SEO specialists use content analysis in order to determine how pertinent a document is for a target keyword within a focus topic. Performing an SEO content analysis requires filtering through stopwords and irrelevant sentences. Then, information retrieval methods are put in place to extract meaningful key phrases, topics, and keywords.
Nowadays, most content publishers rely on a single software, or suite of tools, to grade their content inventory. However, not all softwares are reliable and effective in their use of proprietary content scores. Software companies understood that blog editors enjoy see a score at a glance; it’s a great selling point. So they came up with random scoring algorithms, most of them opaque at best, and pseudorandom at worst.
What is the goal of an SEO Content Analysis?
An SEO content analysis is performed to evaluate and grade a piece of content against specific topically-related keywords. Results of content analyses are then used to refresh a blog’s content library with more relevant SEO-friendly information.
Most digital marketers try to produce a evry high volume of content. Finding blog writers is relatively cheap when outsourced. However, content quality and topic coverage suffer in most high volume publishing platforms. For too long, brands favored writing a lot about their domain instead of writing less but more impactful articles.
A great content analysis should be able to consider both human and robotic readers – search engines and actual people! Although natural language understanding has drastically improved recently, most bloggers write for Google rather than for actual people. This is a huge mistake in the long run.
What are the Pillars of an SEO Content Analysis?
A comprehensive SEO content analysis must take into account parameters such as on-page SEO, latent semantic analysis, topical coverage, content depth and breadth, possibilities of knowledge expansion, and readability.
Even as a reader, we know that how much we love a blog post comes down to several parameters. For example, you may like a blog thanks to how easily-digestible the content is, how deep the writer went into their research phase, and how quickly pages loaded. If the writer used relevant lingo you would expect for the specific topic you are searching about, it shows more relevance.
We are reviewing the most important aspects of a successful content analysis below. Remember that writing quality content takes time and requires a lot of editing in order to get the below points right.
Word and keyword co-occurrences are the cornerstone of modern contextual searching algorithms. Indeed, in order to judge how well a topic is covered, search engines build a database of all words a given keyword often appears next to, or in the same articles of.
For example, if you want to build a list of co-occurring terms for the keyword “best coffee beans”, you can expect semantically-related words such as “arabica”, “robusta”, “caffeine”, “caffeine content” to be present. Therefore, any page trying to rank for the same keyword that you index should have these words in the article’s body. Otherwise, the content may be too thin and superficial.
Semantic connectivity is a statistical measure computed using several variables, such as:
- Co-occurrences in many documents
- Distance between the co-occurring phrases
- Term frequency, and Inverse document frequency
- Anchor links and connected pages
- Common modifiers
Using the graph theory, you can use functions in order to dismiss articles that present very little interest semantically in comparison to a user’s search query. By reverse engineering this process, even approximately, you can make sure your content contains enough expected words.
Any given article contains core topics and some more secondary concepts it touches on. Search engines want to establish whether a given article has a lot of depth or breadth.
Content depth is how deep the blog post goes in terms of knowledge about a very precise problematic or topic. If your blog is known for writing several in-depth blog posts about topical SEO, you will inevitably be authoritative over topical SEO subject matters (topical authority, content quality, semantics, etc).
Depth of content is very much sought-after by all contenders in any industry. But by reaching for deep content, you may alienate some of your readership interested in a topic related to yours but not exactly yours. Depending on your target audience and sales funnel, you may prefer content breadth over depth.
The breadth of a piece of content is how well it covers its core topics by also covering related topics – not subtopics! If you blog a lot about digital marketing, you will also touch on traditional marketing. If you have a blog about TV, you would also discuss radio at times.
Content breadth relates to the widening of the knowledge offered in opposition to content depth which relates to the focus on a very narrow topic covered comprehensively.
A blog post should offer possibilities of knowledge expansion through internal and external links covering specific parts discussed in more depth. These links should point to pages related to your piece of content, too.
The same way a topic is the sum of its nearest neighbors, a web page should be topically relevant to the pages it links to, and is linked from.
As a reader, you want to find your information fast on a page, right? But we also love having the possibility to click on a link to expand more on a specific concept or idea. Links helps with narrowing down or broadening up the score of our learning experience. Both will help with increasing the depth and breadth of your content.
As a rule of thumb, you should link internally for more depth and externally for more breadth.
A highly informative blog post will only be useful to a reader who actually reads it. A searcher can be enticed to read an entire article and learn from it. Or, a searcher can scan a well-outlined blog post to quickly jump the right section.
Regardless of how you make a reader read, you must make the reader read for them to gain value from your great content. Unfortunately, not all technical writers can outline and break down an article in a digestible manner.
