How AI Has Made Me Better And Efficient At Work
How AI Has Made Me Better And Efficient At Work
I don't think AI will replace me any time soon. After incorporating AI, I started to stand out from the crowd. Here is a list of 20 prompts that you can start using now.
Artificial Intelligence is a trending technology that many companies are trying to incorporate into their products. No wonder why, its ability to process and analyze data surpasses human capabilities. But this has raised concerns among product managers about the possibility of being replaced by AI in their roles.
It's true. AI has revolutionized industries around the world, since the last decades thanks to its ability to process and analyze large amounts of data and in a much shorter time than it would take a human to do so.
The following article in
in collaboration with , after all the analyses and experimentation, is clear that AI is making strides in performing certain tasks associated with product management but is still a considerable distance from fully replacing humans in this role. Its current utility lies more in assisting with specific tasks rather than taking over the entire role.
Another example, it has been almost 2 years since
posted this as a humorous way of the AI situation back then.
Although many would have thought in 2022 that this is impossible, we see how it has also become more than a trend. Like when people started adding Microsoft Office in their toolkit or on their CV. Now, knowing AI may become a must-have in job descriptions.
However, I see AI as a tool to enhance my skills and make me better at what I do.
For example, when it comes to handling large amounts for data analysis, it’s difficult for me to visualize myself or any professional trying to analyze many spreadsheets with thousands of rows at the same time. I’m not a spreadsheet manager, I know my time would be better spent doing something else. Specially, relevant to my job.
In the following article,
describes quite well this situation:As a product manager, staying ahead of the curve is essential, and AI is a powerful tool to help you achieve that. A recent LinkedIn survey shows 91% of professionals are incorporating AI into their daily workflows. Despite its widespread use, AI often remains an under-discussed topic among product people.
I think AI won’t replace Product Managers, but it will enhance them. At least, those who don’t want to be left out of the game. And I can see how others are also aware of it too. 👇
“AI won’t replace PMs. PMs who use AI will replace PMs who don’t.”
Product areas where AI can be useful
In this sense, AI opens the door to unmatched insights, streamlined workflows and customized processes. In my opinion, AI is most useful in product management competencies such as:
📊 Product planning is improved through data analytics for development optimization.
🗺️ Roadmap and backlog management is enhanced with data-driven insights for prioritization and resource allocation.
💡 Benchmarking, Market and Data Analyses are gained to identify patterns, trends, and needs for better decision-making.
*Does not mean that all the activity or responsibility is being taken away from PMs.
However, I consider AI is less useful in areas such as Leadership and People management:
🤖 AI is less effective in empathetic engagement with the customer or stakeholders, where intuition and human emotions are important.
🤝 Relationship building is a challenge for AI as it cannot fully replicate personalized interactions and understand human nuances.
💡 Even if the AI lacks the human touch and empathy, it can still be useful by providing ideas on how to convey a message and communication tips.
How I use AI to be more productive
I need to admit, I have increasingly embraced the power of AI to improve my performance and efficiency. It's like a framework that you rely on but have to adjust from time to time, because every situation is unique, but at least you're not starting from scratch.
But before we look at the prompts I use, let's recap the basic structure of AI prompts so you can come up with better ones later. 👇
A prompt structure that works
The first time I used ChatGPT of course my requests were very limited and most of them were simply questions because I was trying to understand what I could get.
And I think that’s how much you can get things done at the beginning, trying to come up with better results over and over again, until AI starts to understanding you even better.
💡 When you need AI help for a task, it is even useful to use a structured approach to get better results. For AI, it easier to answer when it has clarity on:
Context - Add details to the task: tone, personality, voice, etc.
Instruction - Define the task it needs to complete.
Input - What else should AI consider? Provide examples.
Output - How AI should display the results? Format, structure, etc.
1. Context
It's like giving a brief introduction or summary about the topic or problem. Set the stage for the AI, provide background information or context about what you need help with.
Why it’s important: Giving context helps the AI understand the scope of the problem and any relevant details that might influence its response.
Example: "I'm trying to increase the open rates of my marketing emails. My current open rate is 20%."2. Instruction
Here, you tell the AI exactly what you want it to do. It's a specific command or request that guides the AI in how to assist you.
Why it’s important: Clear instructions make sure the AI knows what form of help or solution you're looking for.
Example: "Analyze the factors that could improve email open rates."3. Input
This is the detailed information or data you provide that the AI needs to perform its task. Depending on the context, it could be a set of numbers, a piece of text, images, etc.
Why it’s important: The input serves as the raw material the AI works with. The quality and clarity of the input directly affect the output's accuracy and relevance.
Example: "Current subjects lines used, time of email sends, email segmentation strategies in place."4. Output
This is the expected result or the kind of response you're hoping to get from the AI. It can be specific answers, a summary, a generated text, predictions, etc. This is a great opportunity to indicate if you need the output in a specific format.
