Best of LinkedIn: Digital Products & Services CW 30/ 31

00:00:00: This episode is provided by Thomas Allgeier and Frennus, based on the most relevant LinkedIn posts about digital products and services from CW-Thirtyandthirtyone.

00:00:09: Frenness is a BDB market research company that supports enterprise product teams with building featurebyfeature competitive intelligence That shows exactly how their product stacks up against the competition.

00:00:21: You can find more info in the description.

00:00:23: So, you know for the last five years product managers have been pretty heavily rewarded.

00:00:27: Well how well they can put together a slide deck?

00:00:30: Right yeah.

00:00:30: or aligning internal stakeholders?

00:00:32: Exactly but almost overnight and we're seeing this everywhere.

00:00:34: The rapid integration of AI has made all that like Brodic Theater virtually

00:00:40: workless.

00:00:41: It really is.

00:00:42: So today we're taking a deep dive into the digital product trends from across LinkedIn.

00:00:46: We want to figure out why the teams that are actually winning right now, or the ones who were you know ruthlessly cutting out the noise?

00:00:52: Yeah because...we are seeing a complete redefinition of the industry.

00:00:56: I mean Product Management Discovery Platform Strategy Operations they all facing massive survival test and common thread in sources we curated for this Deep Dive is that organizations pulling ahead They aren't just buying the most AI tools.

00:01:12: Right, it's not just an arms race of who has the most tech

00:01:15: Exactly.

00:01:16: they're using AI to force a fundamental evolution in how they make decisions.

00:01:21: so The focus is really shifted away from maximizing output To you know sharpening their actual judgment

00:01:28: which brings us straight to the product manager seat.

00:01:30: honestly We constantly hear this mainstream narrative about AI like generating images or writing front-end code.

00:01:38: sure But the most aggressive disruption is happening right to the PM job description.

00:01:43: I was reading a breakdown from Tom Forilli, he's chief product officer at what not and his take completely flips conventional wisdom.

00:01:49: Oh yeah!

00:01:50: His post was fascinating.

00:01:51: Right.

00:01:52: He argues that AIs biggest unlock for a PM isn't rapid prototyping it actually data science.

00:01:57: And that completely changes the bottleneck of product development Like just think about workflow back in say twenty seventeen or twenty eighteen.

00:02:03: Oh man night & day.

00:02:05: Seriously If a PM wanted...I don't know pull cohort retention data or analyze user logs to figure out why a feature was failing, they had to submit it ticketed to dedicated data scientist.

00:02:17: And

00:02:17: then just sit around and wait for the result?

00:02:19: Exactly!

00:02:20: But now that same PM can run those queries.

00:02:26: But

00:02:26: how does that actually change the culture of the product team?

00:02:29: Because if I'm a PM and suddenly have data scientist in my pocket, My first thought is just about speed.

00:02:35: I can go faster!

00:02:36: Yeah but it really takes you to step further.

00:02:38: You talk so much about the collapse of product theater.

00:02:40: Right i love this term Product Theater.

00:02:42: It's

00:02:42: SO accurate right?

00:02:44: It encompasses all those stakeholder management The endless alignment meetings.

00:02:48: They're really polished internal storytelling.

00:02:50: Basically they work looking busy Exactly.

00:02:53: And that theatre was allowed to thrive because Real verifiable work like deep data analysis was expensive and it's slow.

00:03:01: Right, when the truth takes two weeks to uncover a really charismatic PM can just fill that gap with great narrative

00:03:07: Yep But now... Data analysis & verification are virtually free And instantaneous.

00:03:15: You cannot bluff your way through strategy meeting With shiny presentation

00:03:19: anymore Because anyone in this room Can literally query actual data In real time.

00:03:24: Precisely The smoke and mirrors are completely gone.

00:03:28: It forces such a complete pivot in what skills are valuable?

00:03:32: because if you can't rely on being a great presenter, You actually have to understand the mechanics of this system.

