In late 2023, Groove leadership was alerted to growing concern from one of our largest enterprise customers regarding the clarity and credibility of our AI roadmap. The feedback was not about a single missing feature—it reflected a broader uncertainty about how AI would meaningfully show up across the Groove experience, and whether the company had a coherent plan for enterprise-scale use cases.

The timing amplified the risk. The customer’s renewal was approaching in the spring, and there was limited runway to respond. Waiting for long-term roadmap cycles or incremental discovery would not address the immediate concern. Leadership needed to demonstrate clear thinking, prioritization, and direction well ahead of renewal conversations.

At the same time, this was not a blank-slate problem. Many of the underlying ideas—AI-assisted actions, AI-authored communications, summaries, and intelligence layered across workflows—had already been discussed across product, design, and engineering. What was missing was not intent, but a shared, organized view of how these ideas fit together, how they would deliver value to enterprise customers, and which bets mattered most in the near term.

Leadership explicitly asked me to step in and help figure out how to bring clarity quickly. The challenge was not simply to produce concepts, but to create alignment—across executives, product managers, designers, and engineers—under tight time constraints and with incomplete inputs.

The stakes were real:

• Enterprise trust risk if we failed to articulate a credible AI direction
• Commercial risk tied to an upcoming renewal
• Organizational risk if teams continued to discuss AI in fragmented, uncoordinated ways

The goal was not to invent a roadmap under pressure, but to rapidly synthesize existing thinking into a coherent, defensible narrative—and to do so in a way that could guide real product decisions, not just satisfy a presentation moment.

This effort began as an executive intervention, not a planned product initiative.

Leadership asked me to step in and determine how to move forward. After assessing the ambiguity, timelines, and number of stakeholders involved, I proposed running a focused design sprint as the fastest way to bring structure, shared understanding, and momentum to the problem.

The intent of the sprint was explicit:

• Get all relevant disciplines working from the same frame
• Pressure-test AI ideas against real enterprise workflows
• Prioritize for impact and feasibility, not completeness
• Produce a concrete prototype that leadership and customers could react to

My Role

I led the effort as a player-coach, owning both the operating model and key parts of the design work.

My responsibilities included:

• Defining the sprint structure, agenda, and cadence
• Creating the initial framing deck to align leadership and participants
• Facilitating cross-functional working sessions and decision points
• Contributing directly to design exploration and synthesis
• Ensuring the work stayed grounded in enterprise needs and real workflows
• Representing design in leadership conversations throughout the process

While I proposed the sprint as the approach, the urgency and mandate were clear from the outset. This was an “all hands on deck” moment, and the sprint became the mechanism for turning diffuse ideas into a shared plan.

Team Composition

The sprint brought together a senior, cross-functional group:

• Multiple Product Managers and Senior Product Managers
• Two Product Designers from my team, working alongside me
• Input from Sales leadership, Sales Managers, and Customer Success
• Engineering partners engaged to sanity-check feasibility and sequencing

Names are intentionally omitted here—the emphasis was on collective ownership and speed, not individual attribution.

How We Worked

We loosely followed the structure of a Google-style design sprint, adapted for enterprise product complexity and a compressed timeline.

Over roughly two weeks (with a short extension for prototype polish), the team worked through:

• AI pattern research and best practices
• Rapid synthesis of existing roadmap ideas, internal discussions, and customer input
• User journey mapping, including connections to Groove Plays and existing workflows
• Structured exploration of seven AI feature areas, each broken down into:
  • How-might-we questions
  • Risks and assumptions
  • Data considerations
  • Interaction architectures
  • Low-fidelity ideation and Crazy 8s

Importantly, not every area was explored equally. We deliberately prioritized depth over breadth—focusing most heavily on the first three concepts with the highest perceived enterprise impact.

The sprint culminated in a demo-ready story flow and a high-fidelity Figma prototype covering:

• AI-prioritized actions
• AI-written emails
• Aggregate summaries

These were chosen not because they were the only ideas, but because they best demonstrated how AI could show up credibly inside real Groove workflows.

The prototype was used to align leadership internally and to communicate a clear, grounded AI direction externally. Several of the concepts explored during the sprint were later shipped, validating both the prioritization and the approach.

The Mandate

The immediate mandate was explicit and time-bound:

Create something credible we could show an enterprise customer—quickly—that demonstrated clear thinking, prioritization, and intent around Groove’s AI direction.

