The steady rise of direct-to-consumer retailers, from the scrappy success story of Warby Parker to the unprecedented scale of behemoth Amazon, mirrors a seismic shift in the changing expectations of online shoppers.
“As technology increasingly reduces a market’s cost of entry, a greater choice of business models, services, and products is now available to customers in something of an endless virtual aisle from which to choose.” (KPMG
1
KPMG, Me, My Life, My Wallet
)
Merchants that cling to competing on price and product alone to serve today’s smartphone-tethered consumers, who have the world’s biggest online mall at their fingertips, risk becoming obsolete.
As Macy’s and Toys ‘R’ Us shutter stores, digital-first brands are taking on Amazon in a few calculated ways: They’re verticalizing, so they can take charge of the supply chain and run a lean, agile operation. They’re also forming hyper-strategic multichannel approaches, using online marketplaces exclusively for brand recognition. But the biggest changes these brands are making are harder to quantify: Unable to compete with Amazon on price, selection, or delivery, they are differentiating through intangibles, like customer community and brand values.
By its very nature, Amazon can’t provide the sense of belonging to a brand community or being a part of a curated lifestyle. These emotional experiences influence purchase and retention, attracting loyal customers who seek connection with a brand. Today, “authenticity is a standout consumer value.” 2 Euromonitor International, The Top 10 Consumer Trends For 2017
“Customer-centric technology and data is critical to unlocking a superior customer experience,” according to a research note from Cowen retail analyst Oliver Chen. “Retail needs to engage customers in creative, on-demand ways to be relevant to new generations, and every company needs an Amazon or Un-Amazon-able strategy.” 3 Barron’s Next, Amazon at 20: Is Any Retailer Safe?
Digital-first brands are rewriting the rules of retail marketing via online product reviews, the humble hero of user-generated content. Using “the wisdom of the crowd” as a marketing tactic is a notion that’s taken on heightened meaning in a retail landscape forever changed by a $2.3 trillion global eCommerce market.
Ironically, the very strategy brands are using to fight back against Amazon was pioneered by the eCommerce giant, the original mastermind behind the online customer review. The much-copied feature gave consumers a voice and a digital bullhorn to sound off on their likes and dislikes to a vast Internet audience. This granted them newfound influence over retailers, while helping to redefine what they demand of their shopping experiences.
Today, a chorus of shopper voices, with influencers chiming in, combine to create a new choir of “brandvocates” who wield more sway over shoppers’ purchasing decisions than professional marketers.
It’s a shift that’s particularly pronounced among Millennials, which have replaced Baby Boomers as the nation’s largest buying group, along with Generation Z. These latter are even more likely than their Millennial counterparts to buy an item based on recommendations from YouTube videos and value the opinion of their “influencer” circles — which include family, friends, and bloggers — when making a purchase. 4 Accenture, GEN Z AND MILLENNIALS LEAVING OLDER SHOPPERS AND MANY RETAILERS IN THEIR DIGITAL DUST
The same shift is stripping marketers of their longtime pull with consumers, while elevating UGC-driven promotions — be it product ratings and reviews embedded in a customer email blast or shoppers’ Instagram photos on an eCommerce site, which are now often deemed more authentic and ripe for monetization than traditional advertising campaigns.
Indeed, it’s the shoppers themselves, who are the voice of the brand today. And as troves of UGC are “shared among consumers due to mobile internet and phone cameras, a number of brands are identifying consumers’ own photographs and incorporating them into their marketing material to make it seem more authentic and relatable,” an Euromonitor report said. “‘Less-than-slick’ is not only acceptable, but sometimes desirable.”
5
Euromonitor International, The Top 10 Consumer Trends For 2017
While reviews were a revolution in customer trust when they were first introduced, they’ve since become a staple for every online retailer, from super niche microbrands to big box department stores. A little more than two decades since the first online review, user-generated content has evolved to become a foundational building block for the success of any brand.
In a multichannel shopping landscape, retailers have no choice but to upgrade to next-generation, born-in-the-web technology that delivers granular insights on consumers, what digital-first brands have been wielding all along.
What’s been key to the success of digital-first brands is now a requirement for all retailers as they navigate through a changed retail space, where shoppers who research online are often more savvy than store associates.
Enter the next era of user-generated content
As the retail paradigm shifts from fulfilling — to generating — customer demand, digital-first brands have been at the forefront of harnessing user-generated content to maximize one key edge they have over physical retailers: a direct link to consumers.
Now, as digital-first brands embark on offline growth with physical stores, they’re leveraging UGC combined with artificial intelligence to enhance content analysis and generate a deeper, smarter dive into the conversional chatter surrounding their products.