00:03:37: your building

00:03:38: Yeah, and Verily even mentioned that the fastest trending skill for his PMs is now systems thinking.

00:03:43: They're actually mapping PMS to specific problems rather than static teams.

00:03:47: Just

00:03:48: a build-up there wraps across different business domains.

00:03:50: right

00:03:51: exactly.

00:03:51: And this shift from presentation two actual business mechanics It's hitting the bottom line directly.

00:03:57: There was a great post from Oxana Drossovich.

00:03:59: Oh yeah She runs product at an AI native financial software company Right?

00:04:04: That's The one She posted about the downstream effects of all this automation.

00:04:08: Over the past twelve months alone, AI has completely absorbed her team's strategy decks — their first draft product requirement documents—all they're

00:04:17: ticker-writing.".

00:04:18: So if the AI is doing all that traditional coordination work… The PM has to own something way more substantial just to justify its role?

00:04:26: Right and Josephish noted ownership is rapidly consolidating toward whoever actually holds a P&L profit or loss.

00:04:32: Wow!

00:04:32: The mechanism there is crucial though... When the cost of generating software approaches zero, feature velocity is no longer a competitive advantage.

00:04:39: No not at all.

00:04:40: shipping ten features which doesn't matter if none of them actually move the financial needle.

00:04:44: So the PM's scoreboard changes entirely.

00:04:47: she's tracking hard metrics now like net retention rate annual recurring revenue annual contract value.

00:04:54: The product manager Is no longer just the backlog administrator.

00:04:58: they are becoming the gatekeeper for revenue

00:05:00: Which is a massive identity shift and honestly it's kind of terrifying one for anyone who built their career purely on like Agile certifications in Gira management.

00:05:10: Oh, absolutely.

00:05:10: But I do want to push back a little on the idea that AI just seamlessly takes over the strategy.

00:05:16: Pawe Herron raised a really great point about the limitations

00:05:19: here at The Context Issue,

00:05:20: exactly.

00:05:21: he pointed out that while AI can summarize an hour-long customer interview in five seconds It has zero understanding of the actual business context.

00:05:31: it completely lacks the intrinsic Understanding of the customers.

00:05:35: unstated pain points like empathy cannot be automated through a language model.

00:05:40: Jerka Helmick actually posted a very strong warning about this specific trap.

00:05:44: He's watching teams adopt AI-native product management tools, and instead of using them to elevate their strategic thinking...

00:05:51: They're just using it to generate cheap automated specs?

00:05:54: Yes!

00:05:55: It gives him the false sense of accomplishment.

00:05:57: You know you click button A beautiful PRD spits out And feel like Hey I did my job

00:06:02: But you've actually insulated yourself from user.

00:06:05: Because generating a document cheaply doesn't mean the problem described in that document is even worth solving.

00:06:11: If you remove the friction of writing this spec, You really risk removing deep thought required to understand if it should exist at all.

00:06:35: Definitely, because that trap of generating cheap specs leads us to a massive structural

00:06:40: problem.

00:06:41: How so?

00:06:42: Well if AI makes writing code and generating requirements incredibly inexpensive the traditional bottleneck vanishes.

00:06:49: right.

00:06:49: execution is no longer the hard part

00:06:51: Right!

00:06:51: So friction shifts entirely figuring out what build in first place.

00:06:55: Exactly this where product discovery becomes dividing line between serious tech companies just feature factories.

00:07:02: Felix von Kuhnhart made this argument that delivery is becoming a complete commodity.

00:07:07: Because if AI allows your engineers to build twenty times faster, the most dangerous thing you can do is just feed that machine.

00:07:13: everything in your backlog?

00:07:15: Yes!

00:07:16: You will drown in fast useless features... ...that just bloat product and confuse user Right.

00:07:22: So Von Kuhnhart insists teams must establish strict kill criteria up front to combat that

00:07:29: bloat.

00:07:29: Yeah,

00:07:30: can you define exactly what?

00:07:31: That looks like in practice.

00:07:32: Like why is kill criteria so urgent right now?

