Technology, Thoughts & Trinkets

Touring the digital through type

Review: The Long Tail (Revised and Updated)

thelongtailI’m in the middle of a massive reading streak for my comprehensive exams, but I’m trying to sneak in some personal reading at the same time. The first book in that ‘extra’ reading is Anderson’s “The Long Tail”, which focuses on the effect that shifting to digital systems has for economic scarcities, producers, aggregators, and consumers.

The key insight that Anderson brings to the table is this: with the birth of digital retail and communication systems, customers can find niche goods that appeal to their personal interests and tastes, rather than exclusively focusing on goods that retailers expect will be hits. This means that customers can follow the ‘long tail’, or the line of niche goods that are individually less and less likely to sell in a mass retail environment.

There are several ‘drivers’ of the long tail:

  1. There are far more niche goods than ‘hits’ (massively popular works), and more and more niche goods are being produced with the falling costs of production and distribution in various fields.
  2. Filters are more and more effective, which means that consumers can find niches they are interested in.
  3. There are so many niche items that, collectively, they can comprise a market rivaling hits.
  4. Without distribution bottlenecks, the ‘true’ elongated tail of the present Western economic reality is made apparent.

longtailgraphicThe long tail of economics (displayed on the right) suggests that the yellow part of the graph is typically under serviced; it doesn’t make sense for shelf-based retailers to keep non-orange products because of the yellow-items’ low traffic flow. What is particularly interesting (for me) is that many of these yellow products include items that would have historically required substantial money to produce, but can now be created and ‘sold’ for low- to -zero costs. Think about blogging: whereas it makes no sense for a newspaper to focus on a relatively obscure topic on a regular basis (maybe banning light bulbs), a blogger will focus passionately on that topic.

This means that the cost of both producing information has fallen, as has the cost of a third-party acquiring information about the effects of banning light bulbs. Traditionally a book, newsletter, or similarly expensive ‘method’ would be required for the light bulb blogger to get their point across; now, because of freely available blogging hosts they can pontificate and only be ‘out’ some time. It makes sense for Google to provide this blogging service because it lets Google capture more and more ‘net traffic as people track down information about banning light bulbs. Google learns about consumer habits and can thus better monetize light bulb interests and the blogger gets to develop a reputation on the basis of the ‘free’ knowledge that they disseminate. All parties, the producer (blogger), aggregators (Google), and consumer (blog reader) win.

Turning to blogging also brings up a key, and regularly noted, factoid about the long tail: the further that you get from the ‘head’ to the tail (left hand side), the more product is available that is both of really good quality, and REALLY BAD quality. Hits and mediocre content tend to be found at the head, whereas hits, mediocrity, and the worst of the worst are found along the right side of the tail.

Accompanying the wider breadth of available products are aggregators, which help consumers actually find product. Aggregators often get a piece of the financial pie from their efficiency in directing individuals along the long tail; think Amazon, and the small cut that they get when a used book seller sells a book through Amazon. Amazon provides a way for individuals to find that book, and have aggregated the massive collection of available books worldwide, and as a result individuals can find books about niche topics that a physical bookstore just couldn’t be expected to carry. In effect, consumers win (more available product, ability to pursue their actual interests), aggregators win (make money from helping to sell product), and many producers win (by increasing their reputation through the wider distribution of their products. Note there isn’t a necessary correlation between reputation and financial gain…).

One real issue with Andersons’ text is the sparse attention that he pays to Cass Sunstein’s worry that aggregators, in providing consumers what they want, will create ‘data cocoons’. Such cocoons are co-developed between consumers and aggregators; consumers supply the search parameters, and aggregators do their best to maximally supply products based on those parameters. This has the effect (per Sunstein) of limiting the range of options that we are presented with; we are less likely to find weird and off-the-wall news items, consumer products, and so forth that we would run into without such cocoons.

Anderson dismisses these worries, claiming that aggregation systems actually have the effect of extending the cocoons we reside in; the ‘surf Wikipedia effect’, or hunt through web forums to find reviews has the effect of shaking off any cocoons as we are accidentally exposed to facts, news, and other disjointing information. Moreover, even when Amazon suggests a book to me, this algorithm is based on the information of all kinds of people who have purchased similar books as I have, which means that some books will be presented that I normally wouldn’t have found. I would suggest that this latter mode of ‘exposure’ doesn’t actually escape Sunstein’s worry, as the aim is to deliver books that I will want to buy according to my niche interests. Amazon is unlikely to market a gardening book to me, which means I’ll never ‘accidentally’ discover my still-hidden love of rose bushes through their automated book suggestion system.

Anderson is, however, likely right that those who are particularly savvy online will commonly escape their cocoons (and, to be fair, this is the audience this book is implicitly written about and even more implicitly, to). At the same time, I question just how much he’s imposing his own attitudes towards online culture on the expected attitudes and actions of the mass public. Where deep packet inspection devices are used to monitor online behaviour and deliver ads that are nuanced for the individual we have an instance where, in addition to self-imposed cocoons, there are corporate threads of digital silk being spun around that person’s preferred search terms.

Somewhat broadly, I have my doubts that Anderson’s ‘solution’ to data cocoons would pass my ‘mom test’: Would my mom be as likely to thrash out of the data silk that Google Reader, Amazon, behavioural advertising, and customized marketing systems spin as Anderson expects? Likely only to the extent that she is subjected to ‘meat world’ environments. Digital systems alone would, I think, be fairly constraining.

This criticism aside, Anderson has managed to write an incredibly accessible book that reveals a new way of approaching economics in an era of digital networks, products, and aggregation. I’d highly recommend it to anyone interested in the emergence of new economic models and possibilities – it will bring discussions of blogging, copyright, and similarly digitally distributed products into new (and productive) lights.

Rating: 5/5


  1. Very interesting article. I like how you went through each of the blogs that you worked with and pointed out different things.

Leave a Reply

Your email address will not be published.