How can a Product Manager be GenAI ready?

Product managers have always been the bridge between tech and business. But now, that bridge is evolving fast, courtesy – generative AI. In case you are in the product management profession and think of GenAI as “just another trend,” you are already quite far behind. GenAI for product managers today is reshaping how products are imagined, built, and scaled.

The good news for you? It is easier for you to become GenAI-ready than you think, that too, without diving deep into the technicalities of things. Here, we break down exactly how to do that.

Let us start with the necessity of the entire exercise – why generative AI is needed for product management.

Generative AI – the new Norm for Product Managers

Why is Gen-AI needed for product management after all? Let me ascertain the necessity with an example here.

Coca-Cola, the world’s most popular beverage, now employs AI across operations. The brand uses AI not just for marketing campaigns, but to guide product decisions through real-time consumer sentiment analysis. To give you a gist, it now analyses data from social media, customer feedback, and regional sales trends.

This means AI helps Coca-Cola identify flavour preferences, and hence launch hyper-localised products and even optimise inventory by geography. A product manager at Coca-Cola can make faster, more confident decisions because AI is constantly feeding them actionable insights.

This is a norm across industries now. Users expect AI-enhanced features as default. Stakeholders are asking for “something ChatGPT-like.” And most importantly, your competitors are already experimenting with copilots, smart assistants, and auto-generation features.

Imagine a competing beverage company still relying solely on quarterly sales reports and manual surveys. Their feedback loop is slow, their response time is outdated, and their product launches often miss the mark. In a world where AI can help you spot, validate, and act on trends in real time, not using it is like showing up to a Formula 1 race with a bicycle.

You don’t want to ride a bicycle on the track, do you? So let’s dive right into your next racecar – generative AI.

Understand GenAI as Your Own Product

Think of GenAI as your own product. You wouldn’t ship it without knowing exactly what it’s great at, where it beats the competition, and what it is simply not meant for. Allow me to shine some light in that area for you.

What GenAI does really well?

• Generate Content: It is right in the name – consider this as the primary strength of generative AI. It can possibly produce content on any topic, across formats. Think emails, tooltips, release notes, UI copy, FAQs, even SEO text. As a PM, you can use it to move faster across documentation, prototyping, and user communication, saving massive time from ideation to rollout and feedback.
• Rapid Ideation: You will hardly find anyone as smart (definitely not as fast) a partner for ideation. A simple query or prompt can yield you tons of ideas across areas where you seek a fresh perspective. It feels like having an always-on brainstorming buddy with infinite post-its.
• Deep Research: Modern GenAI tools can perform extensive research in a matter of minutes. As you gear up to introduce your next product in the market, it can possibly tell you any and every similar product rollout in the entire history, giving you key insights on the best practices and the failures you can learn from.
• Simulation and Testing: Generative AI can mimic personas. This basically means that it can roleplay as a confused first-timer or a power user trying to break the system, helping you stress-test the UX before it ever reaches your real users.
• Personal Assistant: This is the most sought-after use of generative AI, to manage the menial and tedious tasks that eat up your precious time. In your everyday tasks as a product manager, you can use it to organise messy meeting notes, customer interviews, support logs, and whatnot, saving hours of mental bandwidth. Meaning, you focus on decisions, it takes care of the documentation.

What it can’t do well?

With all the pluses, there are some shortcomings. Generative AI, in its present state, faces a few struggles, for instance:

• It can’t perform complex, step-by-step reasoning as well as humans do.
• It doesn’t truly understand your user’s intent. It can guess, but not think as they do.

This basically means that as a product manager, you can treat GenAI like a product partner. You should know when to lean on it and when to put guardrails in place.

Learn the GenAI Language (No PhD Required)

Now that you know how generative AI can help you, you’ll need to learn how exactly to put it to use. For that, learning the language of GenAI is super important. Here is what you need to focus on:

Prompt Engineering

For instance, at the most basic level, you will need to learn prompt engineering. Context – a prompt is the query or the direction you provide to your AI tool. For example, you may ask ChatGPT to “write an email to the team for a meeting at 5 pm.” Though this is a very basic example, your prompts will get more and more technical in nature as you increase your use of generative AI.

That is when you will need to know how best to write your query, for the AI to yield best results. Here is an example of a bad prompt and a very good prompt from the context of a product manager:

Bad prompt:

“Write some suggestions for improving user experience.”

Great prompt:

“You are a UX researcher for a SaaS analytics dashboard. Suggest 5 UX improvements for the onboarding flow of a first-time marketing manager. Keep it data-informed, and focused on reducing drop-off.”

Prompt engineering is nothing but learning the art of providing prompts to generative AI. You don’t really need to take a course for it. Simply read through our detailed guide on prompt engineering here, and you will be well on your way to giving highly specific and fruitful prompts with some practice.

