Oct
15
Keyword Research For Recommendation Engines – Followers
Filed Under Recommendation Engines | Leave a Comment
In the previous article about Keyword Research For Recommendation Engines – Views (First Step), we covered the basics of one submission, stuff that many of you already know. The next stage is gaining followers or subscribers. What does keyword research has to do with it? Apparently, a lot. The part of keywords doesn’t end with a success of a single submission, but to understand it, let’s first analyze how subscriptions work.
Article Number 3 – Keyword Research for Recommendation Engines – Followers (Second Step)
One good post can give you subscribers, but most people are reluctant to subscribe, or friend somebody, just because they liked one post. People subscribe not because of what you’ve already given them, but for the promise of what you will give them in the future. Impressing them is not enough, they must believe your future posts to be worthy of their YouTube main or Stumblers dice. The key to subscribers is good history. Videos with a common theme, posted regularly, will cause their target audience to subscribe. Systematical stumble history of a certain subject, especially if your discoveries and positive reviews mark successful, high ranking content, will encourage people to friend and subscribe your profile. However, if you repeatedly submit content with the same tags, your submissions will compete with each other, and thus only few will be able to succeed. In YouTube case, this can be solved by grouping them in a channel, but that isn’t always appropriate, and other engines, like StumbleUpon and Digg, still leave you with a problem. We will put aside YouTube’s channels, as they can be utilized for subscriber hunting by simply posting a stream of episodes of the same show, or a similar technique. We want to learn more ways of getting subscribers, especially those that work on all recommendation engines.
Here is where keyword research and subscription hunting come together. Recommendation engine optimization is not only about submitting once with tags that help you build up from low to high competition. It’s also about maintaining a common theme of submissions, without having your submissions compete between them. To do such a thing, you have to plan on your common theme keywords, preferably words that have high competition anyway, and bring high traffic. In YouTube’s case, plan so you can have a channel on those words, then find derivations from those words, long tails, synonyms and tags relevant to specific posts, and make sure they are your low competition tags. Meaning, the tags that you must dominate right away, those with the same purpose like “black humor” from our example, must be unique between your posts. You shouldn’t bet on the same horse in that case, because if your posts compete there, and some of them fail to dominate their weaker niches, they won’t grow to challenge the stronger niche. On weaker niches the traffic is low, and being second just won’t cut it. On the other hand, if different posts of yours compete for the top keyword, each one of them already has a traffic source in the form of its secondary keywords. Then you can still get good results from the second, third and fourth place of the top keyword, and completely push your competition out[1].
It may be easier to post the same type of content each time then to innovate. Tempting as it is, you have to remember that variation around a common theme is the key to success. Rank high consistently on your main theme and you will gain a following. Keep up with the heat of competition by varying your keywords, and pulling traffic from previously untapped sources. When it comes to recommendation engines, keyword research is about understanding which keywords are constantly used by the users you want as your followers, and also, what variations those users tend to explore, when “zipping” through content. When the same person finds you time after time, on different searches but with a clear continuous theme, this is when your profile gains another follower.
In summary:
Recommendation engine optimization forces you to complete all of your keyword research beforehand, leaving no place for on-the-run corrections. Tools for SEO and PPC keywords should not be your call. Investigating the current ranks of the engine can help, but spending time to plan out your keywords in detail is a must.
Each time you promote something on a recommendation engine, think! Don’t just fill up randomly those tags with whatever comes to mind. Think how this new post connects to your posting history, what keywords link it all together. And then think of what makes it different, what is new about this post that you haven’t posted before. And finally, think like your potential subscriber, and filter from all you brought up, the keywords which make this post part of a surfing pattern. Relevant yet new, something the same user who liked your previous post will enjoy, yet something that can also present you to other users as well.
[1] Note that on StumbleUpon you can’t corner your competition with multiple posts from the same domain, as StumbleUpon posts a limit of one page per domain for a single tag. But you can still post from multiple different sites, which is also more efficient in regard to StumbleUpon’s algorithm. The same goes for many content share engines.
Next time – researching keywords, and how it should be done. Subscribe to the RSS feed to be among the first to read it.
