What Should Product Leaders Be Talking About?

We have lots of crazy stuff happening throughout the world, especially with anything tech or software worker related. I’m curious on what are truly the most important things that people should be talking about (or what they want to talk about).

A couple of topics I’m super invested in:

  • AI Agents - what makes them work
  • The Future of Work - what this means and how to position yourself for success
  • Metrics - how measurement matters in early process to gauge what you should be doing

Curious if anyone has any thoughts, topics, or ideas that they’d like to be talking about

1 Like

Well, those are all interesting to me. They all overlap in my world. The function of product management is changing just as quickly as design. So these topics are super relevant.

I continue to be curious about how companies and leaders put value on these things.

2 Likes

I think that a big one for me is how leaders are supporting their teams through these waves of uncertainty. From a talent perspective, I’ve seen people chirping about how AI is increasing output expectations faster than it’s increasing support or clarity.

I think people see it as their jobs are being redefined without any real conversation about how its going to be supported or what it means for their longevity.

2 Likes

This is a really good one @nathaliesmith !! Business leaders can definitely be helping with either easing the pain or helping figure out what’s next.

We need help grounding ourselves in a world of chaos (aka innovation)

I think this comes down to keeping agency, relevance, and credibility. Especially when tools are faster and organizations are on shakey ground. Many of these decisions feel more political than rational.

Based on my online browsing of LI, YouTube and Reddit. Here’s how I’d frame this:

  1. AI is changing jobs faster than teams can keep up
  2. How to stay valuable when the job market feels unstable
  3. Using proof to avoid opinion-driven decisions

Here are some topic ideas:

  • How AI agents really work in practice
  • Where human judgment still matters and where it does not
  • Fear of becoming made obsolete by a “prompt middle layer”
  • How to trust systems that act on their own
  • Layoffs and role compression across design, PM, and engineering
  • Whether leadership titles still matter
  • How to move from IC to influence without a title
  • How to prove value when output is no longer as big a deal
  • Frustration with “vibes-based” decision making
  • Executives asking for outcomes without know what success is
  • Analytics overload without insight (just going to get worse)
  • Early signals versus lagging KPIs

For many of the leaders here, I don’t think it’s about new tools or trends (those are happening too fast). We should be focused on making good decisions under uncertainty and then prove those decisions mattered with evidence.

1 Like

Loving the list- I think the hardest pill to swallow is that whether or not the skills that people have been learning for decades will now be close-to-worthless when it comes to creating an income that can support a family.

These are definitely needed conversations. I’m fully on-board with being flexible and know that a lot of this comes from a sunk-cost fallacy.

haha, this is the kicker and nail biter, IMO. “We should be focused on making good decisions under uncertainty and then prove those decisions mattered with evidence.”

Gotta have a strong gut feeling to follow for this to work. :wink:

1 Like

We have to create culture that supports experimentation and quick failures! Creating output matters less now. What matters is whether those outputs actually move the outcomes we care about (wether that’s visuals, code, workflows, etc)

For everyone who never wanted to manage people, this is the job now: managing your bot, without the feelings. :slightly_smiling_face:

It’s like going back to Tamagotchis.

2 Likes

haha, you know, I never actually owned one of those if you could believe it! There is such an interesting line we’re walking here with the human + AI combo.

Humans are notoriously predictable & unpredictable all in the same breath, which makes the work unique, fun & inspiring (perhaps hard, too) . I tend to think that the more collaborating or (managing) a robot that happens, we’ll lose that spark. I HOPPPEEEE that’s not the case, but I do think there is some beauty in the chaos. But I suppose that now, the chaos isn’t gone, it just looks different. Ya know?

(Story of a business’s life – not less problems, just new problems.) haha.

1 Like

In the mid-90s, as a toy inventor, I spent a month building a prototype similar to Tamagotchis using Macromedia Director (precursor to Flash) and Lingo for logic. It wasn’t even highly functional… more of a prototype. Maybe 20 people saw it?

That same prototype can now be built into a functioning app in less than an hour. I tend to see this opening up all kinds of new ways to collaborate and spark creativity among creators.

Like @EricZ’s tool, this probably wouldn’t have been worth weeks of effort a few years ago. Now it’s a real talking point and a collaboration artifact that supports handoff. Maybe it’s a throwaway app. Maybe it’s the start of a more advanced workflow. In the short term, that distinction doesn’t matter. What matters is the outcome it helps create.

The future is bright!

1 Like

Agreed! Or at least gotta keep leaning into the positives.

The key here is to have fun with the process. If this isn’t exciting for you now, what is it going to take to get you to that point?

Maybe you’re not seeing the benefits as much as you’d like yourself. My recommendation is to experiment and explore!

I saw this in a reddit thread.. “Mine PM is so hands off that I get no feedback and he has no interest in defining how we work just leaves us to the chaos to figure it out and monitors deliverables. This has definitely impacted my ability to learn and grow and I haven’t learned exactly what I should and should not do in any optimal way.”

I thought this was interesting because I think we often see people walking this line of wanting to be successful but the leadership being so disjointed they forget their people need help. A comment I’m seeing more and more online these days. “Every man for themselves” type of mentality.

1 Like

To add to this… Product leaders don’t need to have all the answers right now, I’d argue that they do need to be talking about the right things with their teams. The things people care about holistically:

What’s changing vs what’s not. (What principles still guide decisions)
People can handle change but what drains them is not knowing what’s still solid and whats out the door. Stability isn’t pretending nothing is changing.

What we’re experimenting with, why (and how the team can contribute) What problem are we actually trying to solve with new technology and how can they actually contribute with impact.

Where judgment matters more than speed. Where is your team willing to take risks or where are we intentionally slowing down for maximum impact?

What skills actually matter for growth. People are anxious about careers in this current climate. Outline what skills are becoming more valuable and what skills do they currently obtain that support scale.

2 Likes

I am looking for ways to build a more comprehensive framework around VFD and how to systemize it within my team. Essentially it means the following:

  1. Viability (hitting the core KPI - do i have a spare kidney for this?)
  2. Feasibility (technical and systemic ability to deliver - do we need a flamethrower to light a candle)
  3. Desirability (will the end users like this - like why do we have earwax flavoured jellybeans at all??) for anything.

Any product leader must think of these three parameters and KPIs associated with them.

3 Likes

That’s a great breakdown. (also appreciate the sense of humor :slightly_smiling_face:)

We’ve defined Desirability as a positive user response in Sentiment and a self-reported likelihood to use: https://glare.helio.app/define/ux-metrics/attitudinal-metrics/desirability

Curious how your team is thinking about Desirability in your context?

3 Likes

Self reported likelihood to use is a good indicative metric. But it has a limitation I believe because people always over or u underestimate their behaviour to align with what’s considered ideal behavior for a hypothetical scenario. Also the source of the question can affect the answer.

We look at it as existing user journeys:

  1. Is the user completing the user journey?
  2. if yes, did the user get the desired result? This is typically a follow up
  3. Did people try edge cases (shows that they might be satisfied with normal user cases and want to test it’s usability more)

These at a few examples but it depends on the workflow being implemented.

I’m still formulating a set template which works in all domains. Will update once I have more clarity.

1 Like

Thanks for sharing Nikhil.

This is why we benchmark. We use UX metrics as relative numbers, not absolute ones. When you see shifts of 10 points, it’s a flag to look more closely at what’s causing the delta! Using additional metrics compounds understanding.

3 Likes