See the previous blog post about Google’s March Algorithm Update? I used AI to start writing that for me – and took it from there.

Let me show you how I did it.

With AI, the prompts are the key.

First, I started with my favorite (at the moment) AI writing tool: perplexity.ai. No, I am not  a partner or paid to write about them – it just happens to be what I’m using – at the moment.

It’s critical to be looking at all the AI writing tools because their capabilities and strengths change by the hour. This means that I used perplexity today, but I may not use it tomorrow. If you’re going to use an LLM to help you write, you need to be testing all the tools, all the time.

For my blog post about Google’s March Algorithm Update, I started with the following prompt:

Please write a 600-word blog post about both the good and bad news about the March Googe Update, using my tone of voice from these two articles: https://twobeaglesmarketing.com/2025/02/21/4-tests-to-see-if-its-time-to-refresh-your-website/ and https://rachelhornstein.wpcomstaging.com/2025/01/24/dont-do-this-bad-b2b-marketing/

Notice how specific I am in this first prompt:

• I delineate exactly how many words to use because this affects the amount of detail provided
• I told it the exact theme I wanted
• I added samples of the type of writing style I wanted, using my own work

What I got back was good and full of facts but it read like an ad for Google, or a Press Release, so I followed up with this prompt:

Please rewrite this getting rid of the sales lingo, like “exciting features” – it needs to be more journalistic and third-party sounding, less of a pitch.

What I got the second round was better and some juicy information was alluded to, but not filled in. I wanted more details so I followed up with this:

Follow-up: please explain what industries have gained and what have lost, to give examples of this sentence: “The update has been particularly impactful for certain industries, with some sites seeing substantial gains in visibility while others face significant declines.”

That third version was the start of what became my blog post. I re-wrote what the AI provided, editing away all the repetitive phrases, the unnecessary adjectives and phrases, and getting rid of passive tense to make it more interesting.

When it comes to using AI as a tool, it starts with the prompts and ends with an editor’s critical eye. I hope this “behind the scenes” view of how I wrote the post is helpful to you as you figure out the best way to use these large language models. If you’d like some help with this, please reach out and let’s talk.