Personalization, sentiment analysis, and machine learning have changed the way brands and shoppers interact with content. AI now offers an added layer that enhances the impact of customer content, giving brands the ability to collect and display smarter, higher-converting UGC at scale throughout the buyer journey.
In the following chapters, we’ll deep dive into how intelligent user-generated content empowers the new growth paradigm — from acquisition to conversion and retention. AI-enabled user-generated content is designed to deliver higher-converting UGC by personalizing one-to-one engagement with the shopper and capitalizing on the right content in the right place at the right time, keeping step with customers by showing them the most relevant content no matter where they are in the buying journey.
The goal of any pre-purchase research is to get answers. The challenge for brands is figuring out how to provide answers to the specific questions of individual shoppers. To do this, they need to first collect as much information as possible from other customers on these topics, and then present it in an easy-to-consume format. Content is the fuel that makes the engine run. So, how do brands collect user-generated content effectively?
In the beginning, the focus was on getting as many reviews as possible. A product with 500 reviews is more trustworthy (and likely to sell better) than a product with five. Since the primary way of engaging customers post-purchase was via email, sending email requests for reviews became the most popular (and effective) way to capture content.
With the rise of mobile, the most important factor for conversion was making the process frictionless. By analyzing billions of behavioral data points based on simple online interactions, requests could be personalized by time, device, and email client. Time personalization uses advanced machine-learning algorithms to send requests when a shopper is most likely to be active online. For busy shoppers, this proved to make all the difference:
143% uplift
An analysis of 4.5 million post-purchase emails shows a 143% difference in response rate between the best and worst times. 6 The Guide to Data-Driven User-Generated Content Marketing
The next phase was using machine-learning algorithms to automatically optimize stores’ post-purchase requests by device and email client. Since shoppers are constantly juggling between their mobile phones, tablets, and desktops, review requests need to be able to easily adapt to each device for maximum content collection.
But no matter how much review response rates increase with the help of hyper-personalized requests, if the content received is of poor quality, it won’t impact a brand’s bottom line.
By definition, shoppers’ pre-purchase research is typically highly specific, both to industry and individual preferences. A shopper setting out to buy a new mattress, for example, is likely to have questions about firmness, quality, size, and comfort. Someone shopping for a dress may be seeking out information related to fit or size, or they might be concerned about receiving it in time for a special occasion and need information related to delivery time. For all of these shoppers, dozens of reviews that simply say “Great!” or “Loved it!” won’t help.
However, a longer review isn’t necessarily a good indicator of conversion impact. Long reviews that don’t speak to specific shopper concerns are less valuable than their shorter, more informative counterparts. After identifying what makes a customer review effective for increasing conversion — namely information on several topics — the next challenge is how brands can actually collect these topic-rich reviews.
Picture this scenario: After buying a new set of sheets, a customer receives an email asking, “What do you think of your sheets?” Unprompted, they may not know what to say. So brands can try another angle: Asking a specific question, like, “How is the quality of your sheets?”
It’s a good start, but this method has its limits. For example, a customer who had an awesome shipping experience may not think to write about it, because they were focused on a question about quality. Or maybe quality isn’t really what matters to that shopper at all. In order to be truly effective, requests have to ask about topics that matter to consumers. Quality may be the main concern of someone searching for sheets while comfort is important to someone looking for a new pillowcase.
For a bedding brand, making these distinctions around topics at the product level is the difference between an abandoned cart and a successful checkout. After identifying the topics, brands can automatically prompt customers to write about them, using machine-learning to adjust requests in real time within the review request itself.
These real-time prompts, couched inside a conversational interface like instant messaging, imitate the experience of sharing thoughts with a friend, and encourage customers to go beyond the standard responses and give more detailed, helpful reviews.
61% uplift
In initial tests, the AI-powered prompts resulted in a 61% uplift in the prevalence of suggested topics in customer reviews.
For brands in industries with low repeat purchase rates, this is especially important. The average consumer is probably not going to be buying a new mattress every month, so it’s critical to make every review request count.
In the saturated home & lifestyle industry,
Brooklinen faced the challenge of earning
customer trust and building brand fans.
They needed to connect with their consumers, develop a loyal customer base, and understand what they really think in order to continuously improve their products and service. Starting out, Brooklinen only had a few reviews on their product pages. Jack Lorentzen, Senior Associate at Brooklinen explains:
“very early on we understood that we needed a review platform.”
Brooklinen wanted to put a user-generated content engine in place, since customer-centricity is integral to their brand philosophy:
“Always speak to your customer as if they’re your friend.”