00:07:35: well it comes down to the sunk cost fallacy.

00:07:38: It's just deeply ingrained in human psychology.

00:07:41: once a prototype exists It becomes incredibly hard-to-kill.

00:07:44: Oh for sure engineers get attached to the code

00:07:46: pms Get attached to launch

00:07:48: and sales team who start promising at two clients.

00:07:51: Exactly When building a feature took three months you had plenty of time to evaluate it, but when it takes Three days the features live before.

00:07:59: You've even had a chance To debate its merit.

00:08:01: so you have to agree on The exact failure conditions Before you start

00:08:05: right?

00:08:05: You need say here is the metric that if not hit means we delete the code and you Have to agree On that before a single engineer touches A keyboard.

00:08:13: I mean That requires a level Of discipline.

00:08:14: most teams frankly do Not have.

00:08:16: no they don't.

00:08:17: But you know, discovering what actually matters to users doesn't always require this massive formalized framework.

00:08:24: Sometimes teams overcomplicate discovery just to justify their own existence.

00:08:28: That's a great point!

00:08:29: Yeah, Gilad Bechar pointed out something brilliant along those lines... He said product teams will waste months debating roadmaps in conference rooms when their ideal roadmap is literally sitting publicly.

00:08:42: In there app store reviews,

00:08:43: the behavioral data has already aggregated and waiting for them

00:08:46: exactly.

00:08:47: he noted that one star review show you exactly what is broken and causing churn.

00:08:52: three Star Reviews Show You What Is Almost Working But Causing

00:08:55: Friction.

00:08:56: And five-star reviews

00:08:57: to show you The Core Value That Things Users Would Absolutely Hate To Lose.

00:09:02: multi-week discovery sprint.

00:09:04: Sometimes you just need to read what people are already screaming at you.

00:09:07: Yeah, Parth Malpathak shared a great real world example of this organic listening.

00:09:11: Oh I saw that!

00:09:12: He built private travel journal app right?

00:09:14: The Journey

00:09:16: That's the one.

00:09:17: he didn't run focus groups or send out complex surveys...he'd paid attention to casual questions his family and friends asked him while planning a cross country drive.

00:09:26: Like what kind of question?

00:09:27: Simple things like can i see where your stopped on map?

00:09:31: Can we comment on the photos?

00:09:33: Those natural, unprompted questions became his entire feature backlog.

00:09:37: Wow!

00:09:39: Genuine customer signals are often just hidden in plain sight.

00:09:43: But let me play Dibble's advocate here for a second.

00:09:46: If execution is practically free and can build things almost instantly with AI Shouldn't we just ship everything quickly and let the live user data sort it out?

00:09:55: A lot of people think that, yeah.

00:09:57: I mean why bother with deep methodical discovery if The cost Of being wrong is Just a few hours of automated coding?

00:10:03: because the cost of Being Wrong isn't just engineering time anymore.

00:10:06: What Is It then?

00:10:07: its User Trust Its Interface Clutter Its Massive Technical Debt?

00:10:12: Bushacuse Cooner And Christian Raebe Tackled This Exact Dilemma Using the GIS Framework Goals

00:10:18: Ideas Steps and Tasks Right

00:10:20: Exactly.

00:10:21: Yeah, they argue that most items in a backlog actually don't need deep discovery.

00:10:25: really yeah.

00:10:26: if an idea isn't highly risky If the complexity is low and the potential impact as obvious You shouldn't waste weeks researching it.

00:10:35: you assign or release date define A strict success metric ship It and measure the fallout

00:10:40: okay?

00:10:40: So you reserve The heavy discovery mechanics for the big existential bets.

00:10:45: yes Discovery Is An Expensive Tool.

00:10:48: its meant For Highly Risky complex bets where being wrong actually damages the brand or wastes significant capital.

00:10:55: That makes a lot of sense

00:10:57: and when you are validating those big bets, The focus has to be surgical.

