<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Curtis Savage on Medium]]></title>
        <description><![CDATA[Stories by Curtis Savage on Medium]]></description>
        <link>https://medium.com/@curtis.savage?source=rss-bb62b1bb6ba2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*1h0jcj_ezksZfEzo7V9k2w.png</url>
            <title>Stories by Curtis Savage on Medium</title>
            <link>https://medium.com/@curtis.savage?source=rss-bb62b1bb6ba2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 05:49:36 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@curtis.savage/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[.]]></title>
            <link>https://medium.com/@curtis.savage/-0502de6b8dcd?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/0502de6b8dcd</guid>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Sat, 08 Aug 2026 17:34:33 GMT</pubDate>
            <atom:updated>2026-08-08T17:34:33.791Z</atom:updated>
            <content:encoded><![CDATA[<img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0502de6b8dcd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Context Rot: The Silent Killer]]></title>
            <link>https://medium.com/ai-for-product-people/context-rot-the-silent-killer-4853e230531f?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/4853e230531f</guid>
            <category><![CDATA[context-engineering]]></category>
            <category><![CDATA[aiforproductpeople]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[ai-product-management]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Sun, 28 Jun 2026 17:47:31 GMT</pubDate>
            <atom:updated>2026-06-28T17:56:20.605Z</atom:updated>
            <content:encoded><![CDATA[<h4>Why your AI gets dumber and what product teams can do about it</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FuqiOSzSLDECM-rvDAbIlQ.png" /></figure><blockquote><em>TLDR: AI performance degrades quietly over long conversations. It’s not a model flaw; it’s a product design problem. PMs who understand it can optimize user sessions and ship AI products that hold up in high-stakes domains.</em></blockquote><p>Your AI was sharp at the start of the conversation. Helpful, precise, following your instructions.</p><p>By message 15, it’s ignoring a constraint you set six messages ago. It contradicted something it said earlier. The output feels… off.</p><p>You might pass it off as a weak model.</p><p>But there’s a name for what just happened: context rot. And it’s silently degrading the quality of AI products every day — including yours.</p><h3>What is Context Rot?</h3><p>To understand context rot, you need to know about the context window.</p><p>Every LLM has a hard limit on how much text it can process at once. This is called the context window. Everything in a conversation fills it up: the system prompt, user messages, AI responses, retrieved documents, tool outputs, memory. When it gets full, something has to give.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N-LG0Ecnlr20_l8cqCUY7g.png" /></figure><p>But here’s the thing most people don’t know: performance starts degrading well before you hit the hard limit. It’s not a cliff. It’s a slow slide. Instructions from earlier in the conversation get deprioritized. The model loses precision. Outputs get sloppier.</p><p>Context rot is fuzzy, hard to reproduce in isolation, and easy to misattribute. Users feel it before they can name it. They just know the AI “started forgetting things.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Vpqapbvt20pzT6sx48oXtA.png" /></figure><h3>Why this is a product problem, not an engineering problem</h3><p>Engineers notice context rot in evals. But the root cause is often upstream, in the session design decisions that PMs own.</p><p>Long sessions accelerate context rot. Multi-step workflows crammed into a single chat thread accelerate it. “Do everything here” UX patterns accelerate it.</p><p>Most AI products are designed like chat apps, optimized for long, continuous threads. That’s the right mental model for human conversations. It’s the wrong mental model for AI agents doing precision work.</p><p>The session is a scarce resource. You have to design for it.</p><p>When you think about session design (how long a thread should run, when to save state, when to reset), you’re doing AI product management. Not just feature work</p><h3>Why precision domains get hit hardest</h3><p>Context rot isn’t uniformly painful. It lands differently depending on what your users are doing.</p><p>At one end of the spectrum: creative brainstorming, drafting, ideation. These tasks are forgiving. A slightly generic response or a missed nuance doesn’t break anything.</p><p>At the other end: financial analysis, legal review, medical documentation. These tasks are unforgiving. Precision is the whole point.</p><p>Here’s what context rot looks like in practice across high-stakes domains:</p><p><strong>Financial planning:</strong> A finance manager asks an AI to help build a Q3 forecast. In the first few messages, they set a critical constraint: no new hires this quarter. Twenty messages later, the AI’s recommended plan quietly assumes two additional headcount. The numbers still add up. The model didn’t flag anything. The manager presents it to their CFO. Not good</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kuP02H7B3U6yvnx3graWSg.png" /></figure><p>Often, the failure mode isn’t “the AI was obviously wrong.” It’s “the AI was subtly wrong in a way that’s hard to catch.” That’s the worst kind of quality problem: the one that doesn’t trigger an error state, just erodes trust one interaction at a time.</p><p>The pattern: any domain where accuracy is non-negotiable and sessions tend to run long is a high-risk zone for context rot.</p><h3>How to detect it in your product</h3><p>You don’t need a sophisticated eval suite to start feeling context rot. Here’s how to find it:</p><p>Watch the token counter. If you’re using Cursor, hover over the pie chart beneath the chat box to see how full your context window is. Watch it fill during a long session and notice when output quality starts to shift.</p><p>Run a qual test. Start a long session on a precision task like analyzing a financial model or legal document. Check quality at message 5, 15, and 30. The degradation is usually visible within 20–30 exchanges</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gCPmCW8gIvKzbOJKT4hd8Q.png" /></figure><p>Know the signals. Context rot tends to show up as:</p><ol><li>Instructions being ignored</li><li>Earlier outputs being contradicted</li><li>Responses becoming more generic.</li><li>The AI “loses the thread” of a complex topic.</li></ol><p>Listen to your users. In research sessions, “the AI started forgetting things” or “it stopped following my instructions” are clear signals.</p><h3>What to do about it: six design patterns</h3><p>Context rot isn’t inevitable. Here’s how product teams can mitigate it:</p><ol><li><strong>Design for shorter, focused sessions. </strong>Give workflows natural end points. A budget review doesn’t have to happen in one infinite thread; it can be a series of focused tasks, each with a clear output.</li><li><strong>Make it easy to start fresh. </strong>Starting a new thread shouldn’t feel like losing work. If the value lives in saved outputs and structured data rather than conversation history, users can start over without losing anything.</li><li><strong>Use subagents for complex workflows.</strong> Break complex tasks into subtasks handled by focused sub-threads. Each one starts clean. Only the output comes back to the main conversation.</li><li><strong>Summarize and reset.</strong> At natural break points, save key outputs to a document and start a fresh thread. Some products can automate this; others can prompt users to do it manually.</li><li><strong>Load context progressively.</strong> Pull in only what’s needed for the current step, not everything that might be relevant. This is the principle behind RAG.</li><li><strong>Set honest user expectations. </strong>If long sessions degrade quality, tell users. “For best results, start a new chat when switching tasks” is a valid UX pattern.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aphVCclFMOAj-XZMaTNvqw.png" /></figure><h3>The deeper insight</h3><p>Context rot forces you to ask a question that most product teams haven’t explicitly answered: what is the right “unit of work” for your AI product?</p><p>A single message? A task? A workflow? A decision?</p><p>The answer shapes everything. It’s one of the most consequential UX decisions in an AI product, and it often gets made by default rather than by design.</p><p>It should guide:</p><ol><li>How you structure conversations</li><li>When you save state</li><li>How you present outputs</li><li>When you prompt a reset</li></ol><p>Your engineering team is already wrestling with these considerations at the infrastructure level. Context rot is the product-level manifestation of the same constraint. Now that you can name it, you can lead that conversation.</p><h3>Conclusion</h3><p>Context rot isn’t a model problem. It’s a product design problem.</p><p>The models will keep improving. Context windows will keep growing. But the fundamental tension (limited space, competing priorities, degrading performance under load) isn’t going away. It’s just going to play out at a larger scale.</p><p>The PMs who understand this will design better sessions, set honest expectations, and ship AI products that earn lasting trust, especially in the high-stakes domains where trust is hardest to build and easiest to lose.</p><blockquote><strong><em>Coming Next: </em></strong><em>RAG vs Memory: The PM’s guide to what your AI actually knows.</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GIxnmRaYOVZR89x-Bk_DCg.png" /></figure><p><strong><em>Curtis Savage works as a Senior Product Manager of AI in the FP&amp;A space and writes AI for Product People — a blog about the craft of building AI products.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Bq7UNp354pBubL6GRlLEvg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4853e230531f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/context-rot-the-silent-killer-4853e230531f">Context Rot: The Silent Killer</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Agents: From Black Box to Mental Model]]></title>
            <link>https://medium.com/ai-for-product-people/ai-agents-from-black-box-to-mental-model-8d197604c477?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d197604c477</guid>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[aiforproductpeople]]></category>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Sat, 20 Jun 2026 11:42:07 GMT</pubDate>
