“Make everything available. Help me find it. Cut the price in half. Now lower it.” Cris Anderson gave us these four rules in his 2006 book, The Long Tail: Why the Future of Business Is Selling Less of More. This made a lot of sense in 2006.

Amazon, eBay, Etsy and others had just smashed the constraints of brick-and-mortar infrastructure and traditional supply chains. Their success demonstrated that their brand’s ecosystem, where the less popular but more personally interesting shiny objects live, held enormous potential.  

Now, 21 years after Anderson introduced us to the long tail, just four ecommerce sites dominate in the U.S.: Amazon, with a staggering 38% share, followed by Walmart at 8%, eBay at 6%, and Etsy at 3%. With four sites owning 55% of ecommerce in the U.S., the sheer volume of available products can be overwhelming, often leading consumers to pick the big hit, the one with the most and best reviews. All head, no tail.

20 Years Later

 Long tail data consists of rare, unique, or specialized data points that fall outside the most common or popular categories. While each individual long tail data point might seem insignificant on its own, when aggregated, these data points can represent a substantial portion of the total dataset.  

Companies like Amazon leverage long tail data based on their over 300M customers’ shopping and buying habits. This allows Amazon to offer niche products that appeal to very specific customer segments which in turn creates value by serving niche interests. That’s an optimal use of long-tail data. 

Long-tail data can also help determine the demand for niche products and infuse prediction models with data to inform new product development. And, since long-tail data is derived from niche, lower-demand products, predictive algorithms gain a more useful, accurate level of granularity. That’s good for new product development too, as well as trend analysis.

Social Media, Forums, and Safety

New research featured in the MIT Sloan Management Review reveals that social media influence has its own long tail. Influence does not all derive from the top. They found that “influencers with lower follower counts delivered dramatically better per-follower returns on marketing investments than macro-influencers, those with a high number of followers. Macro-influencers in the researchers’ dataset had 32 times more followers but generated only about four times as much revenue compared with nano-influencers.”

Another rich source of long-tail content and data are message boards and forums. In spite of the behemoth social platforms like TikTok, Facebook, and even Reddit, according to SocialPlus, about 80% of internet users participate in niche online communities. And that number has remained steady for several years. 

The authentic connections users look for but don’t always find on social media are part of the fabric of online communities. According  to GWI, that’s a major reason why people are turning to online communities more and more. Gaming, gardening, fashion, and health are a few of the most popular niches that online communities form around.

At least 40% of web traffic is made up of fake users, or computerized bots, according to cloud-services provider Cloudflare. That share is expected to grow with the proliferation of AI systems that regularly scan the web for data to ingest, along with the recent decision by Facebook to eliminate fact-checking. 

While this is an increasingly major issue on the largest social media platforms, it is less so with online community forums. One reason is that forums typically have sharper moderation. Another is that bots are more easily identified in niche communities. Far fewer bots, tighter moderation, and passionate, authentic commenting provide us with highly useful data that we can trust, leading to long tail insights that accurately reflect real human tenor, along with the voice and the opinions of community members. 

Socialgist provides the data access layer between the massive social media networks and niche online community sites. We make it easy for you to access and analyse the world’s conversational data. Accessible, trustworthy social data is good for business.

‍