This was not a request to ship production-ready features or finalize a long-term roadmap. Leadership needed a tangible artifact that could:

• Demonstrate that Groove had a coherent, enterprise-relevant AI strategy
• Show how AI would meaningfully integrate into existing workflows
• Signal prioritization and sequencing, not a scattershot list of ideas
• Restore confidence that AI investment was grounded in real customer value

Crucially, the mandate was as much about communication and alignment as it was about design. The output needed to work externally—with a skeptical enterprise customer—and internally, giving executives and product teams a shared reference point for future decisions.

What the Product Actually Needed

Early in the sprint, it became clear that the risk was not a lack of AI ideas. The risk was ambiguity:

• AI concepts existed across slides, Slack threads, and roadmaps
• Different teams described AI value in inconsistent ways
• There was no single narrative explaining how AI would show up across the Groove experience

What the product actually needed was not more ideation, but structured convergence.

To meet the mandate, the work had to:

• Synthesize existing thinking into a single, legible direction
• Prioritize a small number of AI experiences that mattered most to enterprise customers
• Ground those experiences in real workflows, not abstract capabilities
• Make trade-offs visible—what we were focusing on now vs. later

This reframing shaped every decision in the sprint, from which concepts were explored deeply to which were intentionally left incomplete.

Success Criteria

Because the goal was credibility under pressure—not feature delivery—success was defined by a different set of criteria than a typical product cycle.

We aligned on the sprint being successful if it produced:

1. A customer-ready narrative
   Could we clearly explain how Groove’s AI strategy worked, why it mattered, and where it would show up—without hand-waving or overpromising?
2. Visible prioritization
   Did the work make it obvious which AI bets were most important now, and which were intentionally deferred?
3. Workflow credibility
   Did the concepts feel grounded in real enterprise sales workflows, rather than generic AI capabilities?
4. Internal alignment
   Did leadership, product, design, and engineering leave with a shared understanding of direction and next steps?
5. A concrete artifact
   Was there a prototype compelling enough to anchor executive and customer conversations—something tangible rather than aspirational slides?

The bar was not perfection. The bar was confidence—earned through clarity, structure, and realism.

Meeting these criteria would allow Groove to engage enterprise customers from a position of intent and momentum, rather than defensiveness or ambiguity, and would give the organization a durable foundation for subsequent AI investment and delivery.

Why this approach worked when others wouldn’t

With limited time and fragmented inputs, a traditional discovery or roadmap process would have failed. The problem was not a lack of ideas or effort—it was the absence of a shared frame. We needed a way to compress weeks of alignment into days without sacrificing rigor.

I proposed a focused design sprint as the operating mechanism. The goal was not speed for its own sake, but decision compression: bringing the right people into the same room, working from the same artifacts, and forcing explicit trade-offs.

The sprint was structured around three principles:

1. Converge before you expand
   Rather than generating new AI concepts, we started by surfacing and organizing what already existed—ideas embedded in slide decks, roadmap notes, Slack conversations, and prior discussions. This allowed us to make implicit thinking explicit before adding new surface area.
2. Prioritize credibility over completeness
   Given the timeline, depth mattered more than coverage. We explored seven potential AI feature areas, but intentionally focused the majority of our effort on the first three with the highest enterprise impact. Incomplete sections were not a failure—they were a signal of prioritization.
3. Design the narrative, not just the UI
   From the outset, the sprint was oriented toward producing a story flow that could be communicated externally. The demo experience was designed as a sequence—how AI would show up across actions, communication, and summaries—rather than a collection of disconnected features.

Over roughly two weeks (with a short extension to polish the prototype), we moved from fragmented inputs to a coherent, review-ready prototype that leadership could confidently stand behind.

From Fragmentation to Convergence

(Using design to force clarity)

The Whimsical board functioned as the central working surface throughout the sprint. It was never intended to be customer-facing. Its purpose was alignment.

At the top, we grounded the team in AI interaction patterns and emerging best practices. This helped set shared expectations about what “good” looked like and avoided defaulting to novelty for novelty’s sake.

From there, we mapped the user journey and identified where AI could plausibly augment real Groove workflows—particularly where it intersected with Groove Plays and execution surfaces already familiar to enterprise users.

The core of the board was organized into seven AI feature areas, each owned by a product manager and explored through a consistent lens:

• How-might-we questions
• Known risks and assumptions
• Data dependencies
• Interaction architecture
• Low-fidelity ideation and Crazy 8s

Importantly, the board becomes less complete from left to right. This was intentional. The diminishing fidelity visually reinforced prioritization and helped leadership understand not just what we were proposing, but what we were not committing to yet.

