AI tools for Product Teams (part 3)
Focus: the art of not getting distracted by the shiny objects and doing part of convincing.
This post is the third one of a serie of publications, where I explore, new or existing AI tools for the different phases of the product development lifecycle..
Today we are digging into the Focus phase, which describes how we use vision, strategy, and target outcomes to prioritise what we work on. If you want to go one step before check part 1
and part 2.
Much of what we do in this phase, is a combination of understanding, forming an opinion, and try to convert that in to something that will drive the effort of the team in the upcoming weeks, months, and/or quarters.
Although we can use the chat-based interface tools to pull knowledge about a specific topic, then the hard work comes when we need to boil down all the data, insights, and into a well articulated piece that, must be aligned with the business priorities, it must be converted into objectives that you want to achieve, and show how it is supposed to play out throughout time, aka “put it on a roadmap".
Some of the tools we analyse in this article are focused on one specific area, and others cover the entire universe of possibilities. Let’s dive in!!!
ChatPRD
This is a custom GPT created by one of my favourite product people out there,
. She is the creator and active weekend contributor into the codebase of the tool.ChatPRD is a PM copilot that can think faster than you can type. Let ChatPRD write your docs and get back to building.
Here are the features you can use after being onboarded as part of the free version:
There is a section to leverage default templates and the option to create your own. I think it would have been useful to know the basic structure/sections of each of them to know which one to use.
Things I really loved after interacting with the bot using the Help me write a PRD:
It is going to ask for more information to build context around the answer. This is great, it feels like the bot cares about giving you the most accurate answer.
As PMs we know how relevant narrative is. Not just the tactics and options of building the thing, and how it is supposed to work, but a more visionary thing, that could be used when telling a story about why and what for the thing you are building is needed.
I love the answer includes the Technical Considerations, that normally are overlooked and can quickly give you some aspects when interacting with your engineering counterpart.
I know that Claire thought about it, but how you can integrate this with many other tools to streamline PMs workflow such as, creating a roadmap, OKRs, Notion, and many others.
wrote a great article comparing several options available to write a PRD (Product Requirements Document), and eventually ChatPRD was the fastest and most accurate of all.
Reforge Browser Extension
This is an extension that can be install on the most famous browsers, and it leverages all the knowledge available from knowledge trusted experts and people submitting content into Reforge Artefacts, to write product requirements, GTM plans, user personas, and more.
When you install the extension, it asks you open a document in GDocs, Coda, Notion, Linear, or Confluence. When you open it, you can trigger the extension and this is what you will see:
When I opened the off-boarding document in Spanish, I asked it to improve this document, and it suggested an interview template, a research plan, and a list of features. All of them forwards you to the Artefact when you click on it.
The feature I liked a lot was “create a draft”, where they leverage the frameworks that are taught in the courses, such as the Use Case Map, PRDs, and more. I’ve tried the PRD draft to compare it with ChatPRD. I have to say that ChatPRD is more robust in terms of the results provided and the same number of prompts and interactions.
A plus should be given to the feedback message that arrived once I signed up. This is a great way to care about the privacy and being outspoken that they don’t know the output of the extension. May be they could add the buttons such as like, dislike, or share as part of the feedback.
ProductBoard
ProductBoard has been one of the tools highlighted for good by many colleagues who have used it. Especially when it comes to capture feedback from many different sources. Something is not longer unique to them (as we have seen with other tools such as Enterpret).
Nonetheless, I think the way they have integrated their AI capabilities, under the name ProductBoard AI 2.0, has a lot of advantages for PMs. Let’s dive into them.
Insights Auto-Linking: Productboard AI can automatically link feedback from the insights board to feature ideas in the product hierarchy. This saves time and ensures all feedback is considered in prioritisation decisions.
From each feedback note, you’ll see all the features it’s been linked to. And from every feature, you’ll see which insights have been linked by AI. If you find an insight that’s not so relevant to the feature at hand, you can easily unlink it, or link it to a different feature.
With the ability to combine insights from multiple source, you can automatically factor into scores you can use to prioritize features by their impact on your customers and business.
AI-Generated Feature Specs: linked insights are also immediately incorporated into any AI summaries or “Ask AI” responses, which will save you a lot of time when you’re writing your next feature brief. It extracts key insights like problem statements, pain points, and common themes, allowing product managers to focus on high-value work
.
With more insights linked to features, you’ll also have a more accurate picture of exactly who needs what and it also helps close the loop with all the right users when you move forward with a feature idea or set it live.
Chisel
Chisel is an Agile Product Management software, brings together roadmapping, team alignment, and customer connection. It allows product managers to collect 360° product feedback, build prioritized roadmaps, and drive alignment across the organization.
Although I never used it, while I was exploring it, I reminded me to Aha; another tool we used at Nexthink when I was there. The UI is quite seamless, smoother, and better organized IMO than Aha.
The AI features are:
Classify bulk feedback: Chisel allows product managers to collect comprehensive feedback from various stakeholders, including customers, team members, and executives. The AI-powered feedback collection system ensures that all voices are heard and considered in the decision-making process.
Create feature descriptions and specs instantly: Make routine writing less of a burden. You can create feature descriptions and specs from all associated content and metadata.
Team Alignment: it helps drive alignment across the organization by involving team members in the prioritization process. It identifies areas of alignment and disagreement, allowing product managers to proactively manage conflicts and ensure that decisions are based on data rather than individual opinions.
airfocus
I’ve never used airfocus, but I’ve heard of it a while ago by someone and when asking GPTs which tools should I consider, it consistently appeared as one of the tools to investigate.
While studying its AI capabilities, it resembles a lot to Chisel and ProductBoard, because all have very similar features (Roadmap, Prioritization matrix, Feedback, and Backlog). In airfocus, as they came to call it, AI Assist offers the following functionalities:
The use of slash command to quickly start with a first draft and iterate from there
Editing suggestions and analyze feedback sentiment and get short summaries to generate insights fast:
To be honest, the AI features didn’t blow my mind. It seems that after reviewing many tools I was expecting to see this, and probably something that surprises me, but it didn’t happen.
Conclusions
As you have observed, many AI tools were presented, some of them with a very narron focus and others with multiple features integrated on them. That’s the beauty of this, despite of platforms attempting to cover most of the workflow from feedback gathering, to specs creation, prioritisation, and summarisation; you can combine multiple tools to get the job done.
For me it’s fascinating how a very manual tasks that involves the creation of a PRD or specs, has advanced thanks to the raise of many AI tools. This not excludes by any means that PMs form an opinion and diagnose what is going on, but accelerate how quickly you can move from thinking into action.
This in my opinion is a clear example of how you can augment your skills as PM and move quickly to deliver more things faster.