            <atom:updated>2026-06-22T04:56:32.342Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RBBbta9ANPHozArGk8dSzg.png" /></figure><blockquote><strong>TLDR</strong>: Every AI agent is built from the same six components. Once you can name the parts, you can decompose any AI product, ask better questions of your engineering team, and anticipate the technology instead of chasing it.</blockquote><p>You’re in a product meeting. Someone says “subagents” or “context engineering.” Everyone nods along. <em>Some</em> know what these terms mean. <em>Most</em> can’t explain why or how to actually use them.</p><p>The problem is the “AI Hype Machine” Most AI content is designed to induce FOMO, not to teach. “This model is INSANE” posts give you the buzz words without the mental model.</p><p>This post is that mental model. Every AI agent is made of the same six components and I’m going to give you six questions to ask (one per component) to understand any AI product. With this framework, everything else starts to click.</p><p>Let’s break it down.</p><h3>The six components of every AI agent</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*wZnVFYzDPdH3qdb2" /></figure><p>The six components of every AI agent</p><h3>1. The system prompt</h3><p>The system prompt is the developer’s layer. It’s set before any user ever shows up, invisible to them, and defines everything about how the AI is supposed to behave: its persona, its rules, its domain knowledge, its guardrails.</p><p>Think of it as the product spec baked directly into the AI. When ChatGPT declines to help with something, or when your company’s AI always responds in a certain tone, or when a customer support bot refuses to discuss competitors — that’s the system prompt at work.</p><blockquote>The PM question: <em>What should always be true about how our AI behaves, regardless of what the user asks?</em></blockquote><h3>2. Tools (and MCP)</h3><p>Here’s a fundamental truth that changes how you think about agents: LLMs can only produce text. They can’t browse the web, query a database, edit a file, or send a Slack message on their own.</p><p>When an AI takes action, it’s calling a tool. The LLM describes what it wants done, and the agent framework executes it and brings back the result. Most tools have familiar names — reading a file, searching the web, running a calculation. The “magic” of an agent is really just an LLM deciding which familiar operation to run next.</p><p>One term worth knowing: MCP (Model Context Protocol). Think of it as the USB standard for AI. Rather than every SaaS company building a separate integration for every LLM, MCP gives them one standard connector that works everywhere. If your product needs to connect to external data sources — your CRM, your analytics platform, your data warehouse — MCP is likely how that conversation will go with your engineering team.</p><blockquote>The PM question: <em>What does our agent need to do, not just say? What tools does that require?</em></blockquote><h3>3. RAG (retrieval augmented generation)</h3><p>RAG sounds technical. It isn’t. It means: “Before I answer, let me look things up first.”</p><p>Before generating a response, the agent searches through relevant documents, data sources, or knowledge bases and pulls in what’s most relevant to the current question. You’ve been doing this your whole life — before answering a hard question, you Google it, check your notes, ask a colleague. Agents do the same, just faster.</p><p>This matters enormously for product quality. When your AI gives a wrong answer, the instinct is to blame the model. But often the real problem is simpler: the agent didn’t have access to the right information. Before reaching for a bigger model or expensive retraining, ask: does the agent actually have the right data?</p><blockquote>The PM question: <em>Does our agent have the right information at the right time — or are we asking it to answer questions it was never given the context to answer?</em></blockquote><h3>4. Memory</h3><p>LLMs are stateless. Every new conversation starts from a blank slate. They have no memory of who you are, what you discussed last week, or how you like to work — unless you engineer that continuity explicitly.</p><p>Memory is how you solve this. It’s persistent context that gets automatically included at the start of every conversation — who the user is, what they’re working on, their preferences, standing instructions. In consumer products, it might be a user profile or a personalization layer. In coding agents like Cursor, it’s a file that gets prepended to every chat thread automatically.</p><p>Here’s the important thing to internalize: memory is just a text file. There’s no magic. It’s context that gets pasted in at the start of every conversation. Simple — but powerful.</p><blockquote>The PM question: <em>What should our AI always know about the user? What’s worth persisting vs. looking up on demand?</em></blockquote><h3>5. Conversation history</h3><p>Everything said so far in the current thread is part of the context. Every user message, every AI response — it all stacks up and gets included in what the model processes on the next turn.</p><p>This is why long conversations can feel different from short ones. The AI isn’t just responding to your latest message; it’s processing the entire history of your exchange.</p><blockquote>The PM question: <em>How long does a typical session run for our users? Are we designing workflows that stay within a healthy context budget?</em></blockquote><h3>6. Subagents</h3><p>A subagent is one agent spawning another to handle a subtask.</p><p>Instead of one long conversation that tries to do everything, the subagent pattern breaks complex work into focused chunks. The main agent delegates a specific task to a sub-thread, that sub-thread completes it and returns just the result, and the main conversation continues — leaner and more focused.</p><p>Think of it like a manager delegating to a team. Each person handles their piece, and only the output comes back up the chain.</p><blockquote>The PM question: <em>Where in our workflow is it worth spinning up a focused sub-task rather than loading everything into one long conversation?</em></blockquote><h3>Honourable mention: skills</h3><p>A seventh pattern worth knowing as the field matures: skills.</p><p>Unlike memory (always on) or RAG (retrieved per query), skills are capabilities loaded <em>progressively</em> — pulled in only when relevant to the current task. Think of it as RAG, but for instructions and capabilities rather than data.</p><p>Claude Skills is a good early example of this pattern. Rather than stuffing every possible instruction into a system prompt upfront, skills let the AI reach for specific expertise on demand — keeping the context window lean until it’s needed.</p><p>Skills are less standardized than the other six components today, but the pattern is growing. Worth keeping an eye on as you design your agent’s architecture.</p><h3>Conclusion</h3><p>You don’t need to write code to understand AI agents. You need a mental model.</p><p>These six components — system prompt, tools, RAG, memory, conversation history, subagents — are the building blocks of every AI product you’ll encounter, use, or build. Once you can name the parts, you can start to reason about tradeoffs, anticipate what’s coming, and ask better questions of your engineering team.</p><p>That’s AI product sense. And it starts with the mental model.</p><blockquote><strong><em>Coming next:</em></strong><em> the context window — the scarce resource that all six components compete for, and what “context engineering” actually means for product teams…</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fwOL2dTGvXgEgVp91ViegA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GIxnmRaYOVZR89x-Bk_DCg.png" /></figure><p><strong><em>Curtis Savage works as a Senior Product Manager of AI in the FP&amp;A space and writes AI for Product People — a blog about the craft of building AI products.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Bq7UNp354pBubL6GRlLEvg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8d197604c477" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/ai-agents-from-black-box-to-mental-model-8d197604c477">AI Agents: From Black Box to Mental Model</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Build Your Own “AI Product Manager in a Box”]]></title>
            <link>https://medium.com/ai-for-product-people/build-your-own-ai-product-manager-in-a-box-db3fc5d082a0?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/db3fc5d082a0</guid>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Tue, 01 Oct 2024 20:39:59 GMT</pubDate>
            <atom:updated>2024-10-01T20:54:09.031Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LVYJOiqXl_dRb2e8n3bdTg.png" /></figure><p>Welcome back to <a href="https://www.linkedin.com/newsletters/ai-for-product-people-7168228164102553600/">AI for Product People</a>, where we dive deep into the craft of building AI-powered products. If you’ve been following along, you know I’m all about finding ways to make Product Managers more efficient, more strategic, and, let’s face it — more superhuman. Today, I’m going to help you do just that by walking you through how to build your own “Product Manager in a Box.”</p><p>So, what exactly do I mean by “Product Manager in a Box”? I’m talking about a custom GPT — an AI assistant tailored specifically to your product, designed to help you with product strategy, definition, and execution across the full lifecycle. It’s your personal product management expert, capable of doing everything from generating user personas and feature ideas to building roadmaps and suggesting go-to-market strategies. Basically, it’s like having a hyper-efficient product team member on call 24/7.</p><p>And if you’ve been following my blog series, you’ll recognize that this assistant will help you across the five key steps of the product lifecycle, also known as the <a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=%2BjiXFo1aQu2sqTf%2BiTGuig%3D%3D">USIDO Framework</a>. (If you’re new here and need a quick refresher, check out the USIDO Framework to catch up on the fundamentals).</p><p>Ready to build your own? Let’s dive into the steps.</p><h3>Step 1: Run the App Store Review Analyzer GPT</h3><p>First, we need to load your GPT with some knowledge, starting with user feedback. For that, I’ve built the <a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer GPT</a>, the leading tool for analyzing App Store reviews.</p><p>Here’s why this matters: understanding user feedback is essential for product success. This tool will analyze your App Store reviews and provide insights into user needs, pain points, top requests, and key themes. It will even perform sentiment analysis to help you understand what users love (and, more importantly, what they don’t).</p><p>Simply enter your App Store URL, and the tool will take care of the rest. Once you’ve run the tool, save the output to a Word document — you’ll be uploading this file into your custom GPT later.</p><p>Click <a href="https://chatgpt.com/share/66fc3406-2194-8005-978c-80dfbfc3a12e">here for an example of the output</a> generated from running the tool for Spotify.</p><p>Action Item: <a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">Run the App Store Review Analyzer here</a>. Then, copy/paste/save the output into a word doc to be uploaded in Step 3.</p><h3>Step 2: Run the Product Manager in a Box GPT</h3><p>Next, let’s expand your knowledge base by adding more strategic insights. For that, you’ll run the <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">Product Manager in a Box GPT</a> — a tool that walks you through the <a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=%2BjiXFo1aQu2sqTf%2BiTGuig%3D%3D">USIDO Framework</a>. It will help you understand your product, strategize your approach, ideate features, define your roadmap, and optimize for success.</p><p>Just like with the first step, all you need is your app’s App Store URL. The GPT will take care of the rest by analyzing user feedback, performing competitor research, and suggesting key product enhancements.</p><p>Again, once the tool has done its magic, save the output to a Word document — you’ll upload this along with the review analysis in the next step.</p><p>Click <a href="https://chatgpt.com/share/66fc40c2-e09c-8005-8feb-f6bc7f7b6687">here for an example of the output</a> generated from running the tool for Spotify.</p><p>Action Item: <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">Run the Product Manager in a Box GPT here</a>. Then, copy/paste/save the output into a word doc to be uploaded in Step 3.</p><h3>Step 3: Create a Custom GPT</h3><p>Now, here comes the fun part — building your own custom GPT. This will be your product management assistant, loaded with the knowledge you’ve gathered from the previous steps.</p><h3>Here’s how to do it:</h3><ol><li>Go to <a href="http://chatgpt.com/">chatgpt.com</a> (you’ll need an account for this part).</li><li>In the left-hand navigation, select Explore GPTs.</li><li>Click the Create button in the top right corner.