10 AI Tools Every Product Manager Should Use

I tried far too many over the past year. These are the ten that actually stayed open on a boring Tuesday, and what each one is genuinely for.

Last November, I opened my password manager and counted. Fourteen AI tools. Fourteen logins, fourteen subscriptions, fourteen tabs, I had convinced myself were essential. Most of them I had signed up for after reading some breathless “tools every PM needs” post, demoed once to look clever in a team meeting, and then never opened again.

So over the following months, I did the boring work of actually paying attention. Which tools did I reach for on a normal Tuesday, when nobody was watching, and there was no demo to perform? Which ones quietly earned their place, and which ones were just expensive guilt sitting on a direct debit?

What follows is the honest shortlist. Ten AI tools that I think genuinely make a product manager’s life easier, grouped loosely by the job they do. You will not use all ten at once, and you should not try to. But if you are building an AI stack from scratch, this is the map I wish someone had handed me eighteen months ago.

Writing and documentation

This is where most of us feel the time drain first. PRDs, release notes, user stories, the endless translation of a technical decision into language an executive will actually read. These three tools attack that pile.

1. Claude

Claude has become my default thinking partner for anything that involves words. Drafting a PRD from a messy set of notes, turning a half-formed idea into a structured one-pager, condensing a long research thread into something a stakeholder will read in ninety seconds. It is fast, the writing quality is genuinely good, and it handles long, messy context without falling over.

The trick is to treat it as a first-draft engine, not a final-draft oracle. I write the bones, it fleshes them out, I edit. What used to be a two-hour staring contest with a blank document is now closer to thirty minutes.

2. Claude Code

This one surprised me, because I had filed “AI coding tool” under “not for me, I am a PM.” I was wrong. Claude Code runs in your terminal or the desktop app, and it works directly on your files. It reads folders, writes documents, and pulls data from tools like Linear or GitHub through MCP connectors, chaining those steps together without you babysitting each one.

The example that converted me: ten user-interview transcripts that I would normally block out half a Monday to synthesise. I pointed Claude Code at the folder, gave it my template, and had a structured first draft in a few minutes. It also drafts my weekly sprint report now by pulling the completed tickets. The catch is real, though. If your underlying data is a mess, the output is a confidently-worded mess. Tidy your workspace first.

3. Notion AI

If your team already lives in Notion, this is the lowest-friction win on the list. The AI sits directly inside your workspace, so you can draft a spec, generate user stories, and organise research without switching contexts or copying text between tabs. It is not the most powerful model on this list, but the value is in the absence of friction. The work happens where the work already lives.

Research and discovery

The quality of every downstream decision, what you prioritise and what you ship, is inherited from the quality of the evidence you gathered up front. Shallow discovery means your roadmap is just a list of opinions. These tools deepen the evidence.

4. Perplexity

I resisted this one because “AI search engine” sounded like a worse Google. It is not. The difference is citations. Perplexity gives you a sourced answer with the links right there, rather than a confident paragraph you then have to fact-check from scratch.

That distinction matters the moment you are putting research in front of an executive. A claim with a live source attached is something you can defend in the room. I use it for competitive analysis, market sizing sanity checks, and the regulatory questions I need to get right before committing a roadmap line. The Pro Search mode does multi-step research, following up on its own findings, which is useful for the deeper “help me understand this whole space” questions.

5. NotebookLM

Where Perplexity searches the open web, NotebookLM grounds itself entirely in your own documents. You upload your interview transcripts, your support tickets, your survey responses, and you ask questions against only that material. Because it is grounded in your sources, the hallucination risk is much lower, and it cites which document each answer came from.

It has become my go-to for synthesising customer interviews. Twenty transcripts that used to take the best part of a day now give up their themes in an afternoon, and I can interrogate them like a colleague who has read everything.