Why it’s important: Defining what output you expect helps in evaluating whether the AI's response meets your needs and guides the AI in structuring its response appropriately.
Example: "Provide a list of recommendations for improving email open rates based on the given factors."When you incorporate this basic structure in your prompts, you help the AI understand precisely what you need, allowing it to provide more accurate, relevant, and useful responses.
AI for repetitive tasks and workflows
As a product manager, my days often start before I even reach my desk or turn on my laptop. Over my morning coffee, I check the latest insights or recaps from calls that were in a different timezone. To do this, I rely on Teams Intelligent recap feature.
It's pretty straightforward, it's natively on most calls if they are recorded (yes, make sure someone always records), and I don't need to watch the entire 1 hour video I missed. Goodbye FOMO.
To make sure I don't miss anything, I like to set up custom flows with Power Automate. In my company, we all use Microsoft, but I am very happy that this tool comes in the package because I can set up everything I want to stay up to date or even handle large amounts of data.
In this case, I have a flow where I receive an email every morning with my action items that I did not close yesterday and mentions or new tasks from other conversations.
Also, 70% of my communication on a daily basis is in English. My organization is spread across the world and knowing how to communicate with all of them is essential. But something I am incredible thankful is how helpful AI has become in this aspect.
I'm not going to lie, I use DeepL translator almost EVERY DAY. At the slightest doubt of a well-structured sentence, I consult DeepL to check if the words I have in mind are in tune with the message I want to say.
AI for Product Management activities
Last week I wrote about how many of the responsibilities and activities of product managers involve writing. This is where I think AI becomes an incredible companion.
Let’s see these examples of how AI helps me with tasks that involve gathering requirements, documentation, and more. 👇
📑 Product Documentation and Help Articles
AI can generate first drafts of product documentation and knowledge base articles from feature lists and technical specifications, drastically reducing the time it takes a person to create them.
Of course it takes some refining after AI can produce user-friendly and comprehensible content, which I then review and refine even more, yet I’m not afraid of the blank page anymore.
Example: "Create a step-by-step guide for new users on how to set up their [Product Name] account and start using the main features."🗒️ Release Notes
Writing release notes and change logs can be time-consuming as well, especially for products that update frequently (or buggy 🐛). I use AI to automate the initial creation of these documents.
In my case, I feed Copy.ai with exported data about the latest product updates, fixes, and new features from Jira.
Example: "Draft release notes for version [version number] of [product name]. Highlight new features, improvements, bug fixes, and any known issues. Keep the language user-friendly and include instructions for downloading the update."📊 Market Analysis
AI tools streamline this process by automatically collecting and analyzing market trends, competitor activities, and consumer feedback across multiple channels. It’s a good starting point.
It provides summaries and identifies patterns, enabling me to focus on strategizing based on the insights rather than being bogged down by data collection and initial analysis. Which we already know that researching anything can become tedious and almost endless!
Example: "Provide a detailed market analysis for [product/industry], including identification of top three competitors, analysis of current market trends over the past year, and a summary of consumer preferences based on recent surveys. Present the data in a structured report with bullet points for key findings and a summary paragraph."💡 It is always good to ask the AI for valid sources so that you can have a look and compare the accuracy of this information. Don't rely totally on the first results it gives you.
🛠️ Feature Specification Docs
The good thing of AI is that it can help organize thoughts, suggest phrasing, and ensure technical accuracy based on existing product documentation and industry standards.
Example: "Explain the purpose and benefits of the [Specific Feature] in [Product Name], and how users can leverage it to achieve their goals."👤 User Stories
Creating user stories and scenarios involves understanding and articulating scenarios from the user's perspective. AI can suggest realistic and relevant user stories that might not have been immediately obvious.
Example:
Request for User Stories: [Specific Functionality or Feature]
Background: [Briefly describe the purpose or the context of this feature. This could include any problems it aims to solve or the value it intends to add.]
Target User Persona: [Describe the user persona for whom you're writing these stories, including their role, goals, and any relevant characteristics. This helps in tailoring the stories to the user's needs.]
Key Features to Cover:
1.[List out the main features or functionalities that you need covered by the user stories.]
2.[Provide any specific details or conditions you want each story to address.]
Constraints or Requirements: [Mention any specific constraints (e.g., technical, regulatory) or non-negotiable requirements that must be considered in the user stories.]
Desired Outcome: [Explain what success looks like for these user stories, including how these features would impact the user or the system.]Requesting AI to consider error handling and potential scenarios where things can go wrong can lead to a more user-centric product development process. This approach helps me considering a wider range of use cases and I can benefit development teams by providing more complete scenarios they may not have initially thought of.