</li><li>In the Configure tab, give your GPT a name. For example, if you’re working on the Spotify app, you could name it “SPOT — Product Manager in a Box”. Note: OpenAI’s usage policy may prevent you from using an in-market app name. If so, use an abbreviation like “SPOT” instead of “Spotify”.</li><li>Add a description: “Expert product management assistant.”</li><li>Provide the following Instruction (Note: replace ‘Spotify’ with your App name)</li></ol><blockquote>&quot;You are a product management expert operating at a principal product manager level for {{Spotify}}.</blockquote><blockquote>Your primary role is to help {{Spotify}} product managers through the end-to-end product management lifecycle by performing research, analysis, and completing a variety of product management frameworks and tasks.</blockquote><blockquote>Start by asking the user if they have a specific question or would like to explore a topic from the menu below (delimited by triple quotation marks). Always redisplay the menu after completing your task.</blockquote><blockquote>### MENU ###</blockquote><blockquote>Please select a number:</blockquote><blockquote>1. Generate User Persona</blockquote><blockquote>2. Identify Biggest Pain Points</blockquote><blockquote>3. Perform Market &amp; Competitor Analysis</blockquote><blockquote>4. Perform SWOT Analysis</blockquote><blockquote>5. Recommend Product Features and Enhancements</blockquote><blockquote>6. Prioritize Product Features and Enhancements</blockquote><blockquote>7. Generate User Stories in Gherkin Format (for highest priority feature)</blockquote><blockquote>8. Generate Release Notes (for highest priority feature)</blockquote><blockquote>9. Identify Success Metrics (for highest priority feature)</blockquote><blockquote>10. Generate Go-to-Market Strategy with Marketing Copy (for highest priority feature)&quot;</blockquote><p>7. Add a Conversation Starter: “Start here”</p><p>8. Upload the two files you saved earlier (from Steps 1 and 2) to provide the GPT with all the context it needs about your app’s reviews and product insights.</p><p>9. Enable all capabilities (Web browsing, DALL-E, Code Interpreter), and click Create.</p><p>10. Et voilà — you’ve just created your very own Product Manager in a Box! You can try out the Spotify example <a href="https://chatgpt.com/g/g-aAvMMHWzz-spot-product-manager-in-a-box">here</a></p><h3>Screenshots:</h3><p><strong>Screenshot</strong>: Configure your GPT with a Name, Description, Instructions, and Conversation Starter</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/0*3qDnBs2g2isKahUR" /></figure><p><strong>Screenshot</strong>: Upload word documents from the first two steps</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/0*NOc4zEG12WUJ3M-h" /></figure><p><strong>Screenshot</strong>: Enable all capabilities</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/0*4jGqY1Z0K870RB51" /></figure><p><strong>Screenshot</strong>: View your published GPT</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*AwJlahagFm4gi7Ux" /></figure><h3>Step 4: Test Your GPT</h3><p>Now that your GPT is live, it’s time to test it. Start by running a few core tasks to make sure it performs as you would generally expect:</p><ol><li><strong>Generate User Personas</strong>: “Based on the user feedback and SWOT analysis, can you create a detailed user persona for our core user type?”</li><li><strong>Generate Feature Ideas</strong>: “What new product features should we build to improve user retention based on the user feedback and SWOT analysis?”</li><li><strong>Build a Roadmap</strong>: “Organize the feature ideas into a one-year roadmap, divided by quarters, and prioritize based on user impact and business objectives.”</li><li><strong>Write User Stories</strong>: “Write user stories in the Gherkin format for {feature X} based on the top pain points identified in user feedback.”</li><li><strong>Generate Success Metrics</strong>: “Generate a list of KPIs for {feature X}, focusing on [retention/engagement/revenue growth/etc.].”</li><li><strong>Go-to-Market Strategy</strong>: “Generate a go-to-market strategy for {feature X}, including key messaging, recommended channels, and a timeline for launch.”</li></ol><p>Need more feature ideas? Just ask. Want to optimize your launch strategy? It’s got you covered.</p><p>For example <a href="https://chatgpt.com/share/66fc452e-bb14-8005-ab9c-2056a6f13bf1">here is the output</a> from the first four prompts above for the Spotify app we created in the example</p><h3>Step 5: Run Your GPT</h3><p>You can also use the built in menu and simply follow the predefined steps in the menu.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oKlpM3U8uxXqs7OU" /></figure><h3>Final Thoughts</h3><p>By now, you’ve built your very own Product Manager in a Box — an expert AI assistant that’s ready to help you ideate, plan, and optimize every aspect of your product lifecycle. Whether you’re brainstorming new features or mapping out your go-to-market strategy, your custom GPT will be there to save you time and help you make more informed decisions.</p><p>It’s insights won’t be perfect but it gets you started and I’m willing to bet it’s a heck of a lot better than a blank page!</p><h4>Call to Action:</h4><p>If you’ve found value in this article, I’d love it if you shared your experience on LinkedIn and left a 5-star review for the GPTs in the store. Let’s spread the word!</p><ol><li>Rate it: <a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li>Rate it: <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">Product Manager in a Box</a></li></ol><p>That’s it for this edition of AI for Product People. Now, go build your GPT and start managing your product like a pro!</p><p>Learn more a about the USIDO Framework below:</p><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=db3fc5d082a0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/build-your-own-ai-product-manager-in-a-box-db3fc5d082a0">Build Your Own “AI Product Manager in a Box”</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OPTIMIZE Your Product with Custom GPTs]]></title>
            <link>https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/65d4476e7a0b</guid>
            <category><![CDATA[ai-product-manager]]></category>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[customgpt]]></category>
            <category><![CDATA[gpt]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Fri, 27 Sep 2024 05:52:06 GMT</pubDate>
            <atom:updated>2024-10-01T20:43:05.607Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WNvPiiBxHYsTRgDAg6Pc3A.png" /></figure><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><p>Welcome back to <em>AI for Product People</em>, where we dive deep into the craft of building AI-powered products. If you’ve been following along, we’re now at the final step of the <a href="http://USIDO Framework"><strong>USIDO Framework</strong></a> for AI Product Managers:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*37__NPMG1mbQdrX-.png" /></figure><p>There are 5 key steps across the product lifecycle in which a Product Manager can leverage AI to launch a new feature.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qBom2IyOQsdeLMKZ.png" /></figure><p>So far, we’ve covered the first four steps of using AI to:</p><ol><li><strong>UNDERSTAND</strong> your users.</li><li><strong>STRATEGIZE</strong> your product approach.</li><li><strong>IDEATE</strong> new product features.</li><li><strong>DEFINE</strong> your product roadmap and user stories.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5YyJK6VsuOqIYs5hylhJ7A.png" /><figcaption>The USIDO Framework</figcaption></figure><p>Now, it’s time for the final step: <strong>OPTIMIZATION</strong>. This is where you fine-tune your product before and after launch to ensure it meets user needs, drives growth, and delivers measurable business impact.</p><h3>The Role of AI in Product Optimization</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DA9Qv14IV4lB4Xtbf-fxkw.png" /><figcaption>Step 5: OPTIMIZE</figcaption></figure><p>Optimization is all about continuously improving your product through experimentation, measurement, and iteration. It involves perfecting your go-to-market strategy, identifying the right success metrics, and running experiments like A/B tests to ensure your product evolves based on real-world data.</p><p>By using a custom GPT, you can streamline this process, leveraging AI to generate optimized go-to-market plans, define success metrics, and create data-driven experiments to maximize your product’s success.</p><p>Here’s how AI can help you in the optimization phase:</p><h3>1. Optimizing Your Go-to-Market Strategy</h3><p>Your go-to-market (GTM) strategy is crucial to the success of any product or feature launch. It’s about defining how you’ll reach your target audience, what channels you’ll use, and how you’ll position your product in a competitive market. The GPT can assist by analyzing your market landscape and generating data-driven GTM plans that ensure your product gets the attention it deserves.</p><p><strong>Example</strong>: Let’s say you’re launching a new feature designed to improve user engagement. The GPT can analyze your target audience and competitors, and suggest the best GTM tactics, such as the ideal marketing channels, messaging, and product positioning to differentiate your product from competitors.</p><h4>Action Item:</h4><p>Visit <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a> and use this prompt: <em>“Based on our product’s new features and user needs, generate a go-to-market strategy that includes marketing channels, positioning, and key messaging.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UVSKtrSMQZM7BKNNaFpOMw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ScB1hrxTfVXwg6ueU8W8SQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wNv-yf0J0LNRrqiO5ASaXg.png" /></figure><p>The GPT will deliver a tailored GTM strategy, including recommended channels (email campaigns, social media, etc.), suggested messaging, and even potential partnership opportunities to boost visibility.</p><h4>Template Prompt:</h4><p><em>“Generate a go-to-market strategy for our [new feature], including key messaging, recommended channels, and a timeline for launch.”</em></p><h3>2. Identifying Success Metrics and KPIs</h3><p>Once you have your GTM strategy, the next step is to identify the key success metrics (KPIs) that will allow you to measure the product’s impact. Whether it’s retention, activation, or revenue growth, defining the right metrics is crucial to ensuring you’re tracking the success of your optimization efforts.</p><p>The GPT can help you identify and prioritize the right KPIs based on your business goals and product type. For example, if your focus is on driving user engagement, the GPT can recommend metrics like <strong>daily active users (DAUs)</strong>, <strong>session length</strong>, or <strong>feature adoption rate</strong>.</p><h4>Action Item:</h4><p>Use this prompt to define your success metrics: <em>“Based on our new feature aimed at improving user retention, what KPIs should we track to measure success?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9q8psEBc3FHVaxUS_qVc7A.png" /></figure><p>The GPT will generate a list of success metrics tailored to your product’s goals, giving you a clear understanding of what to track.</p><h4>Template Prompt:</h4><p><em>“Generate a list of KPIs for our [new feature], focusing on [retention/engagement/revenue growth/etc.].”</em></p><p>By identifying and tracking the right metrics, you’ll have the data needed to continually optimize your product’s performance.</p><h3>3. A/B Testing and Experimentation</h3><p>Optimization doesn’t stop with tracking metrics. To truly maximize your product’s potential, you’ll need to run experiments, such as A/B tests, to understand what works and what doesn’t. A/B testing allows you to experiment with different variations of features, user flows, or even marketing messages to see which performs better.