Keyword Research Workflow: Study Demand, Understand Search Intent

One of the most powerful features of InfraNodus is its ability to understand search intent: keywords that people use to find what they are looking for. 

Knowing the search intent gives you unprecedented insight: you will have a much better idea of the product and content you need to create in order to get in front of the people's eyes. This is useful not only for marketing and SEO applications but also for researchers and writers who want to make sure that their ideas can actually be found on the internet. 

Here's the workflow described step-by-step. It is also available when you use the Search Intent / Keyword Research App in InfraNodus at the top right corner. 

Video Tutorial

Step by Step Workflow

Think of a topic you're interested in and come up with the most general search query that your potential clients might be using. For instance, "ai agents".

Step 1: Import the search intent keywords 

Go to https://infranodus.com/apps and open the Marketing > Search Intent import app. This app will extract the keywords suggested by Google to a search query in the "people also search for..." field. You can also use AdWords suggestions, which are more general and might reveal other topics that people who search for this topic are interested in.

In our case, we will be using the following search query: "keyword research"

Step 2: Reveal the top keywords and topics

InfraNodus will visualize the top related search queries to your original query as a graph. The keyword lemmas are the nodes, their co-occurrences are the connections. The clusters of keywords that occur more often together will be closer to each other on the graph and have the same color (using the force atlas layout and modularity algorithms). The keywords that connect different clusters will have higher influence (based on betweenness centrality) and shown bigger on the graph. 

This visual representation lets you quickly see the most important topics and keywords in search intent. You can use those to better understand what people search for when they look for a query / product that you are researching and adopt your content strategy accordingly:

Note, that InfraNodus automatically hides your search terms from the graph to show the context around (stored at the top right). You can return those nodes to the graph but as they are in almost every search query, those keywords will take all the focus and it will be more difficult to explore what else people search for.

For instance, in our case, seaching for "keyword research" reveals that the most influential keywords are "youtube",  "channel", "google", planner", and 'free". This is an indicator that people are interested in keyword research for YouTube channels — a very interesting insight for us if we want to create a keyword research tool. We could then compare it to the graph of the search results for this same query using the search results analysis workflow to see whether our readers actually find this content easily when they search for it. A gap here would indicate an opportunity for a business that would like to rank higher for this query.

The main topics shown to the right also include "Youtube tags", "Digital marketing", and "Keyword analytics" giving us a high-level overview of the main themes people are interested in when they search for "keyword research". Targeting these themes in our content strategy will ensure we will get constant search traffic.

Actionable insight: understand the general interests of the people who perform this search to know what's happening in their head when they search for a specific topic.

Step 3: Identify top keyword combinations

The next step is to shift our attention from the most influential keywords and top themes and zoom in on specific keyword combinations. These can reveal interesting insights about user intent that is not visible on the surface.

In order to do that, click some smaller nodes that are close to each other and switch the the Analytics > Keyword Relations tab to see what else they're connected to. The statements on the left will show the search queries that those keywords are used in and the number of searches per month:

We can see that "youtube tag generators" is a popular search query related to keyword research. Therefore, we would do well to offer a free youtube tag generator on our website if we want to promote our keyword research tool to users. 

Actionable insight: look for the patterns of other keywords that people use when they search for a certain query and target those to maximize your exposure

Step 4: Save your insights to project notes

Remember to save all those keyword ideas to your project notes. You can use those notes later to generate content:

Step 5: Use AI to generate content ideas from the keyword graph

A huge advantage of using the graph is that you can generate content from the keyword graph by selecting the topics or concepts and using the AI idea button to generate a business or content idea for them. 

You can display the high-level topics in the graph using the Analytics > Main Topics panel and make sure that your content covers all of those topics. 

You can also use the Analytics > Missing Content panel to have InfraNodus identify potential opportunities for you, reveal the gaps, and generate ideas based on them. For example:

The InfraNodus AI here generated a really good idea for us: make a free demo tool that uses this very search intent keyword research tool to generate tags for YouTube videos. We will definitely implement it on our website to make sure we cater to this existing demand.

Step 6: Reveal latent keywords

Last, we can also focus on Gateway Keywords shown in the Analytics > Missing Content panel. These are the keywords that connect different clusters together but are not the most frequently used ones. They can be useful for discovering how the search intent for this query can be connected to other interests that your potential users have:

For instance, in our case we find out that there's a tool called "keyword magic" from SemRush. So our content strategy could include making a page where we compare our product with that tool to attract the search traffic that looks for alternatives. 

Another approach you can use is to remove the most frequent keywords from the graph to reveal what's hiding underneath. This is akin to slicing off the top layer of concepts to reveal underlying ideas and latent keywords:

For instance, in our case we also learn that people are searching for "reddit search trend", which can be an interesting long-tail search keyword opportunity. 

Step 7: Add more search queries and reiterate

Once you finished studying search intent for a certain query, import the search intent for another related query. Or use the same search query but import the AdWords suggestions to expand the graph to the other topics to understand what else your potential audience is interested in.