00:11:01: Annalisa Reinsen and Otto Augustine highlighted takeaways from recent master classes that show how most teams get this completely wrong.

00:11:08: How so?

00:11:09: Well,

00:11:09: future teams usually just test for basic usability.

00:11:11: They put a mock-up in front of the user and ask hey can you find the checkout button

00:11:15: which tells you literally nothing about whether they actually want to buy the product?

00:11:18: exactly The problem.

00:11:20: true product team's tests For value will be used or actually switch To this solution from whatever work around there using today.

00:11:27: Will they pay for it while they advocate for?

00:11:29: yes You have to de risk those specific assumptions through targeted customer conversations long before you build.

00:11:36: As the saying goes, AI won't stop you from building the wrong thing.

00:11:40: It will just help you build the wrong things much faster.

00:11:43: Let's follow that logic to its natural conclusion.

00:11:46: then Okay Say you nail the discovery phase.

00:11:49: You use the GIS framework To validate the big bets Use AI to build a solution in record time And launch it.

00:11:56: Sounds like a dream

00:11:57: Right, but you're releasing it into a market where every single one of your competitors also has AI and they are building just as fast.

00:12:06: The speed advantage completely cancels itself

00:12:08: out.

00:12:08: It really does

00:12:09: So in a flooded market.

00:12:11: what actually separates the winner from noise?

00:12:13: Well the battleground shifts entirely to positioning and platform strategy.

00:12:18: Its no longer sufficient to have a highly functional product.

00:12:22: How do you frame that product against the rest ecosystem?

00:12:25: That's what determines survival.

00:12:26: Yeah, Anthony Peary had a very aggressive take on this.

00:12:29: He argues that standard Silicon Valley product advice like ignore your competitors and focus on the user is actively destroying company positioning...

00:12:39: Oh totally!

00:12:40: But that goes against a decade of tech startup culture.

00:12:43: why is ignoring competitors suddenly fatal?

00:12:46: Because that advice only works in immature brand new categories where the market doesn't even know who the players are.

00:12:53: yet If you're launching a Challenger brand into a mature established category, ignoring your competitors is suicidal.

00:13:00: Because the user's already comparing to them anyway.

00:13:02: Exactly!

00:13:03: Piri points out breakout tools like Gamma, Linear and Figma.

00:13:07: They didn't ignore massive legacy incumbents in their spaces.

00:13:11: they won by building positioning explicitly against them.

00:13:14: They identified the exact bloated, slow workflows of The Incumbents and built a surgical strike against those specific pain points.

00:13:22: Yeah?

00:13:23: Ali Memoji brought up an historical example that perfectly illustrates this dynamic —the battle between Zoom & Skype—.

00:13:29: Oh!

00:13:29: That's

00:13:29: classic

00:13:30: one... Right Back in twenty eleven Microsoft bought Skype for eight point five billion dollars.

00:13:35: They had hundreds of millions of users, unlimited engineering resources and just total market dominance.

00:13:41: But Microsoft tried to turn Skype into a broad everything-to-everyone communications platform.

00:13:47: Yeah they bloated the product trying to serve every conceivable use case

00:13:50: Leaving the door wide open for Eric Yuan and found Zoom.

00:13:55: And zoom did not try to be a comprehensive platform.

00:13:58: in those early days?

00:13:59: Not at all, they focused on one highly specific maddening frustration making video meetings frictionless.

00:14:06: no account required just click the link and the camera turns on.

00:14:09: They absolutely dominated the pandemic era because they refused to be distracted by broader feature parody.

00:14:15: It's

00:14:15: The Difference Between A Swiss Army Knife & A Scalpel.

00:14:18: Sometimes being incredibly precise at making one specific cut is infinitely better than offering fifteen mediocre clunky tools.

00:14:26: Yeah, but you know that's couple approaches brilliant for early stage market entry.

00:14:31: once You have that foothold the long-term survival strategy actually requires a pivot.

00:14:35: well Khalid M shared A strategic insight indicating That The rigid One size fits all sauce model Is actively dying?

