1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

The 5 Best Keyword Metrics

Discussion in 'Black Hat SEO' started by judaculla, Jul 23, 2016.

  1. judaculla

    judaculla Jr. VIP Jr. VIP

    Joined:
    Oct 11, 2014
    Messages:
    337
    Likes Received:
    124
    Location:
    USA
    I'm building a pretty in-depth script that auto-posts content with injected keywords to my PBN sites.

    I'm at the point in my script that I need to make some forward looking decisions, and was hoping for some input from those with a little more experience in IM than myself, specifically keyword metrics which are viewed as most-important.

    Currently, I have no intention of selling any sort of access to this script, and only intend to use it for my own purposes. However, I can see potential in the future of allowing others to use it as well, and I am trying to imagine the type of data feedback one might want (since it would serve me as well.)

    So, if you would, imagine yourself as a potential user to the following system and think of the types of feedback you would expect if you were paying money for the service, and please don't be minimal.

    The script scrapes all kinds of websites, and more are being added constantly. It takes this content and posts to many sites within my PBN as non-original content, which gives credit to the original source. Think; newsblog type sites. It stays within the range of 'fair-use' and only posts ~300 word teaser articles with 'Read More' type links at the bottom of each article.

    Part of the process of the posting of this content to my PBN, is to make keyword-substitutions in the content, which are made by searching the original content against a spreadsheet I have of keywords and links. So, if I pulled in an article about fishing from the BassPro blog, and it had the keyword 'best fishing pole' in its' text, the 'best fishing pole' keyword in my keyword spreadsheet would automatically create a link using the keyword anchor text, which links to whatever target url is paired with that keyword in my spreadsheet.

    So, in effect, every time my scraper runs and posts content to my PBN sites, it automatically adds links for keywords that I'm targeting.

    Now, every article doesn't have those keywords in it, so the frequency of use isn't very predictable, but so far in testing it seems to work well for most all keywords. I certainly plan to accommodate some integration for partial matching, LSI matching, or just general random topical matching, but I'm not getting that specific just yet.

    My question is this, if you gave me a keyword and url to link to, and paid me money for the above service, what type of feedback would you be expecting?

    So far, I've thought the following would be useful:

    # of times keyword used
    Text of entire sentence in which it was used
    Dates on which it was used
    Topical analysis of content in which it was used (30% news, 20% tech, 15% security, etc...)
    # of links per article where it was used
    no-follow / do-follow count (maybe an option for control over the ratio?)
    rel= attribute usage for other attributes such as noreferrer

    Another consideration I'm attempting to integrate, would be the ability to choose what 'types' of content keywords would be inserted into, with topical consideration. Many of my PBN sites are simply news-blog type sites because they are easy to manage and maintain and can include just about anything and still seem 'natural.' However, I am considering adding flags for certain sites that are developed in the future that are more niche-relevent, so that the keyword input could be as follows: [keyword, link, topic(s)] in which case the links would only be inserted into articles which had content analyzed as being weighted towards the chosen topical categories. How much more would you be likely to use a service if it included that type of accommodation compared to if it didn't?

    I would really appreciate any feedback anyone would be willing to express. Like I said, this isn't planned to have any type of commercial access, but I've always found that consideration for commercialization often brings about powerful insights.

    Thanks guys, and I apologize for the cheesy title for the thread.
     
    Last edited: Jul 23, 2016