MindTheProduct

MindTheProduct The world’s most engaged product community, providing articles, meetups, training, and events that support the careers of 300,000+ product people globally.

Join the community at www.mindtheproduct.com.

"I've shipped three products I was genuinely proud of with clean architecture and features people actually asked for. Al...
07/10/2026

"I've shipped three products I was genuinely proud of with clean architecture and features people actually asked for. All three are gone."

Good engineering without strategy is just an expensive way to fail.

In this community contribution, fractional CTO Alex Ivanouski shares what he learned building three products that failed for three different reasons: a discovery failure, a risk failure, and a pricing failure.

He shares the discipline he uses now, and five questions to pressure-test any strategy before you commit.

Read the full article 👉 https://bit.ly/3VQmedm

On LinkedIn, it seems all PMs have become AI builders overnight, posting the results of what they are building before yo...
02/10/2026

On LinkedIn, it seems all PMs have become AI builders overnight, posting the results of what they are building before you've finished reading the last AI summary.

In this article, Marta Rolak describes two camps forming under AI pressure: the introverts, who go quiet, compare themselves to a distorted version of LinkedIn reality, and distrust their own judgement. And the extroverts, who build faster, ship more, and quietly outsource their product thinking to the tools.

Her argument: both are making the same mistake. Neither is applying logos, a Greek word meaning reason or rule.

"The real risk for product managers is not that we are failing to use AI fast enough. It is that the pressure to keep up is making us forget the core of the job: what to build and why."

Read the article for her three ways to pay attention to the logos 👉 https://bit.ly/4jwBZQq

September was a busy month across the ProductTank community, as we hosted 69 meetups across the world, a lot of conversa...
01/10/2026

September was a busy month across the ProductTank community, as we hosted 69 meetups across the world, a lot of conversations around AI, and careers on almost every agenda.

Some highlights:
🇬🇧 Birmingham: James Gunaca on running your career like a product, with a skills assessment to find the gaps
🇩🇪 Cologne: three different career paths with Marlene Jung, Esjona Musta, and Shehrina Kamal, different experiences and different stages of the journey, one common thread: stay curious.
🇦🇪 Dubai: what AI can't build for you

Do you want to join in the fun? Check out if there is a ProductTank near you 👉 https://bit.ly/4wyZBaN

Or join ProductTank Global, coming very soon 👀 👉 https://bit.ly/3U2kAEX

Most teams reach for ROI after someone questions a decision. However that’s too late.“ROI doesn’t need precision. It nee...
01/10/2026

Most teams reach for ROI after someone questions a decision. However that’s too late.

“ROI doesn’t need precision. It needs written assumptions.”

In this community contribution, Olha Kashpur, Senior PM at Flo, shares two frameworks she uses to make ROI thinking part of everyday product decisions, not just annual planning.

1️⃣ The Feature ROI Scorecard: forces you to define what “gain” actually means before you build. If you can’t agree on that, you’re not ready to build.
2️⃣ Invest, Harvest, Watch, or Exit: a matrix for existing features that makes the “should this still exist?” conversation a lot easier to have.

What matters is the process, which will surface disagreements that would otherwise show up six months into delivery.

Read the full article 👉 https://bit.ly/4jwBZQq

30/09/2026

Do product people need to learn to code? Kendra Vant goes beyond a simple yes or no.

Kendra, CPO at Tapi, says product people are more effective when they understand how software is built, which is not the same as being able to build it securely yourself. She also wouldn't hire a PM who isn't comfortable in git.

She joins Randy Silver to explore the “reliability layer”: the people, checks and engineering that make AI products dependable. She explains why enthusiastic AI users often provide that layer themselves, and why product teams can’t assume their customers will.

Watch the full episode to find out the 4 questions to ask before building AI into your product, and how to design for failure. Do product people need to learn to code? Kendra Vant goes beyond a simple yes or no.