</p><p>With the GPT, you can easily generate a set of A/B tests for different aspects of your product. Whether you’re testing different onboarding flows or messaging strategies, the GPT can help you quickly create hypotheses, define test groups, and set parameters for measurement.</p><p><strong>Example</strong>: Let’s say you want to test two different versions of your onboarding process. The GPT can generate variations based on your product data and suggest what to measure, such as completion rates, time to complete onboarding, or user activation rates.</p><h4>Action Item:</h4><p>Use this prompt to generate A/B tests: <em>“Based on our goal to improve onboarding, generate two A/B test variations for the user onboarding flow. What metrics should we track to measure success?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*auvMQA6N1eFiu954J5JJkA.png" /></figure><p>The GPT will output specific test cases, detailing the variations, expected outcomes, and metrics to track, such as <strong>conversion rate</strong>, <strong>time spent in onboarding</strong>, and <strong>feature engagement</strong>.</p><h4>Template Prompt:</h4><p><em>“Generate A/B test ideas for [product feature] and suggest key metrics to track for each variation.”</em></p><p>Running these experiments will allow you to make data-driven decisions, ensuring that your product is optimized for real user needs.</p><h3>4. Analyzing Results and Iterating</h3><p>Once your A/B tests and experiments are complete, it’s time to analyze the results and iterate on your product. The GPT can assist by quickly summarizing the data from your tests, providing insights into which variations performed better and why. From there, you can use this information to make informed decisions on the next iteration of your product.</p><h4>Action Item:</h4><p>After running your A/B tests, use this prompt to analyze the results: <em>“Analyze the results of our A/B tests on the new onboarding flow and provide insights into which variation performed better and why.”</em></p><p>The GPT will summarize the test outcomes, highlighting which variation achieved better results and suggesting further improvements based on the data.</p><h3>Final Thoughts: GPTs in Product Optimization</h3><p>The optimization phase is all about using data to make continuous improvements to your product. A custom GPT helps streamline this process by generating a GTM strategy, identifying success metrics, and automating the creation of A/B tests to ensure your product is constantly evolving.</p><p>To recap, the GPT can help you:</p><ul><li><strong>Create a go-to-market strategy</strong> tailored to your product and audience.</li><li><strong>Identify success metrics</strong> that align with your product’s goals.</li><li><strong>Generate A/B tests</strong> and other experiments to optimize your product.</li><li><strong>Analyze test results</strong> and iterate on your product for continuous improvement.</li></ul><p>With a custom GPT, you can accelerate your optimization efforts, making smarter decisions faster and ensuring your product evolves based on real data.</p><h3>What’s Next?</h3><p>Now that you’ve completed the <a href="http://USIDO Framework"><strong>USIDO Framework</strong></a>, you’re fully equipped to leverage AI across the entire product lifecycle. From understanding your users to optimizing your product’s performance, GPTs can help you work smarter, faster, and more effectively.</p><p>Stay tuned as we explore more advanced topics in AI product management, and don’t forget to share your experiences using the USIDO Framework with a custom GPT!</p><p><strong>Call to Action</strong>:<br><strong>Start Optimizing Now</strong>: Ready to optimize your product with the help of AI? Use <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a> to create A/B tests, identify KPIs, and generate GTM strategies — all in a few clicks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*l7nYHppK_soRPgKR.png" /><figcaption><a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box</a></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65d4476e7a0b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE Your Product with Custom GPTs</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DEFINE Your Product with Custom GPTs]]></title>
            <link>https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/56f4eaa60662</guid>
            <category><![CDATA[customgpt]]></category>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <category><![CDATA[user-stories]]></category>
            <category><![CDATA[chatgpt]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Fri, 27 Sep 2024 05:23:05 GMT</pubDate>
            <atom:updated>2024-10-01T20:42:47.884Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hnj6hYsaAoveBedwB2xzYg.png" /></figure><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><p>Welcome back to <em>AI for Product People</em>, where we dive deep into the craft of building AI-powered products. If you’ve been following our series, you know we’ve been walking through the <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">USIDO Framework </a>— a powerful tool for Product Managers to leverage AI across the product lifecycle.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*k-hofpf_9HG3fSqi.png" /></figure><p>There are 5 key steps across the product lifecycle in which a Product Manager can leverage AI to launch a new feature.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NDq6Q0ECBBS5AqHq.png" /></figure><p>So far, we’ve covered the first three steps of using AI to:</p><ol><li><strong>UNDERSTAND</strong> your users.</li><li><strong>STRATEGIZE</strong> your product approach.</li><li><strong>IDEATE</strong> new product features.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5YyJK6VsuOqIYs5hylhJ7A.png" /><figcaption>The USIDO Framework</figcaption></figure><p>Now that you’ve brainstormed and generated a set of features, it’s time to <strong>DEFINE</strong> them into a Product Roadmap and generate product requirements like user stories.</p><h3>The Role of AI in Product Definition</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AsGDzdsm8uulSQ7XmFiI8g.png" /><figcaption>Step 4: DEFINE</figcaption></figure><p>Product definition is where your ideas begin to take shape. It’s the phase where abstract features take shape into product roadmaps and user stories. Traditionally, this process takes a lot of time and coordination, but with a custom GPT in your toolkit, you can streamline it significantly.</p><p>To help you get started, use <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a> — a custom GPT that will help you take those rough ideas from ideation and turn them into actionable roadmaps and user stories.</p><p>Here’s how AI can take your product definition process to the next level.</p><h3>1. From Ideas to Roadmap</h3><p>Once you have a list of feature ideas, the next step is to organize them into a clear, prioritized roadmap. Your roadmap will act as a blueprint for how and when to build these features. This is where the GPT shines: it can analyze your backlog and prioritize features based on impact, user needs, and business goals.</p><p><strong>Example</strong>: Let’s say you’ve ideated several features aimed at improving user retention. The GPT can analyze your list, rank the features based on potential user impact, and help you map them out over the next year.</p><h4>Action Item:</h4><p>Visit <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a>, and use the following prompt: <em>“Based on the feature ideas from the ideation phase, what are the top five features I should prioritize for the next product release?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UivZ93XiIc-UI0AqFn4Yww.png" /></figure><p>The GPT will then generate a roadmap for your top features, helping you organize your development efforts.</p><h4>Template Prompt:</h4><p><em>“Organize the feature ideas from ideation into a one-year roadmap, divided by quarters, and prioritize based on user impact and business objectives.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qR8wYq-rRGiTj8usQ6aNDg.png" /></figure><h3>2. Writing User Stories and Defining Features</h3><p>Once you’ve built your roadmap, you’ll need to create user stories for each feature. The GPT can save you hours of manual work by generating these stories for you. With the context provided from your ideation and roadmap, the GPT can quickly turn each feature into a detailed user story.</p><p><strong>Example</strong>: Let’s say one of the features on your roadmap is a new onboarding experience. The GPT can generate user stories that capture the specific goals and needs of your users, which you can hand off directly to your engineering team.</p><h4>Action Item:</h4><p>Use this prompt to generate user stories: <em>“Write user stories in the Gherkin format for the onboarding feature based on the top pain points identified in user feedback.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OOvo1vyRPcb63dU-rud-Pg.png" /></figure><p>This helps ensure your features are clearly defined and ready for development.</p><h3>3. Define Edge Cases and Acceptance Criteria</h3><p>One of the trickier aspects of defining a product is identifying edge cases and writing detailed acceptance criteria. The GPT can help you brainstorm potential edge cases and write out the criteria for each feature.</p><h4>Action Item:</h4><p>Use this prompt: <em>“Generate edge cases and acceptance criteria for the new user onboarding feature.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vZVKsIY3Rac7HvHqJjVUhg.png" /></figure><p>This will help you think through all possible user scenarios, ensuring that the feature is robust and thoroughly tested before release.</p><h3>4. Automating Release Notes</h3><p>Beyond roadmaps and user stories, you’ll also need to create other key documents like release notes and stakeholder updates. A GPT can generate drafts for these documents, saving you time and ensuring consistency.</p><p><strong>Example</strong>: After defining your features and user stories, you can use the GPT to create draft release notes for your next product update, summarizing what’s new for your users.</p><h4>Action Item:</h4><p>Use this prompt: <em>“Generate release notes for the upcoming product update based on the features in the roadmap.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bh7Is5BhNilsjL8gK_-Cvg.png" /></figure><p>The GPT will give you a clean draft to share with your users, allowing you to focus on refining and personalizing the message.</p><h3>Final Thoughts: GPTs in Product Definition</h3><p>Defining your product requires detail, precision, and careful planning. With a custom GPT, you can streamline this process, ensuring your product roadmap, user stories, and documentation are clear, actionable, and aligned with user needs.</p><p>To recap, a GPT like <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a> can help you:</p><ul><li><strong>Generate detailed roadmaps</strong> that prioritize the most impactful features.</li><li><strong>Write user stories</strong> and other key product documents.</li><li><strong>Define edge cases </strong>that may have been missed.</li><li><strong>Automate release notes</strong> and stakeholder updates.</li></ul><p>Using a GPT not only accelerates the process but also ensures your product definition is comprehensive and data-backed. This frees you up to focus on high-level strategy and creative problem-solving.</p><h3>What’s Next?</h3><p>With your product now well-defined, the next step in the USIDO Framework is <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b"><strong>OPTIMIZE</strong></a>. In the final step of this series, we’ll explore how AI can help you continuously experiment and optimize your product for growth.</p><p>See you there!</p><h4>Call to Action:</h4><p><strong>Start Defining Now</strong>: Ready to define your product with the help of AI? Use <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a> to generate user stories, roadmaps, and more in just a few clicks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*LB9pwD5hv5S8Hizy.png" /><figcaption><a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box</a></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=56f4eaa60662" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE Your Product with Custom GPTs</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[IDEATE Your Product with Custom GPTs]]></title>
