By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.
Product SchoolMural CPO on How To Put AI in Multi-Player Mode for Your Team | Elaina O'Mahoney
Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.
CHAPTERS
- 0:00 – 0:17
Ideas got cheap in the AI era—visual collaboration is what’s missing
Elaina O’Mahoney frames a core tension: AI makes it fast to generate compelling artifacts, but that doesn’t automatically help teams collaborate or decide together. She sets up the theme that the bottleneck has shifted from coming up with ideas to making decisions with shared context.
- •AI accelerates creating prototypes, docs, and slide decks in minutes or hours
- •Visual collaboration is described as “absent” in the AI era
- •The episode’s premise is that decision-making and shared context are now the constraint
- 0:17 – 1:36
Meet Elaina and the question of Mural’s competitive landscape
Carlos González de Villaumbrosia introduces Elaina as CPO of Mural and tees up the discussion around competition. The conversation sets expectations for a product-leadership lens on how fast the space is changing.
- •Elaina is introduced as Chief Product Officer at Mural
- •Carlos asks directly about Mural’s real competitive landscape
- •The discussion signals a forward-looking, product-strategy perspective
- 1:36 – 2:51
How Mural started: collaboration canvas born from distributed game-making
Elaina explains Mural’s origins and what it was designed to do: help teams brainstorm, diverge and converge, and then decide on a path forward. She also connects Mural’s pandemic-era adoption to the urgent need for new ways of working.
- •Mural emerged from co-creators in Argentina collaborating across locations
- •Core workflow: brainstorm, diverge/converge, then decide
- •Pandemic drove rapid adoption as teams needed new collaboration methods
- 2:51 – 4:23
AI feels like solo-player mode: lonely chat boxes and docs replace visuals
Elaina argues that many AI workflows isolate people in private interactions with an LLM, which reduces shared understanding. She contrasts this with visual collaboration that can make information more digestible and speed up decisions.
- •AI collaboration often happens alone via words or voice with an LLM
- •Coworker collaboration shifts toward docs and text-sharing
- •Visual collaboration is positioned as a faster path to decision-making
- 4:23 – 6:38
From artifacts to decisions: connecting context spatially to move faster
Elaina expands on why cheap idea generation isn’t the win by itself—teams still fall into silos when sharing outputs. She describes the need for a system that connects artifacts so everyone can quickly catch up and make decisions together.
- •Cheap creation leads to more artifacts but also more siloed handoffs
- •The goal becomes faster decision-making rather than “process” adherence
- •Spatially connecting artifacts helps teams share context
- •More ideas means more decisions, increasing the need for alignment
- 6:38 – 9:23
Tool-stack convergence and the “blank canvas” of decision context
Carlos points out how many tools claim to be “where work happens,” and Elaina describes a convergence toward similar chat-based interfaces. She identifies an unresolved gap: capturing decision context—the why—in the moment, not just post-meeting summaries.
- •Many tools converge toward similar collaboration claims and UIs
- •Slack/Teams, docs, and issue trackers cover planning and communication
- •A key gap remains: where decisions are made and why
- •Meeting notes summarize after the fact and often miss decision context
- 9:23 – 11:33
Lowering the bar: everyone can design and visuals surface disagreements faster
Elaina explains how Mural’s Luma practitioner approach pushes people to stop thinking “I’m not a designer.” She argues visuals accelerate comprehension and help teams reach disagreements, bottlenecks, and missing info more quickly.
- •Luma training aims to break the “I’m not a designer” mindset
- •Everyone can design, even if not at professional design proficiency
- •Visuals can be absorbed faster than reading a document
- •Visual collaboration helps teams reach disagreements and bottlenecks sooner
- 11:33 – 13:05
“Everybody should be playing and shipping”—but not always to production
Elaina describes how she expects teams to build and ship, while distinguishing experimentation from production releases. She highlights areas that remain scarce—like data science, taste, and analytics synthesis—and how leaders can use AI workflows and agents to speed teams up.
- •She expects broad experimentation: playing and shipping across roles
- •Shipping can happen in safe environments rather than production
- •Leaders can democratize decision bottlenecks by shipping workflows/agents
- •Scarce capabilities still matter: data science, taste, analytics synthesis
- 13:05 – 14:04
Kill the five-page strategy doc: prototype the North Star to align faster
Elaina critiques the traditional long strategy doc that circulates and “slowly dies.” She argues product leaders, PMs, and engineers can convey direction faster with prototypes that communicate principles and unlock quicker team decision cycles.
- •Large strategy docs create slow, fragile alignment loops
- •A prototype can communicate North Star direction faster
- •Prototypes aren’t necessarily the final shipped product
- •Visual conveyance accelerates ideation and decision-making cycles
- 14:04 – 17:10
Enterprise adoption: safe spaces, learning systems, and breaking rigid process
Carlos asks how to make large enterprises comfortable with new AI horizons, and Elaina emphasizes low-risk environments for learning. She describes a split between “haves” on the frontier and others doing basic use cases, and argues big companies must set up adaptable systems rather than lock into early processes.
