purchase with perfectly matched products.
We’ve eliminated cumbersome manual product mapping, along with the need for your merchandisers to keep recommendations up to date with inventory and trends. You can now leverage existing data within the RichRelevance dashboard to create new recommendation strategies that group products by attributes or compatibility.
Advanced Merchandising makes recommendations more relevant, intelligent and personal. Mobile and online shoppers demand—and deserve—the same personalized attention found in a brick-and-mortar store, where sales associates recommend what shoes and accessories to pair with a fall season dress, or which brand of headphones are most compatible with iPhone 6. Office Depot is successfully using Advanced Merchandising to connect consumers with their perfect match. You can watch this webinar to learn more.
To inspire you, read on for a few ways that RichRelevance customers are already using Advanced Merchandising.
- Complete the look: Fashion complete head-to-toe looks with recommendations based on the latest trends and shoppers’ own personal styles.
- Don’t forget accessories: Recommend compatible accessories and warranties to consumer electronics or office supplies shoppers.
- Discover the long tail: Aid shoppers in discovering relevant new products that rarely surface with top sellers or other commonly used strategies.
Match compatibility using CNET
Through our exclusive partnership with CNET Content Solutions, consumer electronics retailers benefit from CNET DataSource™, a database that contains over five million technology and consumer electronics products that leverage specific product attributes such as key selling features, price points, and technical details. Learn how a top-20 consumer electronics retailer gained a 3% increase in revenue per session using Advanced Merchandising with DataSource™ to recommend compatible products.
Try Advanced Merchandising today
To present perfectly matched recommendations using Advanced Merchandising, get in touch with us now!
Our retail publishers see their highest volume of summer shopping on July 11 and 12. In fact, our retailer trends have shown up to a 22% increase in revenue throughout the month.
The upscale chain strives to make it easier for online shoppers to find products and see recommendations.
Online customers of Barneys New York Inc. face challenges that might seem alien to shoppers who tend to buy from less luxurious retailers such as Amazon.com Inc. and Wal-Mart Stores Inc.
So, you just started as a Sr. Product Manager at an online apparel retailer and “personalization” falls squarely in your lap. Product recommendations are already implemented on the site and are doing reasonably well—but, being the over-achiever that you are, you’ve promised your boss an additional 2% revenue per session lift in the next quarter, and are now scratching your head wondering just how to accomplish that feat. Don’t worry—we’re here to help.
First off, it’s important to understand that personalization isn’t one-size-fits-all. Consumers shop differently across verticals; what elicits conversion in electronics or consumer packaged goods is vastly different from what drives incremental revenue in apparel. With the former, shoppers primarily buy based on features and capabilities. In apparel, inspiration and aesthetic compel shoppers to buy. A shopper envisions herself wearing a jacket, incorporating it into her wardrobe, and a compulsion to buy is born. Fundamentally, apparel retailers must deliver a different set of experiences to drive the buy decision and maximize incremental revenue.
With that in mind, you can meet your goal and impress your boss with these 5 apparel best practices—which you can execute and validate within the next 90 days:
1. Remove price from email recommendations. Unless your pricing strategy is to appeal to bargain buyers, do not display price in email recommendations. Price can deter clicks by introducing another dimension with which to evaluate a product. With price exposed upfront, the customer must make a decision as to whether she can afford the item. Remove the price and you mitigate customer objection—particularly on high-ticket items. Instead, dazzle the shopper with product imagery and let them view price once they click through to the site—where they’ll have the opportunity to absorb other salient information that communicates value and endorses the offer. We’ve observed up to a 40% increase in click-through rates after removing prices in apparel email recommendations.
2. Recommend higher priced alternatives on the item page. Influence AOV to drive up per-session revenue. You may take a hit on conversion, but the increase in order values makes this a winning proposition. Start by testing the status quo (the control) against boosting the display frequency of similar items priced at least 85% the price of the product on the page. Or, to improve your chances of identifying the proper threshold, conduct a four-way test, adding treatments that boost items priced greater than 90% and 95% the price of the seed, respectively. It’s important to boost rather than only recommend these items as the latter can result in no products being available for display.
3. Strategically place recommendations to keep external search traffic onsite. You spend a lot of money on SEO and SEM, but you’re still seeing high bounce rates from traffic arriving on item pages from external search. Rather than hoping that the lone merchandised product will resonate with the shopper, provide more relevant opportunities to engage with your catalog. To achieve this, display recommendations for similar items across the top of the page, only when the shopper lands from external search. Across a number of sites, we have been able to reduce bounce rates for external search traffic—landing on the items page—from 5%-20%.