In each of these writing-intensive tasks, AI dramatically reduces the time required to produce the first draft, allowing me to allocate more time to strategic thinking, creative problem-solving, and high-level planning.
You too can benefit from this by downloading a PDF I created with 20+ prompts. Use them and let me know what you think. 😉 👇
Areas where human touch is needed
To be honest, I would not rely on AI for any kind of communication that is in person and I don't think this is something that AI can improve on any time soon. Not as soon as many product managers would like.
What happens is that AI lacks a lot of the human touch. AI can't hold a conversation with your stakeholders on your behalf. You'll still have to attend those meetings and have difficult conversations with a lot of people. That's part of the job.
So AI cannot
❌ Have conversations with stakeholders for you
❌ Have a good product sense
❌ Generate ownership mentality
❌ Leading and guiding product teams
❌ Inspiring and motivating others
❌ Read beyond expressions in customer interviews
❌ Build relationships
❌ Be empathetic
❌ Make an organizational change
And so on…
It is important to recognize and appreciate the value of human skills and qualities in this role. Embracing these aspects of yourself will help you stand out and truly make a difference in the product management.
Remember, it's your individual experiences and skills that shape the products you create and the impact you have on the world.
Where I think AI can help, however little, is by providing structure, advice, drafts and perhaps incorporating communication frameworks to deal with those difficult moments.
💡 For example, you can tell to the AI how a stakeholder behaves and the background of this person: where does he comes from, how old is he, words he tend to use often, etc.
Then you can ask for a suggestion of what communication framework would be suitable in this scenario.
After that you can play and ask for examples of how to address the situation.
Example:
AI Prompt Template for Difficult Conversations with Stakeholders
Background Information on the Stakeholder:
- Demographics: [Insert age, gender, and any relevant background information]
- Professional Behavior and Communication Style: [Describe the stakeholder's behavior and preferred communication style]
- Commonly Used Phrases: [List phrases that the stakeholder uses often]
- Current Situation: [Provide context on the current issue or decision that requires discussion with the stakeholder]
Objective:[Clarify what you intend to achieve through the conversation with the stakeholder]
Ask for a Communication Framework Suggestion:
Given the information about the stakeholder’s background and communication style, I would like the AI to suggest a suitable communication framework that would effectively address the conversation objectives and align with the stakeholder’s preferences. The strategy should focus on [Insert specific focus areas, e.g., building trust, presenting data, managing risks].
Request for Example Phrases or Strategies:
Please provide example sentences or questions that would likely resonate with the stakeholder, addressing areas such as [Insert areas to focus on, such as financial outcomes, risk management, etc.]. Additionally, suggest strategies for presenting data or arguments in a compelling way to the stakeholder, and recommend techniques for keeping the conversation constructive if met with resistance.Using AI correctly depends a lot of the individual’s creativity, if you ask me.
The result is a good starting template that can make you feel more comfortable when faced with difficult situations or communication barriers. But the AI won't be there to defend your position. Be prepared, but also be flexible.
Final thoughts
Overall, can excel at decreasing the time it takes you to generate a lot of the documentation that is part of the product management job. In addition, with the right tools, it is useful in activities that require analysis such as:
📊 Product planning is improved through data analytics for development optimization.
🗺️ Roadmap and backlog management is enhanced with data-driven insights for prioritization and resource allocation.
💡 Benchmarking, Market and Data Analyses are gained to identify patterns, trends, and needs for better decision-making.
AI can be a helpful tool in improving communication skills, but it should not be solely relied upon. It can provide guidance, suggestions, and even simulations of conversations, but at the end of the day, it is up to us to effectively communicate with others.
Human touch, empathy and emotional intelligence remain crucial aspects of effective communication that AI cannot fully replicate. So while AI can be a valuable resource, it should be used as a complement, not as a substitute for humans.
I’ll close with a great quote I read in
post:"You know what the biggest problem with pushing all things AI is? Wrong direction. 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
What are your thoughts on this? What do you think of AI and how it relates to your daily work activities? Drop your thoughts in the comments! 👇
This one has everything Elena! Great post. Could print this out and turn it into a nice little book to pass out to your office friends. And thank you for the shout!
Great post!
Two things I'd add:
1. AI's terrible with negative prompts. For example, "Answer this question but make sure not to use more than 50 words" or something like that -- very prone to hallucination. Keep things 'positive' as much as possible.
2. I saw somewhere (Harvard Business Review? Not sure) that the term 'knowledge worker' is transitioning to 'wisdom worker' because of AI. I find that quite accurate and relevant to your last point (on what AI doesn't do). Our PM work might get easier thanks to AI, enabling us to skip some of the 'knowledge' part. But that only means we need to focus on 'wisdom' to make a difference!