- •Enterprises need lower-risk spaces for people to try and fail with AI
- •AI adoption can create a divide between frontier users and basic users
- •Early transformation needs system thinking more than fixed process
- •Innovation centers and smaller-team champions help scale change
- 17:10 – 19:14
Forward-deployed champions—sometimes the best “engineer” is in legal
Elaina argues forward-deployed engineers can accelerate learning by showing others how work gets automated, but she broadens the idea to any function. She shares an example of someone in legal automating a large portion of their processes and showcasing the work to trigger “light bulb moments” across the company.
- •Forward-deployed expertise spreads practical “how-to” examples
- •Champions don’t have to be engineers; any department can lead
- •Example: a legal team member automates a significant share of processes
- •Showcasing work publicly helps other teams learn and adopt faster
- 19:14 – 22:34
Building community: sharing how you work when there’s no playbook
Carlos explores Mural’s community angle and why people contribute without obvious ROI. Elaina describes inviting a subset of customers into a newer community and emphasizes the value of sharing practices, asking vulnerable questions, and learning outside one’s internal echo chamber.
- •Mural created a newer community with invited customer subsets
- •The community focus is sharing how people work, decide, and think
- •Community matters most when change is new and there’s no playbook
- •Trusting forums enable asking questions and learning from peers
- 22:34 – 24:40
Pricing in an AI world: seat-based SaaS vs token usage and the “black market”
Elaina says pricing is what keeps her up at night, especially as an older model collides with AI usage economics. She describes how token limits at large customers can lead to “black market” behavior and stresses the need to avoid disincentivizing use or forcing people into token accounting.
- •Mural’s legacy pricing reflects seat-based SaaS assumptions
- •Usage-based/token economics introduce new tradeoffs
- •She describes token constraints creating a “black market” inside enterprises
- •Pricing must work for solo users through very large enterprises
- •They want to avoid making customers track tokens like an accounting ledger
- 24:40 – 26:14
Measuring real AI impact: costs, fit-for-purpose usage, and predictability
Carlos asks how to ensure token usage reflects outcomes rather than gaming, and Elaina emphasizes measurement and appropriate use. She argues not everything should be done with AI, and says customers need predictable costs that don’t swing wildly month to month.
- •Start with a measurement layer to understand how AI is being used
- •Usage alone can be a misleading proxy for value and outcomes
- •Not every task should be done with AI; some uses may be unnecessarily expensive
- •Businesses need predictable costs and incentives that drive the right behavior
- 26:14 – 28:25
How Elaina structures teams: traditional triads vs hybrid, high-experimentation pods
Elaina describes running teams that vary by product area and risk tolerance. Core product work uses more traditional cross-functional triads, while newer bets can use hybrid roles and faster production cycles with less change-management overhead.
- •Team structure varies by area: broad capabilities plus specific personas
- •Core product uses more traditional designer/engineer/product triads
- •Some teams use hybrid roles (product engineer; designer-product hybrid)
- •Newer areas can run more untraditional, high-experimentation models
- 28:25 – 30:03
The innovator’s dilemma in practice: allocating bets while protecting the core
Carlos asks about balancing new bets with “keep the lights on” work, and Elaina names the innovator’s dilemma. She explains allocating resources across bets, learning what works, and reallocating quickly—while still grounding decisions in the current revenue engine and Mural’s collaboration focus.
- •Resource allocation must balance core revenue needs and future evolution
- •They spread investment across bets and adjust based on what’s working
- •Reallocation decisions are made faster than traditional long-range planning
- •Future work stays anchored to collaboration, bringing people together, and visual thinking
- 30:03 – 31:05
From annual planning to monthly loops: prioritize learning over shipping
Elaina explains that future-bet reprioritization happens about monthly, with her and the CTO reviewing what teams learned. She stresses asking “What did we learn?” rather than “What did you ship?” and describes operating like multiple mini-startups inside a decade-old company.
- •Future-bet teams are reviewed and reprioritized roughly once a month
- •The review question is centered on learning, not just output shipped
- •If thinking hasn’t changed, the team likely didn’t gather net-new information
- •They operate like multiple mini-startups within a company that’s been around over 10 years
- 31:05 – 31:45
Closing reflections: Mural’s decade-plus journey and a World Cup sign-off
Carlos reflects on early conversations with Mural’s founder and how far the company has come. Elaina notes the founder is still around, and the episode wraps with a friendly exchange referencing Argentina and Spain around the World Cup.
- •Carlos recalls meeting Mural’s founder in the early days
- •Elaina says the founder is still involved and remains an “idea person”
- •The conversation ends with a light World Cup-related exchange
- •Host thanks Elaina and closes the interview