Learn about LLMs

LLMs are Large Language Models – what you avidly know as ChatGPT and Claude. These are AI systems trained on massive datasets to understand and generate human-like language. You can read about LLMs in detail here.

As a product manager, you don’t need to train an LLM. Though you do need to understand how they work, what their limits are, and how fast they’re evolving. Knowing the difference between GPT-4, Claude, and open-source models like LLaMA isn’t trivia for you. It has a practical application – it helps you choose the right model for the right use case.

You see, while the world runs after the benchmark scores of different LLMs, the fact is that each LLM has its own area of expertise. This simply arises from the data fed to them while in training. That means a particular LLM may be more suited for your needs than others. As you try your hand on the various models available, you will eventually find your suit.

Know the AI Lingo

Part of a product manager’s job is to coordinate across leadership and departments. In such meetings, you should be able to talk to your engineers, vendors, and leadership without sounding lost. That is exactly why you need to know, at the very least, the meaning of some keywords associated with generative AI. Some of these are:

These elements can directly impact your product’s speed, accuracy, and UX. Once you know them, you will know all areas for improvement.

Rethink User Experience with GenAI in Mind

Generative AI has changed the UX game already. In case you think any differently, let me just honestly and boldly tell you here that you are wrong! The old product flows just don’t apply when a user can just “ask” for what they want.

Look around, and it is easy to spot. Search boxes have turned into chat windows. Instead of typing keywords, users now ask: “What’s the cheapest flight to Goa next weekend with extra legroom?” GenAI assistants from Google, Bing, and countless other services spit out the answers instantly.

In Canva, users no longer click through icons. They just type “make a minimalist logo in green and black,” and the AI creates it. The interface is conversational now.

The change is not just digital. Samsung’s smart fridges now use AI to recommend recipes based on what’s inside. Even BMW is rolling out GenAI-powered voice experiences that can explain dashboard alerts, answer follow-up questions, and handle natural conversation, far beyond the old “set temperature to 22” era.

So if your product still expects users to tap through endless tabs or menus just to get something done, well, I think you can make an educated guess.

As a product manager using GenAI, you will need to rethink interfaces, user journeys, and error handling in a world where outputs are probabilistic, not deterministic.

Lightning-fast Prototypes: With APIs

AI accessible today has evolved to the point that it can itself act as the implementation tool, for itself. Meaning, no more waiting for a full tech team to build an AI feature. Tools like OpenAI’s API, Claude, LlamaIndex + LangChain, let you prototype GenAI features in hours.

Want a content suggestion tool inside your product? Build a demo with GPT-4 and a Notion frontend. This is where you don’t need to make an excuse or have patience to bring a whole new feature. Simply build the prototype through these tools, and once it gets you the well-deserved applause, get your tech team onto building it in-house.

Start Asking AI-First Product Questions

The best GenAI-ready product managers have already shifted their approach. I am not sure if you have or not, but I am sure you would not mind learning from the best in your role. At Microsoft, product managers are now acting as AI trainers for agent-based products. Mondelez, known for its snacks like Oreo and Cadbury, is using AI to iterate and launch new food products faster. At PepsiCo, PMs leverage AI for real-time data-driven decisions in operations. You name a known brand, and AI is probably already a part of its product journey now.

If you wish to be included in this list, here are some questions you can ask about yourself and your brand that will help you align your needs with GenAI:

• What part of your workflow can be automated or enhanced by GenAI?
• Can you personalise the experience using user data + LLMs?
• How do you measure success when outputs vary?
• What’s the fallback when the model gets it wrong?

These questions will act as a roadmap for your AI implementation, or at the very least, will help you have a fair idea of how best to put GenAI to use in your organisation.

Be the Ethics and UX Gatekeeper

Remember, the use of AI introduces new risks – bias, hallucinations, and privacy. As a product manager, you are to build trust much more crucially than you are to build features. For this, you should put GenAI to use ethically and aptly as a product manager.

At different points of a user’s journey, own questions like:

• Are we exposing user data to an external AI model?
• Can the AI say something offensive or misleading?
• Should the user know they are interacting with a model?

Being GenAI-ready means thinking beyond features. It means building responsibly.

Conclusion

Being a GenAI-ready product manager doesn’t mean you need to code a model from scratch. It means you understand the possibilities, the risks, and the value it brings to the table. With the use of AI in your operations, you can potentially test fast, fail faster, and win super-big, all through products that make sense in an AI-native world.

So if you’re a product manager, change your job description today. Include: “understanding AI well enough to use it wisely.”

Because the best product managers won’t just adapt to AI. They will make it their edge and redefine what product even means.