Brooklinen began to leverage user-generated content throughout the buyer journey, using product recommendation upsells and coupons in review requests to incentivize customers to leave reviews and make repeat purchases.
For brands like Brooklinen, customer experience isn’t just about satisfaction and support, it’s about building relationships from the first interaction. By commenting back to customer reviews and leveraging community Q&A, Brooklinen builds a two-way conversation with their shoppers.
Getting the right content is only half the battle. The other half is making sure it gets seen. Reviews were historically displayed in chronological order. The addition of filters made it possible for consumers to rearrange reviews by star rating, etc. But that sort of display doesn’t cut it today. Consumers have no patience for unnecessary legwork. Given the abundance of choices, any added distraction will send them straight to Amazon.
With the help of natural language processing, large amounts of content can now be synthesized into relevant, digestible highlights and the most effective content can be automatically identified and displayed. Coupled with personalization, brands can not only show content in the right format, but show the precise content that a specific shopper is looking for. In short, they’re able to better tailor the on-site experience to individual preferences.
Using data on buyer preferences — which products they viewed, added to their cart, and purchased, which reviews and photos they engaged with, what they searched for or which ads they clicked on — brands can selectively display reviews and customer photos that address an individual shopper’s specific concerns about a product.
Additionally, brands can understand on an individual level which products a specific shopper is likely to consider buying and what their personal purchase considerations are. They can recommend relevant products and display customer content that relates directly to a shopper’s specific interests.
For vertical brands in industries like health and beauty, this is game-changing: Beauty shoppers tend to be product enthusiasts who are highly attuned to quality, ingredients, color, and myriad other idiosyncratic details. At the same time, the skin creams, perfumes, and hair products sold are as varied as the customers themselves. So the category is ripe for UGC-driven marketing.
Health and beauty has among the highest conversion rates of any eCommerce industry. At 3.8%, it’s second only to the food & beverage industry. 7 Beauty is in the Eye of the Buyer It’s unsurprising that a vertical propped up by customer communities swapping recommendations on sites like MakeUp Alley or Reddit’s SkincareAddiction is seeing such high conversion rates as UGC becomes commonplace on site.
213% uplift
Beauty shoppers who see UGC on site convert at a rate 213% higher than those who do not.
Using hyper-specific content to target high-intent shoppers who are actively hunting for their “holy grail” products gives beauty brands a powerful way to increase conversion. For example, by connecting that a shopper clicked on multiple moisturizers in the past, a beauty brand can display product recommendations for a moisturizer alongside reviews that discuss how much it helped dry skin.
Founded in 2010 by Chris Riccobono and Aaron Sanandres in New York City, UNTUCKit has perfected the ideal shirt for the untucked man. Alberto Corral, UNTUCKit’s Director of Marketing explains:
“We launched with a single idea: Shirts designed to be worn untucked.”
The brand identified a need in the market for a casually elegant shirt with an emphasis on fit. Despite fierce competition, UNTUCKit upended established workwear brands by addressing a prevalent problem in the market, making its mark as a product-focused company.
The digitally native brand has also found success offline. After opening the first brick-and-mortar location in 2015, UNTUCKit now have dozens of stores all over the US, with at least five in New York City alone.
“Our customer philosophy is that with every engagement we have with our customers, we want to make sure they come away from that experience as true ambassadors of our brand. We want it to be that impactful.”
In order to build a strong community around their brand, UNTUCKit wanted to collect and display more user-generated content from their happy customers throughout the buyer journey. Corral explains:
“When shoppers are looking at user-generated content it definitely helps them make that purchasing decision. This relatable content enables them to look at other shoppers like them and understand: ‘How does it fit them, do I relate to this person?’”
Optimizing on-site UGC for conversion is important, but it only speaks to one component of the modern buyer’s path to purchase. Today’s fragmented buyer journey means it’s common for shoppers to visit an eCommerce site on three different devices, from two different traffic sources more than 10 times before finally purchasing, making cross-channel orchestration more important than ever.
Consumers expect content and products shown in email, and all other digital channels, to be personalized to them. It’s a bit like dating: When today’s shoppers interact with a brand, they expect a personal relationship. If a customer has purchased lotion for dry skin repeatedly and then receives advertisements for oily skincare products, it can be jarring.
Since social networks are increasingly working to show products and brands that suit the tastes and tendencies of their users, a failure to hit the mark on intimate channels like Facebook Messenger can be even more glaring. Just as a friend’s instant message would reflect a basic knowledge of her buddy’s personality and preferences, a retailer’s should, too.
