Topical SEO is what drives the growth of a website’s topical authority. Keywords are still alive but topics are what search engines use to index and rank web pages, for the most part, today. In order to improve your content depth, your focus topics must be carefully chosen and written about. This is where topic research tools come in handy.
Nowadays, keyword research tools are very easily available and with some tweaking, most can be used for topic research, too. Yet, they may be a little too literal and intentionally too close to the initial seed phrase. Topics are broader concepts with a keyword universe sitting right under. Remember that topics are abstract subject matters expressed using keywords — therefore, topics and keywords work together, not against each other!
The best topic research tools should be easy to use in a few clicks. Generally, you type in your core topic and expect key phrases, questions, and other related topics in return. Some keyword research tools are real-time and fast, while others take a tad more time but provide more accurate results.
1. Keyword Planner
Google is by far the best search engine ever, but it is also an advertising machine. The giant has gathered so much insight from search queries and the trillions of web pages indexed that they have their own tool for topic research. Using Keyword Planner is very easy since their recent 2018 update.
Type in one or a group of topic ideas into the search box, and instantly receive a list of related sub-topics and key phrases to write about (or advertise against). And, this tool is virtually free — you just need to have topped up your Google Ads account, even for a dollar!
The only con with this tool is that the suggestions will often be very close variations of the seed topic. So it remains very narrow. See the screenshot below, I entered
Caffeine and most suggestions contain the word
caffeine in them. Although this was to be expected, we do not want to stuff our content with a single core word. There are still some welcomed variations such as
Coffee or inflections like
caffeinated. Remember that using a wide range of terms helps with boosting your topic coverage.
Our free topical research tool! WikiBrowser allows you to visualize Wikipedia pages in a truly refreshing manner. We remove all textual content to focus on related Wikipedia pages. Indeed, a given topic is defined by the keywords used to express it, but even more so by the topics it is a neighbor of.
Little aside here: topic neighbors are how modern algorithms like classifiers find topics. They put all concepts in a three-dimensional matrix (vector space) and see which ones are close to each other.
German Shepherd and
Dog would obviously be closer than
This tool makes total sense. Type the page name, hit enter, and visualize what truly matters when you are doing your topic research:
- a topic’s thesaurus-like outline
- a list of ranked concepts
- keywords pertaining to the topic
For instance, if you want to be authoritative on the topic of
German Shepherds, you must also discuss the
Herding Group it belongs to,
Search-and-Rescue tasks and training, as well as
Police dogs and many other related subjects.
3. Answer The Public
When you know the key phrases you want to cover, Answer The Public gives you search engines’ autosuggest data in a stunningly visual way. So once you submit your topic, it automatically calls Google servers (it takes a few seconds), and then displays common questions, and other search queries associated with your seed topic. Some topics really yield no results like
topical authority (how sad, ha!) but others will get you hundreds of hits.
Check it out, it is a cool way to find some common questions to answer within the body of your blog post, or even pick some questions and answer them with a full article each. There is no search volume data on Answer The Public, but you can export the data as or spreadsheet, to then copy and paste it on Google Ads Keyword Planner to get the historical search volume for each suggestion.
The data is sorted by question tag (what, how, which, who, when, where, etc) but there are also suggestions that are preposition-based. This is an overall helpful tool for common topics. You could complete the data yourself by typing your phrase on Google and look at the recommended searches, the “People also ask” box, and so on. All the contextual data from your search helps you understand what people want to know, so you can better cover it in your blog post.
Topical SEO is so new because algorithms required to detect topics within textual content have only been perfected recently. Before, such algorithms were somewhat inaccurate and tremendously hungry in terms of computing power. Sense2vec is one of those algorithms and a tool has been provided by the machine learning company explosion.ai.
Sense2vec is both a natural language processing method of discovering topics, but also a topic research tool if you want to use their demo. It’s totally free and although not designed or built for SEO purposes, it does the job!
In their own words, the sense2vec demo “read every comment posted to Reddit in 2015 and built a semantic map”. What does that mean exactly? Well, without getting into too many details, the algorithm read all Reddit comments and extracted the main keywords (noun phrases mainly). But each time, the algorithm looked at the closest neighbors and took such factor into consideration when building the knowledge graph.
So when you type in a topic name or keyword, the application finds the associated entity in the model and returns all the co-occurring neighbors that were found in these millions of Reddit comments. Because it is based on a huge amount of data, noise and irrelevant neighbors simply disappear. You may get an odd result here and there but overall, it’s great!
The best for the end — topicseed is our own topical SEO platform! It is a suite of Herculean tools helping content strategists grow their topical authority, find new article ideas, and reach content depth. We also have powerful visualization tools for you to compare a single, or many, web pages against your competition.
During your topic research phase, you want to analyze what your competition is writing about for the same topic:
- how many pages cover a given topic
- what key phrases and words are these pages using the most
- are they using simple or complex variations for their keywords
- what outbound links have they added, etc.
The topicseed platform does just that. Take a look, below, at our phrase table. We analyze a blog post from a hypothetical competitor, and at a glance, we can see the phrases they use the most. If they rank very high, you de facto know what Google wants to read! Each topic on Earth comes with its own ever-expanding keyword universe. Meaning, you must use as many of these keywords to show search engine that you know what you are writing about.
An actual phrase table can hold hundreds of keywords so we allow filtering by terms (just type it and it automatically filters out), phrase score, or even word count. Unigrams (one-word keywords) are rarely useful and salient. We recommend content marketers to focus on bigrams or longer noun phrases. However, focusing on a particular term is important in order to view the list of keyword variants for SEO purposes.
Take a look at the same document’s analysis with a filter added over a particular term we want to focus on:
Here, after filtering added, you can start seeing the long list of keywords variations this document was holding. Sure, the document is clearly about dog clippers, but what does this assumption translate into? Well, they use a lot of variants in order to show Google their topical breadth and depth. Plus, this makes it more interesting for the reader as we avoid the undesirable keyword stuffing.
To wrap up this article, I just want to recommend every content marketer to adopt a wholesome topical SEO strategy. What I mean by that is simple. Take a step back from your article-centric approach. Do a lot of research independently of any singular blog post. Find focus topics you want authority over, create pillar pages, make long-form evergreen articles, shorter articles, and so on.