            <link>https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/fd541b703f2f</guid>
            <category><![CDATA[custom-gpt-models]]></category>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Fri, 27 Sep 2024 04:32:53 GMT</pubDate>
            <atom:updated>2024-10-01T20:42:34.357Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EyANwcGYe-2X9xFAvYnoNw.png" /></figure><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><p>Welcome back to <em>AI for Product People</em>, where we dive deep into the craft of building AI-powered products. If you’ve been following our series, you know we’ve been walking through the <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">USIDO Framework</a> — a powerful tool for Product Managers to leverage AI across the product lifecycle.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DHWB2-gSemgajhyc.png" /></figure><p>There are 5 key steps across the product lifecycle in which a Product Manager can leverage AI to launch a new feature.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*pq_nJHo00QTFqKIM.png" /></figure><p>So far, we’ve covered the first two steps using AI to:</p><ol><li><strong>UNDERSTAND</strong> your users.</li><li><strong>STRATEGIZE</strong> your product approach.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5YyJK6VsuOqIYs5hylhJ7A.png" /><figcaption>The USIDO Framework</figcaption></figure><p>Now that you’ve got a solid understanding of user needs and a strategic direction, it’s time to move on to the exciting part: <strong>IDEATION</strong>.</p><h3>The Role of AI in Product Ideation</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L9IjmpnE07nmm4ZFMtAJ1g.png" /><figcaption>Step 3: IDEATE</figcaption></figure><p>Ideation is where the magic of AI truly shines. It’s the phase where you transition from understanding and analyzing data to actively creating new product features that address your users’ pain points. In traditional product ideation, this step could take weeks of brainstorming sessions, feedback loops, and research. But with a custom GPT in your toolkit, you can fast-track this process significantly.</p><p>To get you started, you can use our tool <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*__6dWunru7PzCrSlu-OwCw.png" /></figure><p>This GPT can analyze all the data you’ve gathered so far — user feedback, market trends, competitive insights — and synthesize these into actionable ideas for new product features. In short, it helps you generate ideas quickly and efficiently.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TDjq6vlzpLzhGdxvrYWYxQ.png" /></figure><p>Here’s how AI can take your ideation process to the next level.</p><h3>1. From Data to Feature Ideas</h3><p><a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a>, armed with your company’s data and strategic insights, is an idea-generating powerhouse. By analyzing your user reviews (from tools like App Store feedback) and strategic documents (like SWOT or market analysis), it can recommend features that align with both user needs and business goals.</p><p><strong>Example</strong>: Let’s say your app’s reviews reveal that users are consistently frustrated with a slow onboarding process. The GPT, informed by this data and your competitive landscape, might suggest building a feature that simplifies onboarding through personalized tutorials or AI-powered user guidance.</p><h4>Action Item:</h4><p>Visit <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a> and follow the steps sequentially.</p><p>Then use this prompt after you’ve generated the your user feedback analysis and market research:</p><p><em>“Based on the user feedback and SWOT analysis, what new product features should we build to improve user retention?”</em></p><p>The GPT will analyze your data and provide suggestions like those below (using Spotify as an example).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BjJb8SCK8aET4c_wBd4sfQ.png" /></figure><h3>2. Custom Prompts for Ideation</h3><p>The power of a custom GPT is in the prompts you feed it. If you’ve followed the previous steps of the USIDO framework, the GPT already understands your market, users, and product strategy. Now it’s time to harness that understanding for creative feature generation.</p><p>Let’s say you want to ideate around improving user engagement. You can use a structured prompt like this:</p><p><em>“Based on our analysis of top user complaints and our SWOT analysis, what features should we build to increase user engagement in our app?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l6FHbpTh9lUPR1H7x2HyVg.png" /></figure><p>With this prompt, the GPT will look at the areas where users are disengaged and align that with competitive gaps, suggesting features that boost engagement. These might include:</p><ul><li><strong>Gamified elements</strong> to encourage ongoing user interaction.</li><li><strong>Push notifications</strong> tailored to user behavior.</li><li><strong>Personalized content feeds</strong> to increase user stickiness.</li></ul><h4>Template Prompt:</h4><p><em>“Based on [the top user complaints from reviews] and [the biggest opportunities from our SWOT analysis], what product features should we build to [achieve our goals] for [target user segment]?”</em></p><h3>3. Ideating for Different Goals</h3><p>The beauty of this approach is its flexibility. Whether your goal is to drive adoption, reduce churn, or increase engagement, the GPT can tailor its ideas to fit your specific goals.</p><p>For instance, if retention is your goal, your prompt might look like: <em>“What product features can we build to improve retention for power users who are showing signs of churn?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rTPXh1zrMBivYkdj4CgdAQ.png" /></figure><p>On the other hand, if you’re looking to increase activation rates for new users, you might prompt the GPT with: <em>“What product features should we build to improve activation for new users in their first week?”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cUMy81KL84eDPXTCjrNXAQ.png" /></figure><p>The GPT will help you come up with feature ideas based on existing insights, helping you quickly generate a targeted list of actionable improvements.</p><h3>Final Thoughts: GPTs as Your Ideation Engine</h3><p>A custom GPT doesn’t just streamline ideation; it also injects fresh, unbiased thinking into your process. By synthesizing vast amounts of data, it can surface feature ideas that may not have been immediately obvious to your team. This allows you to innovate faster and make more data-driven decisions when designing new features.</p><p>To recap, GPT can help you:</p><ul><li><strong>Synthesize user feedback and competitive analysis</strong></li><li><strong>Generate creative, goal-driven feature ideas</strong></li><li><strong>Prototype new ideas quickly</strong></li></ul><h3>What’s Next?</h3><p>Now that you’ve used AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f"><strong>IDEATE</strong></a> new product features, you’re ready to move on to the next step in the USIDO Framework: <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662"><strong>DEFINE</strong></a>.</p><p>In our next edition, we’ll dive into how AI can assist you in defining and prioritizing features, building roadmaps, and even generating user stories.</p><p>See you there!</p><h3>Call to Action:</h3><p><strong>Try It Now</strong>: If you’re ready to start using GPT for product ideation, head over to our <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a>. Just enter your product’s key data and let GPT do the rest! You’ll have a list of feature ideas in minutes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qBYIH1gP-2qDuopkz63Avg.png" /><figcaption><a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box</a></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fd541b703f2f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE Your Product with Custom GPTs</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[STRATEGIZE Your Product with Custom GPTs]]></title>
            <link>https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/1c71ebf1be04</guid>
            <category><![CDATA[ai-product-management]]></category>
            <category><![CDATA[gpt]]></category>
            <category><![CDATA[aiproductpeople]]></category>
            <category><![CDATA[product-manager]]></category>
            <category><![CDATA[ai-product-manager]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Tue, 13 Aug 2024 02:40:22 GMT</pubDate>
            <atom:updated>2024-10-01T20:42:05.059Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w12My24Jjc4v_TyYiRo_RQ.png" /></figure><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><p>Welcome back to another edition of AI for Product People. For those new here, we cover the craft of building AI products.</p><p>In the <a href="https://www.linkedin.com/pulse/understand-app-store-reviews-customgpts-curtis-savage-rqeoc/?trackingId=by9snY2fQJWYan%2FyPsJDmw%3D%3D">last edition</a>, we looked at how a custom GPT can help you better UNDERSTAND your users and product by analyzing App Store reviews. As discussed, the ability of LLMs to recognize patterns and trends in unstructured data sets is a game-changer. LLMs can sift through troves of unstructured data like user reviews and give you a direct lens into how users interact with your app — what they love, what they hate, and what they wish for.</p><p>This week we’re continuing our journey and looking at how you can use a GPT to help build out your product strategy. Using a GPT to STRATEGIZE is the logical next step now that understand user needs.</p><p>This brings us to the second step of <a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=%2FtxBXFJrTfqB6Fndk%2BAO6g%3D%3D">The USIDO Framework for AI Product Managers</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kvIQNppbeTX2ZQNub5OYPQ.png" /></figure><p>There are 5 key steps across the product lifecycle in which a Product Manager can leverage AI to launch a new feature.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5YyJK6VsuOqIYs5hylhJ7A.png" /><figcaption>The USIDO Framework</figcaption></figure><p>This week we’re going deep on GPTs and product strategy. We’re looking at how you can use a GPT to generate three key strategic documents:</p><ol><li>Market &amp; Competitor Analysis</li><li>SWOT Analysis</li><li>Product Strategy</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zmCObvTarnXsPXTWCL7FGA.png" /><figcaption>Step 2: STRATEGIZE</figcaption></figure><p>Together, these three outputs can lay a solid foundation for your product strategy. They can help you make more informed decisions and ensure that you are more confidently guiding your product towards success.</p><h3>Building a Custom GPT</h3><p>Perhaps the thought of putting in time to tailor a custom GPT seems daunting. It’s actually not very difficult and can be done rather quickly once you learn the basics and some nuances. Spend an hour watching some tutorials on youtube and you’ll be well on your way.</p><p>But not to worry, in the meantime, I’ve built one to get you started immediately 😉 The <a href="https://chatgpt.com/g/g-tGFjmSHTJ-ai-assistant-for-product-strategy">AI Assistant for Product Strategy</a> will start by analyzing your user reviews for context. Then it will perform market research before generating the strategic documents mentioned above.</p><p>Simply follow the link above and select the “Start Here” prompt.