|
|
|
Source: Lucas Amunategui
I am not advocating plagiarism, not even close. I am talking about finding top-rated, online material that closely resembles yours and comparing the keywords, tags, and meta-data. If it’s different than yours, you may have a problem.
This doesn’t mean that your content is wrong, it simply means that your word choices may not align with current tastes, fashions, or with the mysterious Google Search machinations. Top-rated blogs have nothing to prove, they’ve made it to the top of the Google Search heap and gained ultimate visibility. If it's more traffic that you’re after, you’ll have to dig in and understand the differences between your content and theirs so you can re-align your wording.
If the above resonates with your goals, then you and I are on the same page, and you may be interested in this approach I’ve recently adopted. I have also built an experimental tool that you are free to check out at http://www.viralml.com/seo_keyword_comparer, but let me first show you how this can all be done manually.
This shouldn’t be brain surgery but due to the opaqueness of search engines, there really isn’t a single way forward. Let’s focus on Google Search as that is the most popular one. From what I’ve read, it seems we can enter about 32 keywords into the Google search bar. So go ahead, enter that title and description or as many keywords that best describe your content in Google Search. The goal is to use similar wording that your potential reader would.
If you’ve chosen the right words, the best matches that surface from the search will be your top contenders - i.e., they do what you do or say what you say but their links show up at the top of the search list. The title and descriptions in the search should be highly relevant to your own content (to the point that it may surprise you that so many people have already written on that exact same topic — but don’t let that discourage you).
Now, pull a couple of the top contenders and read their titles and first paragraphs. Do they do a better job describing your topic? Are they using different words, different phrases, and tones? Do they have better graphics? Are they more or less salesy, transparent, to the point, etc? These are critical content areas that the Google Search engine will parse to rank your posts.
Though this will already reveal plenty, we can still dig deeper. Now, look at the source code for those top contenders (in chrome, its as simple as selecting ‘view’ in the top toolbar then ‘developer’ and ‘view source’. Once inside, look for the content inside these critical HTML tags:
Also look at the META tags in the header section of the HTML source:
If these top sites are using their HTML properly, you should see the distilled essence of that writer’s content hidden in plain sight within these tags. From my research (and truth be told, the information on this subject is numerous and all over the place), this is one of the areas that Google Search will use to index and rank content.
Also, even if you post on sites like Medium where you don’t control the HTML, knowing how other writers are using their tags and the text they put in it can help you tweak your own content. Either way, it would be foolish not to beef up yours.
Just as I did with the http://www.viralml.com/writing_analyzer, I wrote a simple web application to automatically compare inputted content or keywords against top-ranked and related URLs. Check it out here: http://www.viralml.com/seo_keyword_comparer. Here is how it works:
Enter your title, keywords, and/or first paragraph that best describes your content. It doesn’t really matter if all you put are keywords or well-formed sentences, it will get cleaned up on the backend into a more generalized format.
Next, you will need to enter up-to-three URLs of closely related content. This can easily be done by dropping your content from Step 1 into Google Search and collecting the top most relevant links. Read the descriptions of each link to make sure that it is closely related and only get HTML files, no PDFs.
Finally, the tool’s code uses embedded-word distances (as I’ve used in previous articles and applications) to calculate the degree of separation between your own words and top-ranked ones. This allows us to identify important words and concepts that may be missing from your content.
Compare your words with theirs, are any of the missing ones important to the message you are attempting to convey? If there is, go back and re-work your content. Also, make sure you are including the typical SEO tags in your HTML source if you can access them.
Sample output from the tool
I entered an article of mine from ViralML, “So, you want to write a top-selling title?”, that I’ve been playing around with to better understand Google’s search ranks. The output is sorted in order of importance and I can already see that the top three words would make great additions to the article’s content or HTML tags. Time to tweak the article again…
I love this stuff; it feels like we’re in the Land of Oz and trying to peek behind the curtain. That said, be wary of big rewrites and constant updates, shouldn’t great writing be able to go viral using its own wings?
For an in-depth look at title writing for visibility and ranking, see How to Write The Perfect Title.
Thanks for reading! And please share this article with a friend or colleague or two :-)
Best!
Manuel Amunategui,
Curator @ amunategui.github.io and ViralML