Using AI for Keyword Research

I’m the SEO Manager for PaperStreet, and, when it comes to my daily workflow, I use AI in a lot of different ways. It has not removed any of the tasks I used to complete prior to AI becoming a part of everyday life, but it does help me accomplish tasks faster, organize information better, look at something differently, and/or avoid doing repetitive work.

Using AI for Keyword Research

Keyword research is a good example of how AI has improved my workflow. AI can help generate keyword root ideas before I start the actual research. For an immigration-related project, I asked AI to identify different practice areas, legal issues, and ways people might search for those services without having to manually research them myself. From there, I could take those ideas into the SEO research tools we normally use and see what was actually being searched.

I use it in a similar way when I already know a topic but want to make sure I am not overlooking keyword variations. AI can provide me with a more comprehensive starting list without having to think of every possible variation manually.

Another way that AI helps me is when two keywords could be similar enough that I am not sure whether we should recommend targeting them both for a campaign. I can use AI to help with a SERP analysis to compare what terms Google is actually returning for both searches.

For example, we might be exploring keywords such as “premises liability” and “slip and fall.” If the same types of pages and results are appearing for both keywords, that is a good indication that we need to pause and reflect before we create two pages that may end up competing against each other.

Working Through SEO Strategy

I also use AI often when working through SEO strategies for campaigns. Usually I’m confident in what we should recommend, but I will use AI to test the idea, see if I overlooked anything, or help me figure out the best way to communicate it to others.

It is certainly not that AI is coming up with the strategy for me. It is simply another way to challenge or verify an idea before I proceed with it.

Turning Data Into Something More Useful

When I am working with a large volume of campaign data, I don’t want to just send a client a bunch of statistics and expect them to understand the report.

AI can help me analyze any information, highlight the main points, and turn those findings into actionable tasks. Some examples would include:

• Keyword rankings and changes in visibility
• Website traffic
• Conversions and leads
• Phone call data
• Several months of performance data that needs to be condensed into a few clear takeaways

From there, it allows me to focus on what actionable items we should do based on the information and how to communicate that clearly to the client.

Improving Emails Without Changing What I Want to Say

Writing emails is probably one of the areas where I use AI the most. I already know what I want to say but I need help making sure I am saying it clearly, not overexplaining something, or getting the tone right.

I also approach client and internal emails very differently. 

With a client email, I am thinking more about it sounding professional, concise and the overall message being easily understood.

With internal emails, I am more concerned whether I explained the issue clearly enough and whether the person I’m assigning a task to knows exactly what I am asking. 

In both scenarios, I take a message I already drafted and ask AI to make it shorter or softer without altering the point I’m trying to get across.

Saving Time on Smaller Technical Tasks

There are a lot of smaller uses for AI, as well. For example, if we find a broken link on a website, AI can help find possible replacement resources instead of me manually searching for every single one.

I have also used it for simple website code changes when the update is small enough that it does not necessarily need to involve a developer. For example, it could be an update to only display information on a website when a particular field has been filled out or changing the order in which certain information appears.

I still must review what code update it gives me and make sure it actually works, but it can save a lot of time being able to make updates without the assistance of a developer.

Some of the Best Uses of AI Have Nothing to Do With SEO

Some of the ways I like to use AI the most are actually the ones that have nothing to do with traditional SEO.

At one point, I needed a list of client accounts from our CallRail dashboard. Instead of going through the dashboard and copying every account name one at a time, I took screenshots and had AI pull the names out of the images for me. There were so many client accounts that manually copying them all would have taken quite a bit of time. The task was simple, but it saved me time on doing less tedious work.

I have also done something similar with training materials. I have taken a transcript from a Zoom training session and used AI to turn it into an organized document.

As an example, there was an internal CallRail training where the recording was useful, but the team needed something they could refer to afterward. If someone needs that information later, they can go straight to a document AI created that outlines the main steps and instructions without having to re-watch the entire recording.

AI Is a Tool, Not the Final Decision-Maker

Overall, AI is in my daily workflow, but it hasn’t taken away my strategic thinking. My approach is not trying to give AI an entire project and rely on whatever it gives me in response. Instead, I am looking for the parts of the task where AI can save me time, help me think something through, or make information easier to work with. In most cases, I am still making the decisions and doing the analysis. AI tools simply help me complete my tasks that much more efficiently.