With the rise of micro-segmentation, brands can now get to know and target consumers at an almost-individual level, creating the potential to exponentially personalize the shopping journey. But for emerging brands that need to reach super niche audiences, micro-segmentation is a double-edged sword. On one hand, the granular degree of targeting available today is unprecedented, making it possible to aim at super specific audiences. On the other hand, the same technology that makes this possible depends on a steady stream of high quality, relevant content to run. It’s nearly impossible to create this amount of content for so many audiences, on so many different channels (Facebook, Instagram, Pinterest, email, etc.), displayed in so many different formats (video, text, photo, and more). But without hyper-specific content, micro-segmentation doesn’t work.
This has made UGC the obvious answer. Who better to relate to a brand’s target audience than their actual customers Brands are increasingly soliciting and curating content from their customers to target potential shoppers just like them. It turns out that this content is not only scalable, it’s also super effective. As long as brands continue collecting fresh content, the potential for new, highly individualized ads is virtually endless, giving upstart brands a real chance to compete without breaking the bank for the first time ever. By using AI-optimized customer content in their Facebook Dynamic Product ads, one brand was able to achieve a 72% decrease in cost per click (CPC), 3x increase in click-through rate (CTR), and 78% increase in return on ad spend (ROAS).
3x increase in CTR
Using AI-powered UGC ads on Facebook, one brand increased their click-through rate by 3X.
Luxury sneaker brand GREATS hews to classic design silhouettes, timeless “greats,” so to speak, that never go out of style, according to the company.
Where it does get adventurous is in its collaborations, partnering with local Brooklyn artists, fashion houses, celebrities, and athletes on new products as it experiments with materials and design.
Taking their customer philosophy from the Charles Eames quote: “The best, for the most, for the least,” their customer-first strategy has powered explosive growth on their eCommerce site and now offline at brick-and-mortar locations in Soho and Venice Beach.
But as the direct-to-consumer brand expanded into stores, GREATS understood the need to put customers first throughout their entire brand experience. Kristin Sword, Greats’ Marketing Manager explains:
“To stay customer-first, it’s important to focus on the feedback loop. We make changes both to our product and our website based on what our customers ask for in reviews.”
Having a full-suite user-generated content platform opened up a rich feedback loop on consumers’ shopping experiences
GREATS has reaped tangible benefits from the reviews in both marketing and in tackling customer-service challenges.
“One of our biggest challenges has always been sizing. Not all of our shoes have the same size across the board, which has been a huge pain point for customer service.”
To address this issue, the retailer lets shoppers help other shoppers by publishing customer-created feedback on how GREATS sneakers fit.
“We collect feedback about how our shoes fit and publish those reviews so other customers know how to order their shoes best. This customer feedback loop helps improve our customer service and inform our product offering.”
Product reviews have also served as a jumping-off point for more cohesive marketing across the shopper journey. According to GREATS, “it doesn’t matter how great your marketing is, there’s nothing stronger than somebody loving your product and telling their friends about it.” Since implementing user-generated content, GREATS has seen:
“a significant lift in conversion from the customers who interact with UGC on our site, versus the ones that don’t.”
When the consumer default is Amazon, there’s no longer tolerance for imperfect brands. A staggering 90% of customers aged 18-54 reported that Amazon was their top place to shop, or named it along with other stores. 8 Gen Z’s Need for Speed and How Retailers Can Keep Up It’s a lot of pressure, but luckily, direct-to-consumer brands are uniquely positioned to capitalize on the new eCommerce landscape. Owning the entire value chain — from product development, to customer experience, shipping, and even marketing — means these brands can make changes quickly. But only if they know what needs to be changed.
Customer content provides a constant stream of feedback about thousands of topics. Looking at aggregated star ratings may seem like a quick solution to analyzing mounds of reviews, but it won’t give you the whole picture. A five-star review can contain important requests for improved delivery time, while a one-star review mistakenly written off as “negative” may contain plenty of helpful details that can entice customers to buy. The big challenge is how to actually extract the most useful information hidden in the text at scale.
That’s where opinion mining comes in. By analyzing reviews, brands can identify a range of opinions about topics in any given pool of text. These opinions are scored along a scale of positive to negative using sentiment analysis technology, making it easy to determine trends in attitude and mood. Sentiment analysis is also designed to distinguish between conflicting sentiments about different topics within the same review. For example: “Great product, but slow shipping.”
Tapping into the text from reviews allows brands to decipher shopper sentiment by understanding customers’ opinions in their own words. Unlike satisfaction surveys, which limit the range of responses and fail to provide a holistic representation of unfiltered customer sentiment, reviews often raise unexpected issues that brands may not have considered. For example, a camping gear retailer that uses opinion mining to analyze its reviews could quickly learn that their best-selling tent has a trend of negative sentiment around the unexpected topic of scent.
