A Deep Dive Into Lemon8, ByteDance's New Social Commerce App That Could Challenge Instagram
Lemon8 is an exact replica of Little Red Book, one of ByteDance's main competitors in China and the go-to destination for gen-Z.
ByteDance just released a new app, Lemon8, that is an exact replica of one of its biggest competitors in China: Little Red Book (XiaoHongShu). Little Red Book (RED) is a social commerce platform that is capturing the attention of gen-Z, who are looking for more useful content in their everyday lives and who think Douyin (China’s TikTok) consists mostly of mindless entertainment content. RED is one of the fastest growing apps in China and has grown from 30M DAU in mid-2020 to over 80M DAU and 200M MAU by 2023. In this post, I will share a deep dive into RED’s business, product, whether something like it can exist in the West, and what challenges RED and Lemon8 face.
First, a brief history of Little Red Book:
RED launched in 2013 as an overseas online shopping community. As China’s economy expanded in the 2010s, there was a rise in Chinese people traveling to the US and Europe. During their trips, Chinese visitors would buy Western products and bring them back to China, leading to the popularization of these brands. RED saw an opportunity to help these shoppers by providing them with shopping guides and an online forum where users would share photos of products they purchased, along with long, written reviews. RED built a product recommendation community made up of young, affluent women who were fans of Western brands.
A year later, RED saw an opportunity to become a cross-border e-commerce platform and shifted its business towards selling products in-app. In other words, it shifted away from being a community platform and towards being more of an e-commerce platform. This change in direction, however, was not successful for the company and in 2017 RED switched its strategy back towards being a community platform at its core. With this shift, the company also changed its business model to rely more heavily on monetizing its community via ads rather than monetizing through e-commerce. The shift towards a community platform monetized through ads became a major breakthrough for the company and was the beginning of its explosive growth in 2020.
Lessons from RED’s early days:
Every social commerce platform has to decide whether it is a social-first or commerce-first company. I believe that many startups in the West lean too heavily into commerce - “how do we help creators sell products and make money” - and ignore social aspects entirely. The company that creates social value for its creators and helps them make money will be the one that breaks through the noise.
Being a commerce-first company often takes away from the trust and authenticity of product recommendations. When a company is primarily reliant on driving GMV, incentives shift towards selling. As a result, creators become salespeople - they are modern-day product pushers. The beauty of RED is that it promotes and emphasizes authentic recommendations from everyday people who are sharing content because of the social value - likes, followers, community - they receive rather than income from recommending products. By being social-first, RED has built a successful business via ad revenue while protecting the trust and authenticity of its community recommendations.
Monetizing as a commerce-first platform is very difficult. If the platform can take a 3-5% fee on GMV, it means it would need to drive at least $2B in GMV to generate $100M in revenue. Instead, by focusing on ad monetization, RED is already generating over $800M/year in revenue. Commerce revenue can be a great addition to a social commerce platform, but ads is a much more lucrative business.
An overview of the business:
The easiest way to describe RED is as a combination of Pinterest’s UI with the trust and authenticity of Reddit communities. While it might seem like RED is a social app at first glance, it really is more of a community platform that helps users discover recommendations from people like them. RED is the photo/video version of Subreddits like /femalefashionadvice or /skincareaddiction. It is the place that shoppers visit to decide what to buy and whether something is worth buying.
In the buyer’s journey (discovery → research → purchase) RED has excelled at solving the research phase. By focusing on this middle step, it has become the link between discovery and purchase. This is the gap that Lemon8 is trying to fill, along with Flip (focused on beauty/skincare) and iykyk (the company I have been working on, focused on gen-Z fashion).
To solve the research phase of the buyer’s journey, there are three principles that RED has followed across every aspect of its product and business philosophy: authenticity, usefulness, and quality. These three principles guide RED’s decisions around user experience, community growth, and commercialization.
Because RED sets very high standards for the content that is promoted in the algorithm, fewer people make content (RED has a DAU contribution rate of 1%), but this ensures that the quality of the content is very high, building user trust.
