Long Tail theory in a wider interpretation is the Wired editor Chris Anderson’s brainchild: „The Long Tail, Why the future of Business is Selling less of More” states that the items that individually have low demand, if accumulated can draw in significantly more demand and therefore can imply multiple times more income than popular products that sell in huge volume.
Anderson made measures in many industries and found a stunningly similar trend outlining a shift from selling a few hits in huge volumes toward selling numerous niche items, each of them in relatively small amounts. Online auction site Ebay success is credited to the presence of lots of auctioneers selling small quantities of “non-mainstream” products. Infrequent sales produces an aggregate sales equivalent to or even exceeding the mainstream product sales. Anderson defines 3 reasons for the proliferation of this phenomenon (in the media content industry basically, but „2,5 of that 3” also apply to the whole e-commerce.)
Technology costs decreased and enabled anybody passionate enough to sell amateur music, films, softwares, apps with professional results and professional incomes. However tools of mass production aren’t democratized so much, but the small quantity, handcrafted – manufactured goods see a throbbing renaissance: people can sell worldwide their artworks, seemingly worthless memorabilias, fiddle-faddle stuffs from the shed via running a garage sale in the Etsy or Ebay.
Not only digital content inventory, transaction costs lowered, but the whole e-commerce is swarmed by „smart & sexy solutions” that can turn anyone to an e-store owner overnight. No store, no stock, no skills needed, the bandwagon is opened to everyone ready to board.
Selling niche content without finding the potential buyers is impossible without powerful search and recommendation engines, pervasive user ratings, and proactive communities of interest that have catalyzed the long tail revolution.
In the majority of recommendation systems there are a few popular items and vasts of less popular ones. If a system recommends the popular items only, the user experience will be inflicted on the other end: the customer will get too familiar with the often recurring products or contents, and might lose her interest. If your e-commerce store shows only familiar items, you can easily lose your customers, they will migrate to other web stores where product versatility is higher.
The Challenge One for the system to predict which item will be useful for the customer in the future. Challenge Two: to moderate your recommendations coming from the long tail and find the equilibrium somewhere between the waterfall of unknown, new items and boring homogenity.
Products in the long tail are rare, often obscure products. E-commerce businesses might have these goods in stock but physically they’re not there: using expression „In Stock” is very pliable, as they are ordered rarely, the stock is often on another continent.
Let’s say, you recommend products from the long tail, and people start liking them. It will present items for the users that they wouldn’t have explored by themselves. When the algorithm decides which item deserves recommendation it has to differentiate mainstream and long tail items: for example the engine thinks you will like 178 batteries based on the algorithm set up by your profile. The first problem here is the „learning to rank”, which item to recommend and in what order.
Long Tail is a problem but an opportunity also, when you get to the boundary where simply offering popular items is not an appealing strategy anymore, because your customers have already reached the limit of buying popular goods. If you sell inferior goods or those very popular goods that can be found in your competitors’ pool you have to open your „flexible in stock- category” and unleash the long tail potential.
Because in the top popular product league you cannot compete with e-commerce giants in price.
1.Thumb rule – Only e-commerce giants can afford themselves to stay out of the long tail hassle because they can bargain a wholesale price on the very popular products.
2.Thumb rule –You have to make users interested in a few contents they can’t find anywhere else. Recommending from long tail is a good opportunity to evade the marketplace controlled by a few e-commerce giants.
You are going to face 2 problems here. Problem 1: the learning to rank problem. When you virtually put lots of stocks into the long tail „in stock” maybe 300 items. Which product to recommend and in which order. Problem 2: the diversity. From the 300 items, with what frequency to show any of them?
Other retailer Long Tail thumb rules
1. If your goal is to keep and nurture your most loyal and best-converting customers, expand your scope of products with more niche ones, because the already loyal group will definitely have a bold interest in your less popular stuff.
1.Keep under control the costs of offering items that rarely sell.
2.However niche products could have higher profit margins, don’t steer the customers to the long tail very often, otherwise you might risk their satisfaction.
Naturally, everybody goes for the top keyword terms. Companies insist on their favored „trophy terms” boosting their product visibility and attracting traffic, in spite of it has proven in several industries that the aggregate demand caused by the long tail – inspired keywords are larger than the value driven by leading terms.
SEO-experts use another segment to represent the long tail concept in a more realistic way: chunky middle means the broad border zone can be still utilized for SEO performance in a well-measurable way.
The top 10,000 keywords that initiated the highest number of Google searches in these 3 month only made up 18.5 percent of all the searches. The long tail brought in 70.0 percent of searches while the chunky middle entailed about 11.5 percent. In case of 50000 searches, head keyword searches raised 9250 ( 18,5 % ), chunky middle keyword searches made 5750 ( 11,5 % ) whilst the rest was driven by long tail keyword searches (35000, 70 %).
You can lose 81,5 % of the searches if you shoot only to mainstream phrases.
The long tail scheme is mirrored in the e-commerce sites: main page and first level category pages should target the mainstream keyword phrases like “shoe”, the second-third depth category pages have to target the chunky middle section in key phrases (flat shoes, smart casual shoes ) while the product pages ( product pages plus recommendations ) usually activated by long tail keyword searches like “purple mary jane shoes”
Some good news for e-commerce owners: optimizing for the long tail + chunky middle can be scaled easily, can be optimized in the taxonomy and in the template dimensions of your store, and on the platform adjustment level, even without SEO knowledge.
The downside is that long tail- and the chunky middle optimization burns design and development time, especially when you’d like to optimize key phrases for product + reco boxes.