Can AI Reduce My Repetitive Tasks?
Let's see how much of my life I can automate!
I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.
Joanna Maciejewska
This tweet from author Joanna Maciejewska has been living rent-free in my mind for over a year. It presents a larger existential question that I definitely won’t be able to answer anytime soon, but it did inspire my experiment for this week. Can I offload some of the more time consuming, but boring, tasks in my life to AI in order to free myself up to do more of what I love?
Agentic AI may be one potential solution, but we are week five into our AI era over here, so I decided to start smaller with Gems for this week’s experiment. The use cases feel endless, but let’s dive into the first few I tried.
The Challenge: Offloading Time Consuming, Basic Tasks
As much as I genuinely love being a Product Marketing Manager (and Podcaster, for that matter), a lot of my time is spent on the laundry list of tasks that keep me from the creative work I love. These aren't the strategic, creative challenges that drew me to marketing. They're the equivalent of doing dishes while my whiteboard sits empty. I don’t mind these types of tasks, but the more time I spend on them, the less time I have for the larger, more impactful projects.
What do I mean by little tasks? Across all my responsibilities, here are a few things I would love to spend less time on:
Writing Slack summaries for new content I create
Writing podcast show notes
Reviewing long-form written content for small details
Checking in on recent competitor news updates
Identifying key legislative activities that could affect our customers
To start, I wanted to focus on tasks that I do on a regular basis that I’d love to spend less time on. To get an idea of how two different tasks could look, I started with the Slack summaries and competitor research.
The Solution: Creating Gems for Quick Tasks
Gemini Gems, ChatGPT GPTs, and Claude Projects are all AI bots that you can easily create to perform a specific task. Creating Gems felt like finally getting that robotic assistant to handle the mundane so I could focus on the meaningful.
We recently did a training on Gems at work (shoutout to the Interplay AI Council!), and they provided a really helpful template. They suggested creating a Gem with three specific pieces of information:
Role
Task
Output
So let’s put it to the test!
Experiment #1: Speeding Up Slack Summaries
Every time I share a new resource (blog, whitepaper, info sheet, etc.) with our go-to-market team, I include the following information about the resource:
What is this?
When should you use it?
How can you share it?
This helps the team understand the point of the resource and know how to use it. But by the time I finish creating something I'm proud of, having to write another summary feels like folding clothes when I want to paint.
So I created a Gem with the following guidance:
Role: You are a product marketing manager at Interplay Learning. You have a great rapport with the rest of the go-to-market team, and you work hard to make it easy for them to understand what new resources are and how they can use them.
Task: Whenever I send you a new resource (guide, blog, etc.) you need to create a short summary I can send in slack to announce the new resource to the rest of the company.
Output: The summary should be modeled after the following format.
What is it? [insert a one sentence summary of the resource that focuses on the main takeaway]
When do I use it? [insert the sales stage that this would be best for]
How do I use it? [this should be 1-2 sentences that sales can copy/paste into their email templates to share this resource out to prospects and customers.]
The tone for this summary should be informational and friendly. It's for an internal team chat, so no need to be super professional. It should sound like a human wrote it to share out to her team why we have a new resource.
With this, I have a Gem now where I can just give it a resource and it immediately creates the summary for me to share with my team.
Experiment #2: Keeping Tabs on Competitors
One thing I like to do is put together a monthly industry newsletter where I highlight any news that would be good for the team to be aware of.
I love the research part of my job, so this is a task I really enjoy. However, checking in on competitor news is a huge pain because, before AI, I would have to Google each competitor one at a time to see if anything new popped up.
So I wanted to see if I could create a Gem to help.
Role: You are a product marketer whose goal is to stay up to date on any publicly available news about competitors.
Task: Search the internet for any news articles that have been published in the last 30 days about [competitors].
Output: When sharing results, please group the articles by company and provide a URL to each news article.
This prompt means I get only recent news, and I can click on the links to read the articles myself.
The Results: What Worked and What Didn’t
Both experiments have definitely saved me time, but the Slack summary took a bit more refining to get me what I needed.
Results: Slack Summary Gem
The Slack Summary Gem took more refining because I really needed it to imitate my voice better. The first few times I used it, I ended up having to rewrite most of the summary, so it wasn’t much of a time saver. What finally solved it was adding actual examples from my past work to the Gem prompt. That gave the Gem something to model itself after. I still have to do light revisions and tweaks, but now that it has good examples, it definitely saves me time.
Ultimately, this use case probably saves me around 10-15 minutes every time I share a resource, which adds up!
Results: Competitor Research Gem
This one was a home run right away. I double-checked with manual research, and Gemini was catching all the articles I would have found on my own. This kind of research is more exploratory for me, meaning that I just want to know what is out there more than I want to answer a specific question. So by using AI to give me all the relevant links in less than 30 seconds, I have more time to read the articles and just learn what is going on in the industry!
My only real note with this use case is that it is possible that it would miss something key. Because I do this monthly, I feel pretty confident I’m catching the most important news, but if you are using this type of research on something that it is critical that you don’t miss out on the details, double checking is always a good call!
Conclusion
While my actual laundry basket remains untouched, I'm slowly building a digital assistant for the professional equivalent. The trick is figuring out which tasks are worth handing off and then being patient enough to refine your prompts until they actually work (looking at you, Slack summary Gem).
Both of these experiments proved that AI can be a solid teammate for handling the administrative pieces, even if it takes some trial and error to get there. The question isn't whether AI will do our laundry, but whether we'll be intentional about which tasks we hand off so we can protect time for the work that matters most (but also…please do my laundry AI).