6. Dovetail

For teams that run continuous user research at any real volume, a dedicated research repository is worth it. Dovetail stores your interviews, tags themes across studies, and surfaces patterns you would never spot by reading transcripts one at a time. The AI layer clusters feedback and highlights recurring signals automatically. If your research currently lives in a graveyard of scattered Google Docs, this is the tool that turns it into an asset you can actually mine.

Prioritisation and roadmapping

Once you have the evidence, you have to decide what to build and convince everyone else of the plan. This is the political heart of the job, and AI cannot make the call for you. But it can do the grunt work around it.

7. Productboard

Productboard has been a go-to product platform for years, and the AI additions have made it sharper. It consolidates feedback from support tickets, sales calls, and user interviews into a single view, then helps you score features and detect trends. The strength here is not really the AI, it is the workflow. The AI just makes prioritisation faster by surfacing patterns you would otherwise miss. Best suited to mid-size and enterprise teams with heavy customer-feedback flows.

8. Linear

Linear is project management with AI woven into the core, rather than bolted on. It auto-triages incoming issues, detects duplicates, suggests priority levels, and can generate sub-issues from a high-level task. The AI triage alone is the feature people rave about, because it quietly removes a chunk of the backlog admin that used to eat an hour of your week. If your team is small or moving fast, it is a genuinely modern alternative to the heavier incumbents.

Analytics

Interviews tell you why. Analytics tell you what and how many. The shift here is that you no longer need to wait on a data analyst for routine questions.

9. Amplitude

The leading product-analytics platforms have made a quiet but enormous change: you can now ask questions in plain English and get a funnel, a retention curve, or a cohort back instantly. Amplitude leans towards behavioural cohorting and experimentation, and the AI layer removes the analyst bottleneck for the simple lookups that make up most of your day-to-day questions. The trap to avoid is treating analytics as a substitute for talking to customers. It is the quantitative complement to discovery, not a replacement for it. (Mixpanel and PostHog are strong alternatives here, depending on your stack.)

Prototyping

This is the newest and, honestly, the most exciting shift. PMs increasingly wear two hats: strategist and builder. AI has made the second hat far more accessible.

10. v0 by Vercel

v0 lets you describe a UI in plain English and get a working prototype back. No waiting on engineering capacity, no fighting with Figma for an interactive flow. Pick one idea sitting in your backlog because engineering is stretched, build a quick prototype, and test the actual flow with users this week. The PMs who gain the most leverage in the coming years will be the ones who can move from insight to a working artifact faster than everyone else, and this is the category that makes that possible.

How to actually use this list

Here is the part that the “tools every PM needs” posts always skip. You do not adopt ten tools. You adopt one excellent tool per recurring job, and you add nothing else until you feel a specific, repeating pain that none of your existing tools solve.

Start with the writing pile, because that is where almost every PM bleeds the most time. Add a research tool the first time you get caught out presenting a claim you cannot source. Add a prototyping tool the first time engineering capacity blocks an idea you are itching to test. Let each tool earn its way in.

The truth is that my own core is closer to three or four of these on any given week. The rest I reach for when a specific job demands them. That is the point. This is a menu, not a shopping list. The goal was never to have the most tools. It was to have the right one open at the moment you need it, and nothing else cluttering your tabs.

So which job hurts the most in your week right now? Start there, pick the one tool that kills that pain, and leave the other nine for when you actually need them. I would genuinely love to hear in the comments which one earns a permanent place in your stack, and which one you tried and quietly abandoned.

If this resonated with you, or even just made you pause and think, I’d really appreciate a clap or two. It genuinely helps the article reach other product managers who’d find it useful.

And if you’re not already following along, I write weekly-ish about product management, AI tooling, and what’s actually working (and what isn’t) in my day-to-day as a PM in London. Hit follow and you’ll catch the next one.

If you’ve got value from my writing and fancy supporting the work directly, you can buy me a coffee here:

👉 buymeacoffee.com/mohit15856

Either way, thanks for reading.