Kendra, CPO at Tapi, says product people are more effective when they understand how software is built, which is not the same as being able to build it securely yourself. She also wouldn't hire a PM who isn't comfortable in git.

She joins Randy Silver to explore the “reliability layer”: the people, checks and engineering that make AI products dependable. She explains why enthusiastic AI users often provide that layer themselves, and why product teams can’t assume their customers will.

Watch the full episode to find out the 4 questions to ask before building AI into your product, and how to design for failure. 👉 https://bit.ly/3TxJlbN

In this community contribution, Andrii Zabluda shares how early operational practices and architectural pivots shaped th...
29/09/2026

In this community contribution, Andrii Zabluda shares how early operational practices and architectural pivots shaped the Trade Flow Builder interface at Kinertic.

He details how the team built their early-validation process, which they refer to internally as Fast-Ideation. It allowed them to recognize — and step away from — the first, deceptively familiar interface they tried, in favour of a full-screen canvas.

The domain itself, not just the interface, dictated the tree-like structure that canvas needed to represent, a structure that turned out to directly support compliance with regulatory standards such as EUMR and CBAM.

That’s what happens when you follow domain logic instead of interface convention.

Read the full article 👉 https://bit.ly/3VcOHde

Has your AI become as native to your workflow as Notion or Jira? Do you have six to seven AI tools open at once working ...
24/09/2026

Has your AI become as native to your workflow as Notion or Jira? Do you have six to seven AI tools open at once working in the background?

Sound familiar? It did for Justice Onyekachi, Technical PM at Evu, and then two friends at separate Big 4 consultancies told him the same thing: commercial AI tools were either heavily restricted or outright banned at their firms.

That conversation made him think about his own usage. If some of the most analytically sophisticated organizations in the world won’t touch the tools he uses daily, what exactly was he handing over every time he pasted a user interview transcript into ChatGPT?

In this community contribution, Justice shares the four things he changed with his AI usage to limit the data exposure.

Read the full article 👉 https://bit.ly/4hcZL12

Structured data becomes a PDF, then the PDF gets sent to an AI, which converts it back to structured data. This loop run...
22/09/2026

Structured data becomes a PDF, then the PDF gets sent to an AI, which converts it back to structured data. This loop runs millions of times a day, and nobody is counting the energy cost.

In this community contribution, Nick Williams argues that the conversation about AI’s energy footprint is almost entirely focused on inputs, and overlooks a different category of waste: generated not by the AI industry itself, but by the data architecture decisions of the organizations that rely on it.

But what if the AI can also provide the solution?

Read the full article to see the five-question decision framework to avoid AI bloat 👉 https://bit.ly/3Vs2v3r

"Language in a product is not a feature. It’s infrastructure. And most of the time, we’ve only built about 20% of it bef...
21/09/2026

"Language in a product is not a feature. It’s infrastructure. And most of the time, we’ve only built about 20% of it before calling it done."

Udit Mehrotra, Head of Product for North America Languages Experience at Amazon, noticed something odd in his conversion data: French-speaking customers were getting decent traffic but purchasing at a lower rate than English speakers.

The only difference was the language experience.

The title and the bullet points had been translated. However, the images (size charts, infographics, installation guides), which carried the most product-critical information, hadn’t. An other important point Udit makes is that translation is not localization.

Read the full article 👉 https://bit.ly/4hfy4EX

Best US city + Product community + World-class lineup = Unmissable conferenceThat's mtpcon Chicago, Harris Theater on O...
17/09/2026

Best US city + Product community + World-class lineup = Unmissable conference

That's mtpcon Chicago, Harris Theater on October 7.

There are only a few weeks to go, and our workshops are already sold out (we did warn you 👀). So don't wait too long to grab your conference ticket. 👉 https://bit.ly/4xLkUXt

Address

London

Website

Alerts

Be the first to know and let us send you an email when MindTheProduct posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Contact The Business

Send a message to MindTheProduct:

Shortcuts

Share

Category