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CtvHI6PqqkUA00Pz" /><figcaption><a href="https://chatgpt.com/g/g-tGFjmSHTJ-ai-assistant-for-product-strategy">AI Assistant for Product Strategy</a></figcaption></figure><h3>User Feedback Analysis with a GPT</h3><p>The tool will first analyze your user reviews from the Apple App Store using an API I’ve integrated after you enter some basic info found in the App Store URL.</p><p>For example, the Apple App Store URL for the Netflix App in the United States is <a href="https://apps.apple.com/us/app/netflix/id363590051">https://apps.apple.com/us/app/netflix/id363590051</a>.</p><p>I will simply enter “us 363590051” into the prompt to get started.</p><p>The GPT then generates a detailed analysis of user reviews. You can access the example and chat history <a href="https://chatgpt.com/share/dcdaa85b-30f1-44c9-a670-af7c6db87f10">here</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*y49uTUcolDUprHMX" /><figcaption>Step 1: Analysis of App Store Reviews with a GPT</figcaption></figure><h3>1. Market &amp; Competitor Analysis with a GPT</h3><p>Conducting market and competitor analysis helps lay a foundation for your product strategy. It provides you with a clear and comprehensive view of the playing field, identifying key players, emerging trends, and potential market gaps. This type of view helps ensure that product development is informed by a solid understanding of the market landscape and not happening in a vacuum or based on invalid assumptions.</p><p>The <a href="https://chatgpt.com/g/g-tGFjmSHTJ-ai-assistant-for-product-strategy">AI Assistant for Product Strategy</a> generates a detailed Market and Competitor Analysis. You can access the example and full chat history <a href="https://chatgpt.com/share/dcdaa85b-30f1-44c9-a670-af7c6db87f10">here</a> (scroll down to the “Market and Competitor Analysis” section).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1004/0*G4IWWSBMEuX5MwK1" /><figcaption>Step 2: Market &amp; Competitor Analysis with a GPT</figcaption></figure><h3>2. SWOT Analysis with a GPT</h3><p>With a good sense of the market and competitors, you’re now in a good position to review your product’s strengths, weaknesses, opportunities and threats (SWOT). A SWOT analysis can help you understand where to play and where not to play; where to double down based on strengths and opportunities; and where to take a more defensive or mitigative approach based on weaknesses. You can automate and expedite these insights using a custom GPT to search the web perform market research</p><p>The <a href="https://chatgpt.com/g/g-tGFjmSHTJ-ai-assistant-for-product-strategy">AI Assistant for Product Strategy</a> generates a detailed SWOT Analysis. You can access the example and full chat history <a href="https://chatgpt.com/share/dcdaa85b-30f1-44c9-a670-af7c6db87f10">here</a> (scroll down to the “SWOT Analysis” section)..</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7ykaohk2k1Y27d2k" /><figcaption>Step 3: SWOT Analysis with a GPT</figcaption></figure><h3>3. Product Strategy with a Custom GPT</h3><p>With a solid understanding of your market, competitors, strengths, weaknesses, opportunities, and threats you’re now in a great position to generate a Product Strategy</p><p>A high level Product Strategy helps you quickly and efficiently identify your strategic direction by identifying the following:</p><p><strong>Executive Summary:</strong></p><ul><li>A high-level overview of the product strategy, including the product vision, goals, and key success factors.</li></ul><p><strong>Market Analysis:</strong></p><ul><li>Industry Overview: Insights into the industry landscape, trends, and growth opportunities.</li><li>Competitive Analysis: Evaluation of competitors, their strengths, weaknesses, and market positioning.</li></ul><p><strong>Customer Analysis:</strong></p><ul><li>Target Audience: Detailed description of the target customers, including demographics, psychographics, and behavior patterns.</li><li>Customer Needs: Identification of customer pain points, needs, and preferences.</li></ul><p><strong>Product Vision and Goals:</strong></p><ul><li>Vision Statement: A clear and inspiring vision for the product.</li><li>Goals and Objectives: Specific, measurable, achievable, relevant, and time-bound (SMART) goals for the product.</li></ul><p><strong>Value Proposition:</strong></p><ul><li>Unique Selling Points (USPs): Key differentiators that make the product stand out in the market.</li><li>Value Proposition Canvas: Mapping of product features to customer needs and pain points.</li></ul><p><strong>Business Model:</strong></p><ul><li>Revenue Streams: Identification of primary and secondary sources of revenue.</li></ul><p><strong>Metrics and KPIs:</strong></p><ul><li>Success Metrics: Key performance indicators (KPIs) to measure the success of the product.</li></ul><p><strong>Summary:</strong></p><ul><li>Recap of the key points and a call to action for stakeholders.</li></ul><p>A custom GPT can be a game-changer here. You can quickly generate a good first-draft. It won’t be perfect but will get you started and can help distill complex market analysis and strategic insights into a clear, concise framework, saving you time while providing clarity. It may even provide some insights that you hadn’t yet considered. Not bad a for a few more minutes of your time.</p><p>The <a href="https://chatgpt.com/g/g-tGFjmSHTJ-ai-assistant-for-product-strategy">AI Assistant for Product Strategy</a> generates a high level Product Strategy. You can access the example and full chat history <a href="https://chatgpt.com/share/dcdaa85b-30f1-44c9-a670-af7c6db87f10">here</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Skf1v1IQfhb1s8ER" /><figcaption>Step 4: Product Strategy with a GPT</figcaption></figure><h3>Final Thoughts</h3><p>As we wrap up this edition, I hope you can see how a custom GPT can elevate your product strategy. Whether it’s analyzing user feedback, conducting market research, identifying strengths, weaknesses, opportunities, and threats, or even generating an initial draft of your product strategy, a GPT can save you time while delivering valuable insights to guide product development.</p><p>If you’ve been following the blog series, you’ve now completed the first two steps of the ‘<a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=%2FtxBXFJrTfqB6Fndk%2BAO6g%3D%3D">USIDO Framework for AI Product Managers</a>’ using a custom GPT:</p><ol><li>You’ve used AI to <a href="https://www.linkedin.com/pulse/understand-app-store-reviews-customgpts-curtis-savage-rqeoc/"><strong>UNDERSTAND</strong></a> your users.</li><li>You’ve used AI to <a href="http://strategize/"><strong>STRATEGIZE</strong></a> your product approach.</li></ol><p>You are now ready for the third step, young Jedi.</p><p>Join us in the next edition as we teach you how to use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f"><strong>IDEATE</strong></a> product features based the user needs and strategic insights you generated from the two steps above.</p><p>See you then!</p><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1c71ebf1be04" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE Your Product with Custom GPTs</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[UNDERSTAND Your Product with Custom GPTs]]></title>
            <link>https://medium.com/swlh/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/66ce9e3eb3f6</guid>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[customgpt]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[app-store-review]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Wed, 26 Jun 2024 23:06:55 GMT</pubDate>
            <atom:updated>2024-10-01T20:41:51.480Z</atom:updated>
            <content:encoded><![CDATA[<h3>UNDERSTAND Your Product with CustomGPTs</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pykUD4CDbqb-I1V2myrHZA.png" /></figure><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><h3>Introduction</h3><p>For those new here, this blog series aims to demystify AI for Product People, helping you foster a deeper understanding and the confidence needed to integrate AI into your product management practices. With all the hype and hyperbole surrounding AI, this blog series aims to help you stay grounded by focusing on the craft, the nuts and bolts of building great AI products.</p><p>Every week we show up to help you answer two fundamental questions:</p><ol><li>How can AI make me a better Product Manager in my day-to-day job?</li><li>How can I build AI features and capabilities into my product?</li></ol><p>Let’s get started.</p><h3>A Framework for AI Product Managers</h3><p>In the <a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=UFdSp2vLTQCmS%2F1k9WPN3w%3D%3D">last edition</a> we covered a framework to help Product Managers leverage AI across five key steps of the product lifecycle. This framework can help you better 1) Understand, 2) Strategize, 3) Ideate, 4) Define and 5) Optimize your Product.</p><p>I call this <a href="https://www.linkedin.com/pulse/5-ways-ai-can-make-you-better-product-manager-curtis-savage-y1hdc/?trackingId=UFdSp2vLTQCmS%2F1k9WPN3w%3D%3D">The USIDO Framework for AI Product Managers</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*acgud5RTnCAveZizly_shQ.png" /></figure><p>This week we’re going deep on the first step. We’re looking at how you can use a CustomGPT to drastically speed up the process of understanding App Store Reviews for valuable insights.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OGd0FeS9yFP3a3A-So0cdQ.png" /><figcaption>Step 1: UNDERSTAND</figcaption></figure><p>The ability of LLMs to recognize patterns and trends in unstructured data sets is a game-changer. LLMs can sift through troves of unstructured data like user reviews and spot trends and themes around customer pain points that may have taken you hours or days to understand.</p><h3>The Challenge with App Store Reviews</h3><p>Product Managers face a significant challenge with App Store Reviews. With thousands of reviews posted across different platforms, sifting through each one to identify trends, pain points, and user sentiments is not just time-consuming but often impractical.</p><p>This mass of data, while valuable, presents a daunting task. Manually analyzing this feedback takes a ton of time. As a result many product teams find themselves chasing one or two of the most inflammatory reviews as they pop up; putting out fires but having no idea just how prevalent and wide spread various issues are.</p><h3>The Shortcomings of ChatGPT for Reviews</h3><p>While ChatGPT can offer assistance, it also has its shortcomings when it comes to analyzing reviews. First, you need to find a way to import the reviews into chatGPT. The recent ability to attach excel documents in chatGPT has made this easier. However, this still comes with its challenges. You need to source a CSV file from somewhere. If you don’t have access to a CSV file you will need to subscribe to an additional service such as appBot to generate CSV downloads of all your app store reviews — an added step and one more monthly subscription to pay for.</p><p>Next, you need to know what to ask chatGPT to get the results you’re looking for. Getting to valuable insights can often be tedious and time-consuming. And if you want to analyze your app again in future or perhaps analyze a competitor’s app, then you need to try and remember how you previously prompted chatGPT to get the kinds of insights you were looking for.</p><p>And what if other team members or a VP wants to do some quick analysis? Now you have to share your prompts with them. Again, not ideal. This is all rather tedious, cumbersome and time-consuming. Likely enough friction that you may not put in all the effort. And without the effort, the insights chatGPT produces will likely be mediocre and rather trivial.</p><p>Well, there’s good news. There’s a much better and easier way to go about this.</p><h3>The Solution: CustomGPTs</h3><p>What are Custom GPTs?