AI in Product Management
What should you know in a noisy space like AI? How can you use AI as a Product Manager, and how much do you need to know?
This post was originally published in my Substack Newsletter, Roadmap Weekly.
The changing tide of AI
You don’t need to be an expert in the field of AI to recognize the rapid growth happening with AI technology. You may wonder if it’s at a tipping point yet and how it will impact various industries. Will AI leave a wake of disruption in its path? Maybe you’ve already been experimenting with Midjourney, DALL•E, ChatGPT, and others.
I’ve been experimenting with AI-generated text, images and voice since 2022. Even the voiceovers for these posts are made with AI. I also created a voice-to-text and text-to-voice conversational AI app. A year before that, AI wasn’t even on my radar, and it probably wasn’t something you paid much attention to either.
Should you learn AI?
Like many things, we must ask ourselves if we should be concerned with the advancements in AI. We should evaluate AI’s role in our lives and careers, both positive and negative, leveraging it for what we can. AI is more than just generative text and images, though.
AI is a large field of study, including computer vision, machine learning, deep learning, robotics, natural language processing, and neural networks. I’m no expert in AI, but you best believe I’m paying attention to this technology and working with it daily.
This recent surge of AI has given me a desire to dive deeper into machine learning and NLP beyond using the existing tools. I wanted to understand how these tools were built and to have experience building them myself.
The basics of AI are getting easier
I just finished a 5-week course on machine learning through Coursera produced by Duke University. Some of it was over my head, but this course was specifically for Product Managers. So, you can probably understand my surprise when the final assignment had us create a machine-learning model from scratch!
This seemed like a lot to ask of a Product Manager, even a more technical PdM. However, it was surprisingly easy. You can now run a multi-linear regression model in Excel! And it’s barely more difficult than creating a pivot table or chart. Once you train the model, you can easily re-create the formula and apply it to future data to make predictions.
How much AI do Product Managers need to know?
But how do you know how much attention to give it? Where does it fit within your product management knowledge, and how deep does your knowledge need to go?
I find it helpful to look at Product Management knowledge and experience in different layers, such as:
- Basic Product Management (what you need to fulfill basic PdM responsibilities)
- Industry-specific, i.e. the industry of your company
- Seniority-related, i.e. Sr. PM, Lead, Group, Director, VP
- Focus area, i.e. Growth PM, Pricing Strategy, Operations, Platform, Integrations
Where does AI expertise fit in? If your company is in the AI industry, but you’re a Growth PdM, your level of required knowledge is different than the Platform PdM at that same company.
Regardless of where you are, you should learn the basics. By now, you should at least:
- have experience with ChatGPT, Bard, or Claude
- have used an image-generating tool like DALL•E or Midjourney
- experimented with the potential uses of these tools
- get experience with and learn about prompt engineering
If your role is very data-heavy, you should become comfortable building basic ML models for data analysis and prediction.
If you’re working with technology in the AI space, what you need to learn will vary depending on what you’re working on. Go as deep as you want, but keep in mind your role as a Product Manager and go wide before going deep.
AI as a product management tool
Don’t forget that you can also leverage AI as a tool to improve your efficiency. In 2022, I used natural language processing to automatically classify and flag tickets, saving my team at least a dozen hours a week and improving our response time significantly. I’ve also experimented with an AI Google Sheets plugin to analyze sentiment in customer reviews. And who hasn’t summarized a document at least once with ChatGPT?
Think about how you can leverage AI in your Product role to give you superhuman powers. Can you summarize documents, analyze video calls, organize your schedule, perfect your writing, automate data analysis, or make things more creative and fun?
How to learn new, multi-dimensional things like AI
Learning new things is essential, especially as a Product Manager, but sometimes we get overwhelmed by the sheer volume of things to learn. To help, it’s worth evaluating your existing skills; you might know more than you realize. Once you have a baseline, identify the areas you want to improve and then implement a learning plan using SMART goals. Remember, it’s important to implement that knowledge quickly when learning something new.
I’m a huge fan of what I call just-in-time learning. Find a problem in your daily workflow or within your product that AI might help solve. Then, learn everything you need to make the necessary decisions or build the tool yourself. There are lots of great no-code options out there to help enable you to do this without bringing in engineering, especially if it’s an internal workflow issue.
Ultimately, let the learning guide you. Focus on finding your starting point, and as you learn, you will uncover new learning opportunities.
Check out this helpful template I built for paid subscribers that will help you track and document your skills and learning objectives. I like to create a new document like this every quarter to track my progression.
Skills matrix and learning plan
Will AI destroy the fabric of society?
Educator, Author and Media Theorist Neil Postman put this well in his book Technopoly: The Surrender of Culture to Technology, saying, “It is a mistake to suppose that any technological innovation has a one-sided effect. Every technology is both a burden and a blessing; not either-or, but this-and-that.”* Worth a read if you want to ponder more about the future of our culture.
Like everything, time will tell what lasting impact AI has on our jobs, society, and culture. I like to see it as a tool; like any tool, it can be used for good and evil.
Will you prevent the AI uprising and save us all when Terminators turn against their creators? Hopefully, we still have a few decades before that becomes an inherent risk. More importantly, how are you using AI now, and are you investing time in learning so you don’t get left behind over the next two to five years?
*Neil Postman (2011). “Technopoly: The Surrender of Culture to Technology,” p.14, Vintage
This post was originally published in my Substack Newsletter, Roadmap Weekly.