00:14:43: The future is moving toward what he calls composable software.

00:14:47: Unpack the mechanics of composable software for me.

00:14:49: Like what does that actually mean?

00:14:51: For The end user,

00:14:53: it means the inherent value Of software is moving away from individual siloed applications.

00:14:59: It's moving toward the orchestration of workflows across multiple tools.

00:15:02: Okay a modern user doesn't want to open fifteen different tabs and manually stitch their data together.

00:15:08: They expect intelligent systems, usually driven by AI agents to adapt their specific context.

00:15:14: Pulling data from one tool and executing an action in another.

00:15:18: So

00:15:18: the user interface of a single app is becoming less important than workflow intelligence that connects it with other users' ecosystem.

00:15:25: Exactly!

00:15:26: The product has become invisible glue.

00:15:28: If you are just siloed feature You get replaced.

00:15:32: And Aiden Ziyapur provided empirical evidence for this, actually when he was analyzing Revolut's latest annual report.

00:15:38: Oh

00:15:38: right!

00:15:38: Revolut grew its revenue by forty-six percent year over year

00:15:42: Which is massive.

00:15:43: and Ziyipur pointed out that This growth wasn't driven By launching one killer feature.

00:15:48: It was driven entirely by platform depth

00:15:51: Meaning they locked the user into an ecosystem Yes

00:15:54: banking, investing payments credit all reinforcing each other within a single architecture.

00:15:59: In that platform depth vastly outperforms Single Feature apps because it drives down customer acquisition cost while massively increasing the lifetime value.

00:16:09: Because customers trust the cohesive ecosystem they choose Revolute as their primary financial relationship And that creates a competitive moat, That a shiny new standalone budgeting app just simply cannot cross.

00:16:21: But running the platform with that level of depth requires flawless internal mechanics.

00:16:25: Oh

00:16:26: absolutely You

00:16:26: can not orchestrate composable software or deep financial ecosystems without robust operations in design.

00:16:33: and yet based on the posts we are seeing both product operations and product design Are facing severe existential crises right now.

00:16:40: Yeah they're under intense pressure from The C-suite to justify their headcount in very measurable financial terms.

00:16:47: The crisis and product operations is rooted in this brutal paradox, Antonia Landy mapped it out perfectly.

00:16:54: Yeah go ahead.

00:16:55: When a product ops professional does their job flawlessly friction disappears from the organization communication flows data is clean delivery pipelines are smooth And whole product machine just hums quietly

00:17:08: Which was literally entire goal of role.

00:17:11: It IS.

00:17:12: But human nature and corporate leadership is reactionary.

00:17:16: When everything's working perfectly, Leadership looks the balance sheet sees.

00:17:20: a high-paid ops team asks wait why are we paying these people if there're no fires to put out?

00:17:25: They

00:17:25: optimize themselves right of job!

00:17:27: Exactly!

00:17:28: Juall Mello validated this from first hand experience.

00:17:31: He noted that he has survived three different corporate restructures And in every single instance The ops roles were the first ones targeted for elimination Man...

00:17:40: So what's his take away?

00:17:41: It's a harsh reality check.

00:17:43: Ops must stop branding itself as helpful

00:17:45: because helpful is completely expendable when budgets get tight.

00:17:48: exactly think about city infrastructure.

00:17:52: if product ops just acts like the street sweepers cleaning up the friction so everyone else can drive fast leadership will eventually cut their budget.

00:18:01: The moment, the streets look relatively clean.

00:18:03: So ops needs to stop being the street sweepers and start being power grid.

00:18:07: Yes,

00:18:08: they have to own end-to-end supplier workflows The core reporting data systems The critical infrastructure that would plunge city into a blackout if it walked out of door.

00:18:17: If you just facilitate You get cut!

00:18:22: And Lizzie Amagus added an important layer by defining where the PM ends in Ops begins.

00:18:29: What's her take on it?

00:18:30: She argues that many companies searching for a product manager are actually desperate.