</p><p>You can now create custom versions of ChatGPT that combine instructions, extra knowledge, and any combination of skills. GPTs are a new way for anyone to create a tailored version of ChatGPT to be more helpful at specific tasks — and then share that creation with others.</p><p>From OpenAI:</p><blockquote><em>Anyone can easily build their own GPT — no coding is required. You can make them for yourself, just for your company’s internal use, or for everyone. Creating one is as easy as starting a conversation, giving it instructions and extra knowledge, and picking what it can do, like searching the web, making images or analyzing data.</em></blockquote><h3>A Game-Changer for PMs</h3><p>CustomGPTs can help solve many of the challenges described above. You build it once and can share it with anyone on your team. As you can imagine, these CustomGPTs can be leveraged across the product lifecycle.</p><p>For example, you could build a customGPT to help you:</p><ul><li>UNDERSTAND your product by analyzing user feedback</li><li>STRATEGIZE by generating a SWOT analysis</li><li>IDEATE by generating feature ideas based on user pain points</li><li>DEFINE your product by writing first drafts of user stories</li><li>OPTIMIZE your product by generating ideas around personalization</li></ul><h3>A CustomGPT for App Store Reviews</h3><p>Now, I have even better news…</p><p>To get you started, I’ve already built one for you! 🥳</p><p>Let’s try it out. Start by following this <a href="https://chat.openai.com/g/g-zxTyUSZAF-ai-for-product-people">link</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GQTC1FlFv7fp-KME" /></figure><p>App Store Review Analyzer</p><p>For the demo, I will analyze the <a href="https://apps.apple.com/us/app/spotify-music-and-podcasts/id324684580">Spotify App</a> listed in the US app store.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*JN1QICK27lvGH3-Y" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xiURW7sYEGeiO6YY" /></figure><p>To use the tool follow the steps below:</p><p>👉 From the GPT landing screen, simply select the prebuilt prompt: “App Store Review Analyzer”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZL0EviseYBs2MODw" /></figure><p>👉 When prompted, simply provide the Country Code and App ID:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*8YtvPF7oY7yRq1ye" /></figure><p>👉 Tap “Confirm” to allow the CustomGPT to talk to the Apple App Store via an API.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-tbSkNQOlGG45wNd" /></figure><p>Et voila! You immediately get <a href="https://chat.openai.com/share/2267a3ea-cefc-48ab-be35-df002b1a278e">an analysis of the Spotify app</a>.</p><p>We begin with some General Information…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7f5proHHMwx-cTHe" /></figure><p>Followed by an Executive Summary to give you a quick gut check of where users are at…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*dLb5ADyMkB5-xaSR" /></figure><p>From here, it provides a quick Sentiment Analysis…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*tifClamRVnUaCOBO" /></figure><p>And explores Main Themes with real-world examples of comments from users…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*--cyTqr8ijPA9ttH" /></figure><p>It lists the the Top 10 Issues users are complaining about…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CvrDso47dEHm_Dao" /></figure><p>And quickly identifies key Trends…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9XP8cpl3Znu9l1tj" /></figure><p>Before offering some high level Recommendations for consideration…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Nb1TEOqFT7_QnBtt" /></figure><h3>A Very Strong Starting Point…</h3><p>Of course, this is just a starting point. But it is a very solid one. One that can quickly point you in the right direction for further exploration. Not bad, to say it was generated in under one minute!</p><p>From this initial analysis, we already know that premium subscribers appear to be relatively happy while the free tier users are much more dissatisfied given heavy advertising and limitations of the free tier. We can now hone in on this (or any other area of interest)…</p><h3>Going Even Further…</h3><p>There are many different directions we can take the follow-up questions based on our specific goals as a Product Manager. Here are just a few of the follow-ups I would be eager to ask:</p><ol><li>What are the top pain points for freemium users?</li><li>How can we reduce churn and drive more conversions to Premium?</li><li>What are the top issues caused by the recent updates?</li><li>Can you provide a more in depth analysis of the Top Issues and provide suggestions on how to address them?</li></ol><p>For example, probing a littler deeper on the issue of excessive ads we can quickly generate a few useful suggestions and potential feature enhancements:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VAudopVB5KIU_rQr" /></figure><p>Again, perhaps not perfect but not bad for a few more minutes of your time!</p><h3>Other Use Cases</h3><p>The versatility of CustomGPTs extends far beyond just analyzing your own app reviews and can be applied to other use cases.</p><p>For example:</p><ul><li>Analyzing Competitor Apps: Dive into competitor app reviews, unveiling user feedback on their features and services. This insight can inform how to refine your product to outpace competitors.</li><li>Understanding a Market: Aggregate insights across multiple apps in your category to spot trends and user expectations, helping you identify gaps your app could fill.</li><li>Preparing for a New Assignment: Analyze the app and market segment, gathering insights on user feedback and industry trends to hit the ground running with informed, actionable strategies from day one.</li></ul><h3>Final Thoughts</h3><p>App Store reviews are a goldmine of insights. They reflect the unfiltered voice of the customer. The ability to rapidly UNDERSTAND this feedback is invaluable. CustomGPTs enable you to have your finger on the pulse of user pain points, preferences, and suggestions, providing a solid foundation of recommendations in just minutes.</p><p>This approach not only accelerates understanding and insights, but can also cut through internal biases and preconceptions. It allows for an unbiased, comprehensive view of how users interact with the app — what they love, what frustrates them, and what they wish for.</p><p>By leveraging CustomGPTs to analyze app store reviews, you’re equipped to make informed decisions, prioritize product improvements, and strategize new features with confidence. This method ensures that product development is aligned with user needs and market demands, fostering a user-centric approach from the outset.</p><p>Armed with a solid UNDERSTANDING of your users, you’re now equipped to leverage AI in the next step of the <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">USIDO Framework</a>. Join us in the next edition as we look at how CustomGPTs can help you <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a>!</p><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=66ce9e3eb3f6" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND Your Product with Custom GPTs</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The U.S.I.D.O Framework for AI Product Managers]]></title>
            <link>https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de?source=rss-bb62b1bb6ba2------2</link>
            <guid isPermaLink="false">https://medium.com/p/df72f8a584de</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[product-manager]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[aiforproductpeople]]></category>
            <dc:creator><![CDATA[Curtis Savage]]></dc:creator>
            <pubDate>Sat, 02 Mar 2024 13:14:36 GMT</pubDate>
            <atom:updated>2025-04-11T00:06:43.343Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zBZCDUDttxnD2znSjy__rw.png" /></figure><h3>The U.S.I.D.O. Framework for AI Product Managers</h3><blockquote>TLDR: AI can make you a better Product Manager by helping you Understand, Strategize, Ideate, Define and Optimize your product (I call this the U.S.I.D.O. Framework for AI Product Managers).</blockquote><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><h3>Introduction</h3><p>There are <a href="https://medium.com/ai-for-product-people/the-4-questions-product-people-should-ask-to-build-great-ai-products-0d6b8e9d8bc5"><strong>four questions</strong></a><strong> </strong>every Product Manager should ask to build great AI Products.</p><p>They are:</p><ol><li>How can AI make me better in my role?</li><li>What are the core use cases for my product?</li><li>What are the best practices for working with teams to build AI features?</li><li>How can AI accelerate value for my users and growth for my company?</li></ol><p>In this blog series, we will exploring the first question:</p><blockquote>How can AI make me better in my role?</blockquote><p>To start, it’s important to note that AI won’t be replacing Product Managers now or in the future.</p><h4>Product Managers are here to stay. Here’s why…</h4><p>Being customer-centric, having keen business sense, translating user needs into compelling features, collaborating with teams and managing stakeholders remain dynamic, indispensable skills. AI, in this context, is not a replacement but a powerful ally that can amplify your efforts so that you can continue to slay dragons and sling products. Think of AI as that super-smart intern who underpromises and overdelivers while you take all the credit…but, of course, you don’t do that, right?</p><p>Coffee, anyone?</p><h4><strong>But you still need to put in the work.</strong></h4><p>Okay, so AI won’t be replacing you now or in the future, but it is a tool you need to learn to<strong> </strong>help you augment, automate and speed up certain tasks. In this way it can help you build better digital experiences.</p><h3>The U.S.I.D.O Framework</h3><p>To structure our approach, it’s helpful to think about how to leverage AI across 5 key steps of the product lifecycle which I call the <strong>U.S.I.D.O Framework for AI Product Managers </strong>because, yes, I’m obsessed with product frameworks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DHWB2-gSemgajhyc.png" /></figure><p>The U.S.I.D.O. Framework is a powerful mental model for Product Managers. To boil it down, there are 5 key steps across the product lifecycle in which Product Managers can leverage AI to launch new features.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*pq_nJHo00QTFqKIM.png" /></figure><p>AI can make you a better Product Manager by helping you Understand, Strategize, Ideate, Define and Optimize your product.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5YyJK6VsuOqIYs5hylhJ7A.png" /><figcaption>The USIDO Framework for AI Product Managers</figcaption></figure><h3>Step 1: UNDERSTAND</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OGd0FeS9yFP3a3A-So0cdQ.png" /><figcaption>Step 1: UNDERSTAND</figcaption></figure><p>Not to pile on to all the hype and hyperbole, but it is undeniable that AI’s ability to recognize patterns and trends in data sets is a game-changer. It can sift through troves of both structured data (think product analytics) and unstructured data (think user reviews). It can spot trends and themes around customer pain points that may have taken you hours or days. And perhaps, even more importantly, it can spot patterns and generate insights that may have gone uncovered without it.</p><p>Rather than dedicating hours to combing through vast datasets, product managers can utilize AI to sift through and condense this information, allowing them to go deeper on specific themes that warrant closer inspection.</p><p>This can then power decision making and become a very powerful tool for things like product discovery, roadmap planning, and growth strategies.