We don’t blame you!
Many claim to know how to analyze hundreds of signals to come up with the best structure for your keywords.
Alas, despite their valiant efforts, the advents in NLP and the sheer amount of variables they analyze, they fall short.. for one simple reason..
The “heart of each keyword” is a mystery known only to the “Black Googlebox”.
Most approaches focus on analyzing different variables like TF-IDF, site topical relevancy, saliency scores and NLP ontologies... which is great!
The problem is that there is always going to be a bias inserted as these methods can only speculate as to what this "black box" truly does.
Our approach is to scratch all of that and focus on the results without taking anything else into account or trying to hypothesize on what variables may be the most important.
There are a number of reasons why our strategies will yield different results than anyone else.
While others focus on analyzing which input signals (keyword auto-suggest, entities, tf-idf, salient scores, etc), affect rankings… we simply don’t care.
We use an ensemble of unsupervised machine learning algorithms to cluster keywords that have a high overlap in SERPs.
We don’t insert any biases around what we believe should be grouped together or what keywords are intuitively associated.
In other words, if Google thinks that certain keywords should be together, we give our blessing as well.
Taste and preferences change constantly.
It takes a lot to mend a broken heart so we take our role very seriously and we want to be certain that we have the latest “preferences” in order to derive the best decisions.
In order for our approach to work we cannot rely on historical or cached data.
Every time you run a report we make custom pulls from dozens of datasets and APIs to ensure we have the latest information at our disposal.
Not all keywords are meant to find mates.
Some very independent ones don’t fall into any category and they are destined to be single.
When a match is far from perfect we indicate this with our color coding in the interactive tree diagram we provide.
Thus, you can judge their compatibility for yourself (maybe get a prenup).
In order provide you with the best and easiest experience, we have developed a sophisticated cloud platform with numerous features that will help you achieve your goals.
To deliver the most value out of our reports, we have included these 3 features that allow you to customize your reports and help you derive actionable items for further steps.
Our system integrates with many popular keyword tools in the SEO space, such as Ahrefs and Semrush.
By selecting our Bring Your Own Data
(B.Y.O.D.) reports, you can directly upload an export from your favorite tool
and we will organize all the keywords for you.
If provided, our tool will also aggregate the impression volume, keyword difficulty and cost per click to coincide with our clustering.
This will allow you to make informed decisions on which clusters to target using wholistic metrics, rather than looking at individual keywords.
Alternatively, you can provide a list of comma-separated keywords through our "Simple Reports" and we will cluster the keywords for you.
Are you targeting a specific location in your research? Are you looking to optimize for mobile traffic?
With Keyword Cupid, you have the ability to select a specific location, search engine, and/or device and we
will provide the most relevant clustering for your target audience.
By targeting different search engines, geo-locations and device combinations, you will be able to capture the variability
of the semantic and contextual relevancy of different keywords and the trends and intents that emerge for each combination.
By employing our customization features, your content will be tuned precisely to your target audiences.
If you’re still wondering what our clustering can do for your site, look no further.
By providing us with your target URLs, we will scrape the pages for your most prominent on-page factors, such as outbound links
(contextual vs header/footer), H-tags, images, bold keywords and so forth.
We also recommend which keywords to use in your on-page elements based on the longest common subsequence
between all underlying keywords of the respective cluster, that are missing from your content strategy.
This information provides you with actionable items for your site that you can implement immediately to rank for the keywords in these clusters.
Ready to join our Keyword Research Family and take your content strategy to the next level?
If you still have questions, that's okay.
Take advantage of our 7-Day trial that allows you to test all our features for just 1 dollar!
We provide you with various options to segment and visualize your results.
This way, you can decide the best strategy to integrate our efforts with your current business pipelines.
We offer two report options for you to interact with our "matchmaking" results.
An interactive graph and a traditional Excel sheet.
Explore thousands of keyword relationships through our interactive dendrogram.
Hover over each node to find total volume and average weighted difficulty for each particular silo.
Our interactive chart allows you to reposition nodes where you see fit and save your work.
Our downloadable Excel file aggregates the pages in silos and sorts them based on their cluster confidence and total impressions. Each page cluster has pre-computed aggregate formulas that calculate the total volume, weighted average difficulty and average CPC for each cluster. You can remove or add keywords and the formulas will update automatically.