00:18:34: For a Product Operations Manager, she draws the hard line.

00:18:37: The Product Manager must own what and why.

00:18:40: So market strategy, adoption metrics, revenue impact

00:18:44: And the Ops Manager.

00:18:45: The Product Ops Managers must own how when they win Internal delivery systems Cross-functional workflows Scaling processes

00:18:52: Right.

00:18:52: Because when you force a single person to quietly juggle both the why and how, The strategic roadmap gets fuzzy And operational deadline slip.

00:19:01: It's just structural mismatch.

00:19:03: So if operations has become the irreplaceable power grid To survive.

00:19:08: What is survival strategy for product design?

00:19:11: Because with AI generating polished high fidelity interfaces in a matter of seconds designers are in danger Of being permanently boxed into the make it look pretty corner.

00:19:21: design only survives if It changes its financial framing.

00:19:25: Okay, so

00:19:26: well Francis Kaiser shared A really compelling turnaround story on this front.

00:19:30: He realized his design team was trapped In the classic feature factory cycle.

00:19:34: They were only being brought in at the eleventh hour to apply pixel polish To decisions that had already been made which

00:19:40: is incredibly common

00:19:41: right.

00:19:42: so to break out of that trap He completely shifted his team's mandate away from design velocity and he anchored it directly to behavioral outcomes.

00:19:50: Wait, how does a designer map?

00:19:51: A behavioral outcome to the company's bottom line?

00:19:54: by tying The user experience directly.

00:19:56: two engineering costs and revenue generation.

00:19:59: Keezer's team started integrating high-fidelity prototyping extremely early in the discovery phase.

00:20:06: By doing cognitive load studies up front, they simplify user flows before any code was even written and this reduced developer rework by twenty five percent.

00:20:16: That is a massive reduction of wasted engineering salaries Huge!

00:20:20: And furthermore... By removing friction in the conversion funnels, they increased revenue by six percent.

00:20:25: So The C-suite instantly stopped viewing design as a cosmetic cost center and recognized it as a margin expanding profit engine?

00:20:33: Because They could finally see the math!

00:20:35: It wasn't some subjective debate about color palettes... ...it was proven formula of saving engineering time and boosting conversion.

00:20:41: That's

00:20:41: brilliant!

00:20:42: Brian Smigzeski reinforced this exact imperative too.

00:20:45: He stated that designers have to stop outsourcing definition success for other departments.

00:20:51: Right, because too often designers accept vague top-down business goals like improve the user experience or modernize the platform.

00:20:58: Exactly!

00:20:59: They get lost in a creative process and then they get judged on arbitrary metrics that had no hand in creating.

00:21:05: So as Majeski argues, designers need to aggressively define their own measurable success metrics.

00:21:11: Yeah

00:21:12: If a business wants to attract new customers The design leader needs to ask how many at what acquisition cost and through what specific interface flow.

00:21:22: They have to pull their own data, map design choices directly to growth retention or operational efficiency.

00:21:29: If they refuse speak the language of business metrics it will always be treated as a subordinate workflow group rather than strategic partner.

00:21:37: It really all converges on single theme doesn't?

00:21:39: accountability, whether you're in product management discovery platform strategy ops or design.

00:21:45: You can no longer hide behind the sheer effort of production now that days have been rewarded for how hard your work to build something are just over because AI has made the act of building sheep and effortless judgment.

00:21:57: precise measurement And direct alignment with revenue Are literally The only currencies That matter Now

00:22:02: which leaves us With a fascinating almost philosophical conclusion To draw.

00:22:06: from all these sources We are watching the barrier to building software essentially drop to zero.

00:22:12: Yeah, very soon every single enterprise startup and solo developer will have the exact same infinite AI execution capabilities at their fingertips.

00:22:21: And when technology is completely democratized and output is infinite The ultimate competitive mode won't be your tech

00:22:28: stack.

00:22:29: it will revert entirely to raw human taste, intrinsic intuition and the sheer

00:22:52: ability.