With this strategy, RED has attracted the majority of high-income gen-Z shoppers from tier 1 and tier 2 cities, particularly women (67% of RED users are female). Unlike Douyin, which is mainly used as a way for people to pass time, RED is a very intent-based platform. Users open the app when they are looking for something in particular. The biggest categories on RED are beauty, personal care, clothing, food, and travel recommendations, in that order. Luxury brands perform particularly well in RED, and those brands have an affinity towards RED because of the platform’s high bar for content quality. Top brands on RED include Dior, Estee Lauder, and Lancome. When I talk to gen-Z users of the app, they say that the app brings value to their everyday lives, whereas they feel like most content on Douyin is mindless.
An overview of the product:
Discovery and search are the two main use cases on RED. Upon opening the app, users are shown a For You Feed that leverages a dual-column UI, similar to Pinterest. A dual-column feed requires users to click into posts to see their full details, which drives more intent-based browsing behavior compared to TikTok’s For You Feed. At the top of the feed, there are tabs for various subcategories, such as fashion, beauty, travel, food, and others. This also allows users to be more intent-based in their content discovery.
RED’s algorithm is critical to its success. Whereas TikTok prioritizes distribution of highly viral content, RED’s content distribution is split into 20% viral content, 70% content based on labels that a user is interested in (typically non-viral content), and 10% commercial content. As a result, RED’s content distribution is more decentralized compared to TikTok’s - users are shown more content from everyday people and nano-influencers, rather than highly viral content made by creators. To drive more informative and useful content, the algorithm also uses a community engagement score (CES) that places a higher weight on engagement actions such as saving/favoriting a post over commenting and liking a post (reach out if you want to dive deeper into how RED’s community engagement score differs from TikTok’s).
Although discovery makes up the majority of time spent in the app (+60%), search is the main trigger that drives users to open RED. Compared to Douyin and TikTok, search is used much more often by users of RED. Similar to the discover tab, the search tab uses a dual-column feed for users to quickly search through various posts and click into relevant ones. The search tab also allows users to use filters to search for more specific content beyond just keywords and hashtags.
RED’s creation format is photo-first (65% of content submissions are images), with an emphasis on long text that drives informative, useful content. For conducting research, photo and text is more efficient and allows users to gather key information quickly across multiple posts. A photo-first approach also allows users to conduct research while multitasking, such as watching TV or commuting.
RED trains its community to create high-quality, informative content by investing significantly into moderation. A mix of AI and human moderators downrank content that is not aesthetically pleasing and content that feels too sales-y. Furthermore, the app forces all creators to use a toggle to tag branded content, it uses moderation to punish creators who do not make these disclosures, and it limits branded content distribution in the algorithm. The result is a feed made up of content that looks high-quality, is informative, and feels authentic.
Lastly, RED has enabled livestreaming to drive live commerce sales. It is important to note that live is not the main product and instead is a feature of RED. Several live commerce companies have launched and failed in the West because solving the matching problem with live content is 10x more difficult than with asynchronous content (i.e. putting the right content in front of the right user at the exact right time is very challenging). Instead, live can be successful when it is interspersed in a robust VOD content ecosystem, similar to RED and TikTok.
While RED and Lemon8 have similarities to Instagram, Pinterest, and TikTok, they also have unique qualities that make them stand out from the rest.
Behavior shifts that RED tapped into:
To understand why RED has become so popular in China, particularly amongst gen-Z, we need to discuss the evolution of the creator and attention economy. In China, the explosion of creators on apps like Douyin led to the expansion of branded content and affiliate commerce. Brands and advertisers leaned heavily into creators, and the more they did so, the more that young consumers started losing trust in recommendations from creators. As a result, gen-Z consumers started seeking more authentic, unbiased recommendations from each other and RED became the community for peer-to-peer recommendations. I believe a similar shift will happen in the West as brands continue to tap creators to serve as their modern-day salespeople.