At the bottom of the board, we defined a single demo flow. This story flow became the backbone of the Figma prototype and the primary artifact used in leadership and customer conversations.

The value of the board was not the ideas themselves. It was the shared clarity it created—allowing a diverse group of stakeholders to reason about AI direction from the same frame and make decisions quickly without reverting to opinion or instinct.

Why we focused on three features—and intentionally left others incomplete

The sprint surfaced seven viable AI feature directions, each with a clear owner, hypothesis, and interaction model. The constraint was not ideation quality—it was time, credibility, and enterprise expectations.

We prioritized AI-Prioritized Actions, AI-Written Emails, and Aggregate Summaries for three reasons:

1. Immediate workflow leverage
   These concepts sat directly inside existing Groove workflows. They did not require sellers to learn a new mental model or adopt a parallel surface. This reduced adoption risk and made the value legible to enterprise stakeholders immediately.
2. Data readiness and feasibility
   Each of the selected features could be grounded in data Groove already had reasonable access to (activity signals, email content, engagement metadata). Other concepts required heavier inference, cross-system dependencies, or assumptions about data quality that could not be validated quickly.
3. Credible path from prototype to roadmap
   The goal was not to demo novelty, but to show a believable, phased AI roadmap. These three features had a clear progression from prototype → early delivery → expansion over time, which aligned with how enterprise customers expect AI capabilities to mature.

Features to the right of the board were not deprioritized because they lacked merit. They were left intentionally incomplete because:

• They required longer validation cycles
• They depended on upstream data or platform changes
• Or they represented second-order optimizations that only made sense once foundational AI patterns were established

This trade-off was deliberate. Depth in a smaller number of features created confidence; superficial coverage across all seven would not have.

Why the demo flow mattered more than feature completeness

The Figma prototype was designed as a storytelling device, not a specification.

Its purpose was to answer a single question for enterprise stakeholders:

“Do you have a coherent AI plan that fits naturally into how sellers already work?”

To do that, the demo flow intentionally followed a realistic seller journey—from signal → interpretation → action—rather than jumping between isolated AI features. Each moment in the flow reinforced a few key ideas:

• AI appears in context, not as a destination
• The system assists without overwhelming
• Automation supports judgment rather than replacing it
• Capabilities can expand over time without rethinking the core model

The prototype did not attempt to show every edge case or future state. Instead, it demonstrated conceptual integrity—how multiple AI capabilities could coexist as part of a single, evolving system.

That narrative clarity was what made the roadmap credible, not the fidelity of individual screens.

Restored Enterprise Confidence in Groove’s AI Direction

The most immediate outcome of the sprint was not feature delivery—it was clarity. Prior to this work, Groove’s AI strategy existed across scattered conversations, roadmap slides, and informal plans. The sprint consolidated that thinking into a coherent, review-ready narrative leadership could confidently stand behind.

The resulting prototype was demoed multiple times across leadership and customer-facing teams and became a central artifact in renewal discussions with a flagship enterprise customer. While positioned as a directional roadmap rather than a delivery commitment, it demonstrated that Groove had a thoughtful, phased plan for how AI would meaningfully support sales execution over time.

This shift—from fragmented intent to a unified plan—was pivotal in restoring confidence during renewal conversations.

Enabled a Credible, Phased AI Roadmap

Rather than forcing premature commitments, the work helped leadership align on sequencing:

• Which AI capabilities mattered most to enterprise customers
• What could be delivered near-term vs. later
• How AI should integrate into existing workflows instead of feeling bolted on

The sprint’s prioritization—focusing deeply on AI-prioritized actions, AI-assisted email writing, and aggregate summaries—created a clear path forward. Those areas later shipped incrementally over the course of the year, validating the original prioritization without requiring rework or reframing.

The prototype functioned as a shared reference point long after the sprint concluded, reducing ambiguity and re-litigation as teams moved into delivery.

Demonstrated Design Leadership Under Pressure

Beyond product outcomes, this work materially shifted how design showed up at a leadership level.

At a moment when senior leadership—including the CEO—was unsure where to begin, design provided the operating model. By proposing and leading a structured sprint, I turned an urgent, high-risk situation into a focused, collaborative effort that aligned product, design, engineering, and go-to-market stakeholders.

This was not about producing polished UI. It was about using design to force clarity, compress decision-making, and help the organization move forward with confidence.

Triangulation Table