Most editorial teams will include writers and editors with very different tasks and tools. An editor is not here to change the substance of an article, but to make it more understandable and scannable. New generations have shorter attention spans so they want to quickly get the information they came for.
Back then, people would accept finding the right information after reading a piece of content. Today, a searcher needs it right here, right now, to then perhaps end up reading more paragraphs of that article.
Different Types of Content Analyses in SEO
Not every SEO product defines the word “analysis” the same way. For many of them, they simply offer a score without further explanation. Others give clear instructions on the article’s readability and on-page SEO.
As an editor, your task is not to use the tools other people use. You need to use what helps you publish better content. If outlining articles and structuring the content flow are your strengths, then you do not need a tool to help you with that.
Most tools stick to offering a content score after evaluating a document. Some tools provide you with a score from zero to one hundred for each focus keyword you enter. Overall, this type of tools tries to mimic some of Google’s indexing and scoring logic.
Scoring is tough because nobody knows the recipe used by search engines. Millions of lines of code and hundreds of ranking signals make content analyses using scores a tad obsolete.
For example, most SEO content scorers will rank a page 90 out of 100 but that page won’t even appear in the first page of SERPS. And then, great ranking blog posts will get a 40/100 grade. Overall, content scores are not so much reliable but they give you a hint on whether you are in the right direction, or not. The number is more of an opinion to be taken with a pinch of salt.
Content analyzers extract the important parts of a document to help the marketer study whether or not their own SEO goals were met. Unlike content scorers, analyzers offer a quick way to visualize what matters in a blog post, without assuming a numbered score. Most content analyzers, though, offer relative indicative scores scoped to the document itself.
Our own suite of topical SEO tools, topicseed, is focused on surfacing the most important data from a document or web page. We do not try to assume what you want to know about it. Instead, we use extremely powerful machine learning and rule-based algorithms to perform accurate information retrieval tasks.
Readability scorers, like Readable.io, are more reliable but less useful for blog editors. Readability is the field of making a written article more digestible for the reader through:
- Outlining the blog post using indicative titles
- Shorter paragraphs with main points highlighted
- Clean sentences flowing well together
- A balance of active and passive voice
- Avoiding repetitions
Famous readability checkers are the suite of Flesch-Kincaid readability tests ordered and developed in 1975 by the U.S. Navy. The tool Yoast SEO uses these evaluations to assess how easily a piece of content can be read and grasped.
Because readability rests on more logical and known principles, such tools are a lot more accurate than content scorers. We all enjoy a good outline, paragraphs properly broken down, and highlighted points.
On-page SEO Checkers
Another range of SEO content analysis tool focus on on-page SEO signals set by the industry as basic guidelines. These on-page signals include inserting focus keywords in important spots (article title, meta tags, introduction, conclusion, headings, links, etc), repeating the main keywords enough but not too often, avoiding linking to competing pages, and so on.
These are the most refined tools since on-page SEO is a domain relatively clear for most experts. We all agree that keyword placement, keyword density and keyword variation all matter. Then, such tools are here to save blog editors and content writers time while crafting an article. Consider on-page SEO checkers as Checklist as a Service applications. They are pretty much in every editor’s toolbox (e.g. Yoast SEO for WordPress bloggers).
Technical SEO Tools
Lastly, technical SEO auditing platforms do not judge a page’s actual content quality but rather how user-friendly the browsing and reading experiences are. Technical SEO has some weight in search engine rankings so it should absolutely be considered seriously.
Page load speed is perhaps the most checked technical aspect along with mobile-friendliness. These tools tend to take measures of various geeky units and then offer ways to improve the UX (e.g. enabling server-side caching, using a CDN, compressing images, turning GZIP on, having HTTPS, etc). Famous technical SEO auditing tools include Pingdom, SiteChecker.PRO, and Google Lighthouse.
Are Content Scores Reliable in SEO Softwares?
Numbered content scores offer an approximate measure of how qualitative an article is. SEO content writing tools try hard to understand whether or not an article is valuable in the eyes of a search engine, but it’s nearly impossible. The only numbered score that is 100% valid is the ranking of your webpage for a particular search query. Anything else than that is pure speculation.
Any scoring system needs a scorecard. A valuable content score is one that is clearly scoping its evaluation. For instance, scoring a keyword density or a phrase within a document is totally acceptable. However, scoring a webpage “as is” with no other input is utter madness.
Because scores are sexy, most SEO software companies try hard to come up with an appealing marketing name for their score. But do not fall for these, only trust scores that you can confirm the accuracy of yourself enough time to deem it as valid.
Not every SEO content quality score is erroneous, but too many exist just for the sake of marketing.