This shift in Chinese consumer behavior changed how brands executed marketing campaigns. Rather than relying purely on branded content deals with creators (Key-Opinion-Leaders), they leaned into activating everyday consumers of their brand (i.e. Key-Opinion-Consumers). This new strategy became known as decentralized, or community-based, marketing. Since RED’s algorithm distributes content in a more decentralized manner, promoting content of nano-influencers and everyday people, it became the perfect platform for this type of marketing campaign.
In the West, we have yet to see a decentralized, community-based platform like RED. Instagram’s algorithm still relies heavily on the friend graph, though Reels is creating more decentralization of content distribution. Since TikTok leverages an interest-graph rather than a friend-graph, its content distribution is less centralized than Instagram’s. However, RED’s algorithm leans even more into decentralization than the rest, which makes it feel more like a community platform than any other player in the space.
Challenges that RED faces:
Although RED has experienced significant growth in recent years, the company faces two major challenges. First, the platform struggles to balance monetization with maintaining authenticity, a core value of the community. The more the company increases ad load and allows creators to engage in monetization, the more the community loses trust in the content on the platform. To address these challenges, RED has had to make UI/UX changes that make it very easy and clear for users to differentiate between organic and paid content. By making these changes and limiting ad load to 10% of total views, RED limits its monetization potential more than other platforms.
Second, while commerce revenue is growing on RED, many users still decide to make the final purchase on other platforms or sites, even after researching the product on RED. To increase in-app purchases, the company has pursued strategies such as blocking creators from linking out to other sites. This, of course, can frustrate creators, but is a tradeoff that RED is willing to make.
Challenges that Lemon8 faces:
So can an app like Lemon8, which is an exact replica of RED, succeed in the West? Some argue that there is already too much competition here. These people argue that while Pinterest, Instagram, and TikTok are not exactly like RED, they have enough similarities and serve the same use cases. However, in speaking with gen-Z women who use Instagram, TikTok, Pinterest, and RED, they express clear differences in how they use the latter. Mainly, they describe RED as a community of people sharing useful, authentic recommendations, whereas they describe Instagram as a way to stay in touch with friends, TikTok as a way to watch entertaining content, and Pinterest as a way to be inspired.
It’s important to note that in spite of Douyin’s success in China, RED has been able to carve a niche of its own and grow quickly. Clearly, ByteDance believes that RED and Lemon8 are also different enough from TikTok, Instagram, and Pinterest to invest in launching Lemon8 as a standalone app. ByteDance is determined to own the discovery-based shopping market and Lemon8 is one more way to get there.
In a previous post, I described some of the differences between Western and Chinese consumers, creators, and brands that contribute to the slower growth of social commerce in the West. One of the main differences is that in China, consumers are less trusting of products sold online and they rely heavily on reviews and recommendations, which bodes well for RED’s business. Since mistrust is not as large of a pain point in the West, the value proposition of an app like Lemon8 may be weaker, which could hinder its growth potential.
Lastly, ByteDance faces both internal and external political challenges that could impede the success of Lemon8. In fact, this isn’t the first time that ByteDance has tried to launch a copycat of RED. In 2020, the company launched an app called Cherry that was a replica of RED with a focus on the beauty and skincare vertical. Despite the app seeing signs of growth, internal tensions and fear of a Trump ban led to the re-prioritization of resources and the shutdown of Cherry. Three years later, ByteDance still faces an uphill battle politically, and having a second app that reaches the masses could put even more unwanted scrutiny on the company.
Conclusion:
Is there room for a company like Lemon8 in the West? I’m hopeful that there is. After all, I rely heavily on subreddits like /malefashionadvice and would love to have access to a photo-first app where I can discover authentic, informative recommendations from other people like me. Will ByteDance be the one to break through in this space? They certainly have the experience and coffers to do so, but I believe a standalone company that is unassociated with ByteDance would fare better in this political climate.