Reading Time: 5 minutes

This is something I’ve written about before—delving into a deep analysis of my blog’s performance, sharing the results, and identifying the changes I needed to make. If you’ve ever tried it, you know it can get complicated and a bit messy, mostly due to the way WordPress saves and organizes information and performance statistics.

I knew there had to be a better way. This challenge turned into a perfect case study for how you can use AI as a productivity assistant to handle the heavy lifting.

When I first explained what I was trying to accomplish and asked how to gather the right data, the AI recommended I create a token for an API call. (Unsurprisingly, this was not a suggestion I got when asking the built-in WordPress AI assistant.)

How the Process Wrapped Together

Here is where it gets even better. Not only did the AI reveal this approach, but it also walked me through creating the token step by step. Once I posted the token and link into our chat, the assistant wrote custom Python code that I could run directly in Google Colab.

A Quick Side Note: I didn’t even know Google Colab existed before this! It is a fantastic development tool. It lets you play around with Python code seamlessly, eliminating the need to install Python locally on your laptop.

Using tools like Claude and Gemini, the AI ran the code and produced three downloadable files containing all the raw information I needed. From there, it executed a full blog analysis, which you can see below, complete with charts and graphs.

What the Data Revealed

Thanks to this analysis, I was able to determine the absolute best days and times to post specific content. On the flip side, it also highlighted the dead zones—days where my posts just drift into the void with virtually no return on the work I put in. Moving forward, those are days I will either eliminate entirely or use very infrequently.

More importantly, the data provided incredibly valuable validation: my work is growing month over month across all key categories, particularly in comments and likes. That’s exactly what we want to see, and it’s backed by a steady subscriber base that continues to add a few new readers each month.

The Bigger Picture

This experience really highlights the true power of these advanced tools. But it also proves something else: this insight doesn’t just happen automatically.

As the creator, I still had to know what I wanted to achieve, how to direct the process, what to ask, and exactly how to ask it. These tools didn’t replace me, nor would they replace a human worker; instead, they made me significantly more efficient.

Ultimately, that is the real goal of technology like Claude and Gemini. While AI will undoubtedly replace certain repetitive jobs, its greatest value lies in its human-multiplier effects—making people vastly more efficient once they learn to use the tools properly.

NotheBookLM

The Strategic Final Piece: Mapping the Future with NotebookLM

Once I had the raw data and charts from my Python analysis, it was time for the final piece of the puzzle. This is where I brought in NotebookLM to act as my strategic coordinator.

I loaded two critical puzzle pieces into the tool: my historical performance data and my upcoming schedule of drafted posts. With those sources uploaded, I could treat the AI as a productivity assistant that actually understood the context of my specific blog. I started asking targeted questions based on the hard numbers and what I had planned:

• Based on past engagement, what should I move around or change by day and content type?
• Which scheduled posts are sitting on days that historical data says will just go into the void?
• How can I reorder this specific lineup to hit the peak audience times we discovered?

Utilizing these tools in tandem allows me to objectively evaluate what I did in the past and plan what I want to do next. It completely changes the game—allowing me to plan smarter, save time, and focus on growing content that is genuinely meaningful and valuable to you, the audience.

Expanding the Workflow: Other Ways to Put This Tool to Work

As I went through this process, it became incredibly clear that simply using it for scheduling only scratches the surface. Once you have your data and writing loaded into a personalized notebook, there are several highly detailed ways to maximize the tool’s output:

• Identifying Unintentional “Content Gaps”: You can ask the tool to analyze your past 50 published posts alongside your future drafts to look for blind spots. For instance, you could ask, “Based on my writing history, what critical sub-topics or angles have I completely missed or neglected that my audience cares about?”
• Cross-Referencing Analytics with Tone: By uploading your readers’ comments alongside your analytics, you can ask the tool to identify patterns in sentiment. You can run a prompt like, “Compare the posts with the highest comment counts against the drafts I have scheduled. Am I matching the specific tone and vocabulary that generated the highest engagement?”
• Creating Instant Content Ecosystems: When you upload a comprehensive, data-heavy analysis, you can instruct the tool to break it down into smaller, bite-sized promotional pieces. I can ask it to generate three short newsletter blurb ideas or a quick summary to build a new post.
• Brainstorming the Next “Series” Dynamically: Instead of staring at a blank page for next month’s editorial calendar, you can ask the tool to look at your top-performing categories and generate logical sequels. A prompt like, “Based on the data showing high engagement for my ‘This Week in History’ series, what are 5 fresh, historically grounded angles I haven’t written about yet?” keeps the momentum going without rewriting the wheel.

Ultimately, this is how you turn a passive notebook into an active editorial partner. It’s not about letting the machine do the thinking—it’s about using the machine to give your own insights a massive head start.

Finally, just a side note to NotebookLM. Since I’m learning portuguese (or trying) I use chatgpt to create me a lesson plan by day. I load this into one of my notebooks and generate learning flashcards, audio, quizzes, and video files. It is very cool!

The full analysis, charts, and data breakdowns are detailed below.