In his book, The Long Tail, Chris Anderson explains that “The theory of the Long Tail is that our culture and economy is increasingly shifting away from a focus on a relatively small number of “hits” (mainstream products and markets) at the head of the demand curve and toward a huge number of niches in the tail.”
Twitter is a great example of the long tail culture: the number of mainstream conversations and mainstream users/influencers on Twitter is relatively small and the long tail of micro trending topics, micro influencers and micro themes is almost infinite. For example, only 20% of the daily conversations on Twitter contain “common keywords” (in this case defined as one of the top 100 frequently mentioned keywords within a certain day). Moreover, the set of top 100 frequently mentioned KWs in any given day is very different from one day to the other. As a result the breadth and variety of conversations discussed on Twitter are immense.
Long tail keyword targeting, has long been a very important tool in the arsenal of sophisticated paid search (or Search Engine Marketing) bidders on Google. For example, given that any advertiser is familiar with the most common keywords in their domain, such as “vacation”, “insurance”, “buy shoes” etc., and the fact that these keywords are highly sought after by many advertisers due to their volume and traffic quality, the cost of acquiring new customers via these few “head term” keywords usually skyrockets, sometimes way beyond any justifiable LTV. Hence, to allow ROAS positive acquisition of new users via paid search on Google, sophisticated paid search bidders have developed a variety of ways to bid on long tail keywords, so instead of bidding on a keyword such as “vacation” they would bid on a cluster of 10,000 keywords containing the names of all the vacation hotels in the pacific, all the honeymoon destinations in the world etc.
As advertisers think about how to leverage Twitter’s unique characteristics for reaching the right users at the right moment with the right offering, the concept of “long tail” comes in very handy. Targeting followers of certain accounts on Twitter, is a very common way to define a relevant targeting audience for paid campaigns on Twitter. For example, a high-end apparel advertiser might want to target followers of the Gucci Twitter account, as these followers have some relevancy/affiliation to high-end apparel. However, similar to the dynamics on Google, given that Gucci is a very known brand, targeting its followers could get very expensive, as many advertisers ate bidding on this, and even if you could afford bidding on the followers of Gucci, at some point, you exhaust or fatigue this audience, so you pretty quickly need to find other things to target in order to continue acquiring new customers at scale.
A long tail approach for follower targeting on Twitter usually includes targeting followers of much smaller Twitter accounts (e.g. with thousands rather than millions of followers) and serves two important purposes: 1. It extends the size of the targetable audience beyond those who just follow the big brands. 2. even if the targeted user follows Gucci (which has 3M followers) the advertiser is better off targeting a smaller and more specific group of users who are interested in a more concrete topic (e.g high end handbags) such as @GiGiNYTweets (GiGi New York: high end leather handbags, clutches, wallets & accessories) with 4K followers. But wait, given that these long tail handles have much smaller numbers of followers, how do I reach a large enough audience? It’s simple, you just need to target the followers of many small, highly relevant accounts to obtain a critical mass. The tricky part with long tail targeting has always been that it is as time consuming as it is effective… By closely listening to advertisers’ input surrounding this very issue, Comprendi has developed a groundbreaking audience discovery product allowing the advertiser to build large scale semantically relevant long tail targeting sets within minutes. Our patent pending technology uses Natural Language Processing, Machine Learning and Deep Learning and has proven to increase scale at highly positive ROAS for top mobile app, brand and direct response advertisers. Want to learn how this could help your campaigns grow? Contact us to learn more! email@example.com