Instead of having to plow manually through hundreds of reviews each month, customers’ opinions can be grouped and analyzed instantly. Each review can be understood and categorized by sentiment, and placed into the context of collective opinion — a feat once only dreamed of by customer success and merchant teams, and a level of insight only reachable through sophisticated AI tools.
These technologies work together to identify and categorize the opinions that are expressed within the text to determine whether topics are mentioned in a positive, negative, or neutral context. This allows brands to zero in, for example, on the topic “shipping” and to see if customers are saying good or bad things about it in the surrounding text.
Yotpo’s data team used this technology to analyze 3,315,650 customer reviews from the fashion industry. A key example of the understanding gap between rating and sentiment came to the fore on the topics of “size” and “fit.” The sentiment analysis demonstrated that “size” consistently ranks with a 50% lower sentiment score than “fit” in all regions. 9 Quantifying The Importance of Customer Satisfaction This indicates that when people purchase clothing in the correct size, it fits nicely, but that getting the right size in the first place is a challenge. This level of insight, at scale, allows brands to re-engineer product information to ensure the right sizing expectations and bring the “size” and “fit” sentiments much more closely in line with each other.
50% lower sentiment
Sentiment around the topic of “size” is 50% lower than sentiment around “fit” (based on a study of 3,315,650 reviews by US shoppers).
While product improvement might be the most obvious use case for this type of data, customer-centricity doesn’t end at product offering. The brands that have mastered customer experience extend the approach to every touchpoint.
Marketing teams can use trends in consumer topics to better inform messaging and campaigns. They can use customers’ own words and experiences to connect with them. For example, they might find out that a top-selling product is most often purchased as a gift, and change the target audience for ads. Customer service teams can also get real-time insights on concerns or issues and use the information to improve help resources or response times. In such a fast-paced industry, this feedback is crucial for reducing lag time and beating the competition.
At a time when eCommerce competition is so fierce, brands that leverage this cutting-edge technology have a prime opportunity to stand out from the pack.
Disruptive lingerie brand Adore Me’s explosive growth has earned them spots on IAB’s 250 Most Important Direct Brands to Watch and Crain’s Fast 50. Throughout their rapid rise to success, they’ve continued to collect and analyze customer reviews as a major source for product development and improvement inspiration.
Naturally, at a certain point, manually plowing through their customer reviews became unscalable. Sandra Negrea, Customer Engagement Analyst at Adore Me, says:
“It was not unusual to manually read through as many as 5,000 reviews each month to extract customer insights — data mining, running different analyses, and sending reports to the relevant internal stakeholders.”
Recognizing the value of using their customers’ own words to propel product development, service improvement, and even marketing efforts, Adore Me sought out ways to analyze reviews at scale.
Using machine learning and sentiment analysis, Adore Me was not only able to gain an at-a-glance view of the topics that mattered most to its customers, but it was also able to pinpoint critical feedback topics that may have been left undiscovered by reading reviews manually. For example, the retailer was able to quickly identify a faulty clasp in one of its product lines and rectify the issue immediately. Adore Me’s marketing team also found that the word “couples” was a frequent topic raised by customers. As a result, its product team decided to produce an entirely new line focused on this vertical, which turned out to be a major success.
“The secret to our growth has always been, and continues to be, intrinsically related to how well we are listening to our customers,” Negrea says. “It’s important to analyze your consumer feedback efficiently, understand what matters to them most and use this data to inform marketing, product, and customer experience initiatives.”
Digitally native vertical brands are experiencing explosive growth. By leveraging their customer communities, they are able to effectively market to niche audiences with highly personalized content. Their deep attention to detail and individualized buyer journeys allow them to differentiate in a world where most shoppers’ instinct is to turn to Amazon.
But in order to thrive, these upstart brands need to scale everything from fulfillment to marketing quickly, inexpensively, and without losing touch with market expectations. Advances in user-generated content collection, display, and analysis mean that new, smaller brands can, for the first time ever, gain a competitive edge in the eCommerce space.
Fostering a customer community gives brands a bank of content that provides them with marketing creative, on-site social proof, and a wealth of product and service feedback to allow for laser-focused improvement. This not only strengthens emotional connections with customers and inspires trust in new shoppers via authentic content, it also gives new brands the opportunity to scale and focus on customer experience without stretching their budget.
In this new eCommerce landscape, only the brands that are collecting detailed, helpful feedback and authentic, relatable content from their customers will be able to cut through the noise of online retail and reach success.