</p><blockquote><strong>Action Item #1:</strong> Use a Custom GPT like <a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a> to analyze user reviews and understand what users love, hate, and wish for in your app.</blockquote><h3>Step 2: STRATEGIZE</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*qqX7teX9UOq2tAvA.png" /><figcaption>Step 2: STRATEGIZE</figcaption></figure><p>The next area where AI can help Product People is at the strategy stage.</p><p>This can be for a new or existing product or feature. It can help you understand the market and competition.</p><p>As an example, you can feed the LLM information about your company/product/users, etc. so it has context. With this context, it can help you identify your strengths, weaknesses, opportunities and threats (SWOT) and generate a detailed SWOT analysis to guide additional decisions across the product lifecycle like where to play, how to position yourself against competitors, and what opportunities to follow.</p><p>Strengths, weaknesses, and opportunities…oh my.</p><p>AI can help you perform market/competitor analysis and generate strategic artefacts and insights to then guide product development efforts in addition to things like positioning, differentiation, and go-to-market strategies.</p><blockquote>The key takeaway is to use AI to help you generate strategic insights to then guide product development — saving you hours or even days of work.</blockquote><p>This gives you faster initial confidence that <strong>the ship is pointed in the right direction</strong>.</p><blockquote><strong>Action Item #2: </strong>Use a custom GPT like <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a><strong> </strong>to perform market/competitor research, generate a SWOT analysis, and output a a product strategy based on its findings.</blockquote><p>It won’t be perfect but it will get you started, accelerate the process, and provide a very good starting point. Much better than a blank page. That blank daunting page. Oh the abyss… I digress.</p><h3>Step 3: IDEATE</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L9IjmpnE07nmm4ZFMtAJ1g.png" /><figcaption>Step 3: IDEATE</figcaption></figure><p>This may be my favourite step because we transition from understanding and analyzing to creating.</p><p>If you’ve followed the <strong>Action Items</strong> above, then you’ve trained chatGPT on your users, competitors, and market. It understands your user’s biggest pain points by analyzing your App Store reviews and it understands your biggest opportunities by performing a SWOT analysis against your competitors in the market. With this understanding it’s well-primed for some good ol’ LLM magic.</p><p><strong>Template</strong>:</p><blockquote>“chatGPT, based on {insert analysis}, what product features should I improve or build to {insert goal} with {insert user-type}”</blockquote><blockquote><strong>Action Item #3: </strong>Using a custom GPT like <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a>, ask it the following: “Based on <strong>{the top five pain points you identified from user reviews and the top five opportunities from SWOT analysis}</strong>, what product features should I improve or build to <strong>{drive retention}</strong> with <strong>{new users}.”</strong></blockquote><p>This is just a boiler plate, but you can insert any goal/user-type of your choice and customize the prompt to focus on Adoption, Activation, Engagement, Retention, Revenue, etc…</p><h3>Step 4: DEFINE</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*0J4kwDR9wJbXC7_l.png" /><figcaption>Step 4: DEFINE</figcaption></figure><p>With a list of product features and enhancements that focus on your biggest pain points and opportunities you can now start prioritizing and defining a roadmap. And guess what, AI can help...</p><h4><strong>Roadmap</strong></h4><p>AI can help you prioritize which features to build based on their potential impact on product success. It can help you define your product roadmap by analyzing historical data and predicting the impact of specific features on strategic areas and KPIs like retention, user satisfaction and revenue.</p><blockquote><strong>Action Item #4: </strong>Use a custom GPT like <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a><strong>. </strong>Prompt it to generate a roadmap using all of the output we generated from the previous action items as context:</blockquote><blockquote>“Based on all of the information provided and generated from previous steps: What are the immediate features/enhancements I should focus on? Can you organize the list of features above into a one year roadmap divided into quarters of the year?”</blockquote><h4><strong>Personas, User Stories, and Release Notes</strong></h4><p>With its ability to analyze large amounts of data from multiple data sources, there is a world where AI can generate user stories and personas based on things like product usage, user surveys and more.</p><blockquote><strong>Action Item #5: </strong>Use a custom GPT like <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>AI Product Manager in a Box</strong></a><strong>. </strong>Prompt it to generate a generate user stories based on all of the context we’ve provided in the chat up until this point.</blockquote><blockquote><em>“Can you generate user stories in the Gherkin format for the next feature to be launched in roadmap?”</em></blockquote><p>This last prompt, I find particularly useful as it can save a huge amount of time.</p><p>It’s important to note, all of these steps require an <strong><em>AI-augmented </em></strong>approach. Meaning there needs to be a human in the loop (that’s you!). The outputs won’t be perfect. You will need to edit, refine, re-prompt, iterate and guide the LLM. And that’s a good thing. That’s why the AI overlords still need you and why you still have a job!</p><h3>Step 5: OPTIMIZE</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DA9Qv14IV4lB4Xtbf-fxkw.png" /><figcaption>Step 5: OPTIMIZE</figcaption></figure><p>Now that you’ve used AI to understand, strategize, ideate, and define your product, it’s time to focus on <strong>optimizing</strong>. The optimization phase is about ensuring your product’s ongoing success by refining your go-to-market (GTM) strategy, identifying key metrics for success, and continuously running experiments to maximize impact.</p><h4>1. Optimizing Your Go-to-Market Strategy</h4><p>A successful product launch is only as strong as its go-to-market strategy. This is where you position your product to reach the right audience, using the right channels, with the right messaging. The GPT can be your secret weapon here, helping you analyze market conditions and generate a comprehensive, data-backed GTM plan.</p><p><strong>Example</strong>: Let’s say you’re about to launch a new feature aimed at improving user retention. The GPT can suggest the best channels (email, social media, paid ads), craft messaging that resonates with your target audience, and even recommend positioning that sets your feature apart from competitors.</p><h4>2. Identifying Success Metrics and KPIs</h4><p>With your GTM strategy in place, the next step is defining success. Which metrics will tell you whether your product is truly making an impact? Whether it’s user retention, activation, or revenue growth, selecting the right KPIs is essential for measuring performance.</p><p>GPT can help you identify and prioritize KPIs that align with your specific goals. For example, if your goal is to boost engagement, GPT might recommend tracking metrics like <strong>daily active users (DAUs)</strong>, <strong>session length</strong>, or <strong>feature adoption rate</strong>.</p><h4>3. A/B Testing and Experimentation</h4><p>Optimization doesn’t end with tracking KPIs. To unlock your product’s full potential, you need to experiment. A/B testing allows you to test different versions of features, user flows, or messaging to understand what works best. The beauty of using GPT is that it can generate A/B test ideas in seconds, allowing you to move quickly from hypothesis to action.</p><p>For example, you could test two different onboarding flows for new users to see which one drives higher activation rates. GPT can help you define test parameters, select user segments, and even suggest the key metrics to track for each variation. This data-driven experimentation allows you to iterate faster and refine your product based on real user behavior.</p><h3>Summary</h3><p>Product Managers aren’t going away.</p><p>Our roles are more crucial than ever: being customer-centric, having keen business sense, translating user needs into compelling features, collaborating with teams and managing stakeholders remain dynamic indispensable skills.</p><p>A world devoid of Product Managers results in ‘capability-driven’ rather than ‘user-driven’ products that nobody wants.</p><p>However, the game has changed. AI is a powerful tool that all Product Managers must learn to leverage to be more effective in their roles. Or risk falling behind.</p><p>It’s helpful to think about leveraging AI across five key steps of the Product Lifecycle: Understanding, Strategizing, Ideating, Defining and Optimizing. Doing so can help you build better digital experiences while unlocking product value for both your users and your business.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DHWB2-gSemgajhyc.png" /><figcaption>The USIDO Framework</figcaption></figure><h3>What’s Next?</h3><p>First, get your hands dirty by trying out the custom GPTs we have built for you:</p><ol><li><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer"><strong>App Store Review Analyzer</strong></a><strong> </strong>to analyze App Store Reviews</li></ol><p>2. <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box"><strong>Product Manager in a Box</strong></a><strong> </strong>to walk you across the entire product lifecycle</p><p>After that, you’re ready to dive deeper into the first step of the USIDO Framework: <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6"><strong>UNDERSTAND</strong></a>. In our next post, we’ll explore how AI can assist you in understanding user needs with practical examples and tools to get you started!</p><p>See you there!</p><h3>Table of Contents</h3><ul><li><strong>Introduction: </strong>The <a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">U.S.I.D.O.</a> Framework</li><li><strong>Step 1: </strong>Use AI to <a href="https://medium.com/@curtis.savage/understand-app-store-reviews-with-custom-gpts-66ce9e3eb3f6">UNDERSTAND</a> your product</li><li><strong>Step 2: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/strategize-your-product-approach-with-custom-gpts-1c71ebf1be04">STRATEGIZE</a> your product</li><li><strong>Step 3: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/product-ideation-with-a-custom-gpt-fd541b703f2f">IDEATE</a> your product</li><li><strong>Step 4: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/define-your-product-with-custom-gpts-56f4eaa60662">DEFINE</a> your product</li><li><strong>Step 5: </strong>Use AI to <a href="https://medium.com/ai-for-product-people/optimize-your-product-with-custom-gpts-65d4476e7a0b">OPTIMIZE</a> your product</li><li><strong>Custom Tool: </strong><a href="https://chatgpt.com/g/g-zxTyUSZAF-app-store-review-analyzer">App Store Review Analyzer</a></li><li><strong>Custom Tool:</strong> <a href="https://chatgpt.com/g/g-it99ZN2Jw-ai-product-manager-in-a-box">AI Product Manager in a Box</a></li><li><strong>Build Your Own</strong>: <a href="https://www.linkedin.com/pulse/build-your-own-ai-product-manager-box-curtis-savage-5i4sc/">AI Product Manager in a Box</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=df72f8a584de" width="1" height="1" alt=""><hr><p><a href="https://medium.com/ai-for-product-people/a-framework-for-ai-product-management-df72f8a584de">The U.S.I.D.O Framework for AI Product Managers</a> was originally published in <a href="https://medium.com/ai-for-product-people">AI for Product People</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>