4. Apply user segmentation to recommendations. More than most verticals, preferences of apparel shoppers vary widely based on geo-location, wealth, gender and other identifiers. Whereas a shopper in Minneapolis might be cross-sold a fashionable coat for a pair of pumps, a shopper in Miami might find a light sweater more suitable. “People who bought this item also bought” should then effectively be “People [in Florida] who bought this item also bought.” If you segment your customer base, instead of using aggregate shopping behavior to inform personalization, it’s best to leverage these segments in product, content and promotion-based recommendations. By modeling recommendations off of sub-communities to which customers belong, you deliver greater relevance throughout the shopping experience.
5. Personalize your category list pages. Category list pages—as many exist today—offer very little in the way of personalization. They’re oftentimes a long matrix of products that can be sorted by price, popularity or newness—meaning the number of page versions equals the number of sort dimensions. Clearly, this doesn’t reflect the diverse interests of your shopper base. Demonstrate that you understand the shopper’s preferences by personalizing the product sort order on these pages based upon her historical behavior. If she has an affinity toward a specific brand or newness, proactively rank the results accordingly and accelerate discovery of the right products. Layering personalization on these list pages can increase the per-session revenue of shoppers exposed to these pages by up to 2.5%.
So, there you have it—five optimizations to land you a 2% increase in online revenue and make you a rock star within your organization. If you would like to learn more about these best practices, you can view this 20-minute webinar. For additional questions on implementing these best practices, please contact your Account Manager or email personalize@richrelevance.com.
Malcolm Gladwell recently popularized the term ‘outlier’ when referring to successful individuals. In data terms, however, outliers are data points that are far removed from other data points, or flukes. Though they will make up a small portion of your total data population, ignoring their presence can jeopardize the validity of your findings. So, what exactly are outliers, how do you define them, and why are they important?
A common A/B test we like to perform here at RichRelevance is comparing a client’s site without our recommendations against with our recommendations to determine the value. A handful of observations (or even a single observation) in this type of experiment can skew the outcome of the entire test. For example, if the recommendation side of an A/B test has historically been winning by $500/day on average, an additional $500 order on the No Recommendation side will single-handedly nullify the apparent lift of the recommendations for that day.
This $500 purchase is considered an outlier. Outliers are defined as data points that strongly deviate from the rest of the observations in an experiment – the threshold for “strongly deviating” can be open to interpretation, but is typically three standard deviations away from the mean, which (for normally distributed data) are the highest/lowest 0.3% of observations.
Variation is to be expected in any experiment, but outliers deviate so far from expectations, and happen so infrequently, that they are not considered indicative of the behavior of the population. For this reason, we built our A/B/MVT reports to automatically remove outliers, using the three standard deviations from the mean method, before calculating results, mitigating possible client panic or anger caused by skewed test results from outliers.
At first glance, it may seem odd to proactively remove the most extreme 0.3% of observations in a test. Our product is designed to upsell, cross-sell, and generally increase basket size as much as possible. So, in an A/B test like the above, if recommendations drive an order from $100 to $200, that’s great news for the recommendations side of the test – but if the recommendations are so effective that they drive an order from $100 to $1,000, that’s bad news because a $100 order has become an outlier and now gets thrown out.
In order for a test to be statistically valid, all rules of the testing game should be determined before the test begins. Otherwise, we potentially expose ourselves to a whirlpool of subjectivity mid-test. Should a $500 order only count if it was directly driven by attributable recommendations? Should all $500+ orders count if there are an equal number on both sides? What if a side is still losing after including its $500+ orders? Can they be included then?
By defining outlier thresholds prior to the test (for RichRelevance tests, three standard deviations from the mean) and establishing a methodology that removes them, both the random noise and subjectivity of A/B test interpretation is significantly reduced. This is key to minimizing headaches while managing A/B tests.
Of course, understanding outliers is useful outside of A/B tests as well. If a commute typically takes 45 minutes, a 60-minute commute (i.e. a 15-minute-late employee) can be chalked up to variance. However, a three-hour commute would certainly be an outlier. While we’re not suggesting that you use hypothesis testing as grounds to discipline late employees, differentiating between statistical noise and behavior not representative of the population can aid in understanding when things are business as usual or when conditions have changed.