The context
Coaching was meant to be the engine of sales readiness at Bigtincan — the place where reps practiced skills, received feedback, and demonstrated readiness for real customer conversations.
It sat at the intersection of enablement, performance, and leadership visibility. When Coaching worked, it drove adoption and confidence. When it didn’t, it created confusion, hesitation, and heavy reliance on Customer Success to fill the gaps.
How it started
When Bigtincan acquired Brainshark in 2021, we inherited a Coaching experience built on early-2000s enterprise patterns. The workflow relied on rigid, state-driven paths that predated modern SaaS expectations.
Over time, incremental enhancements were layered on — new features, shortcuts, and patches — but the underlying structure never evolved. By the time this work began, Coaching had become fragile, opaque, and increasingly difficult to reason about.
The future pressure
By late 2025, as Vector Capital took Bigtincan private and the Showpad integration began, Coaching needed to function as a coherent, platform-level system.
While a full backend rebuild wasn’t possible, this moment created the opportunity for targeted backend changes in service of clearer structure — allowing us to move beyond surface-level UX fixes and finally define a more predictable lifecycle.
The tension
The inherited Brainshark architecture hard-coded two primary paths — submission and feedback — and couldn’t be re-architected end-to-end, even though limited, targeted backend changes became possible later. We still needed to deliver incremental value, support new capabilities, and signal platform maturity.
The challenge wasn’t identifying what was broken.
It was improving clarity and usability inside a system that had never been designed as a lifecycle.
The problem
Opening an activity dropped users straight into either the submission flow or the feedback flow, with no expectation, no orientation, and no sense of progression. Recording lived on a separate page, disconnected from the rest of the experience, while core context — instructions, criteria, and participant selection — was buried inside flows rather than surfaced upfront.
Iteration technically existed, but it wasn’t visible as takes were hidden, there was no feedback on non-submitted attempts. Reviewers were thrust into a single rep’s submission with little context, managers relied on a separate product (Scorecards) for limited visibility, and CS teams spent hours explaining how the system worked.
Fundamental issue
Fragmented pages, unpredictable paths, and no sense of “what comes next.”
Understanding the system
Working closely with CS and Engineering, it became clear that these weren’t isolated UX issues — they were symptoms of a system without an organizing model.
The workflow behaved as rigid legacy paths with scattered context and unclear transitions. Without a predictable sequence, reps couldn’t tell what came next, reviewers had to reconstruct intent manually, and managers lacked a reliable picture of progress.
A pattern began to emerge: Coaching needed to be understandable as a sequence — expectations, attempts, and review — not a set of disconnected screens.
The constraints
Legacy backend that couldn’t be fully re-architected, limiting workflow states to incremental additions
High-stakes submission model with no true drafts
Multi-role complexity across participants, reviewers, and managers
Phased engineering capacity requiring incremental progress
What I owned
I led design direction from 2022–2025, from stabilizing the inherited Brainshark experience to shaping the later unified Coaching designs.
Strategy & Direction
Defined clearer structure across the Coaching workflow with Product and CS
Sequenced improvements to deliver value within backend constraints
Established consistent role-aware patterns across participant, reviewer, and manager experiences
Ran regular CS reviews that shaped key decisions, including removing a “quick submit” shortcut that would have created dead ends
Design & Delivery
Designed a unified Activity Page to reduce fragmented entry points
Reworked Instructions, Practice, Review, and Feedback into a more connected flow
Improved dashboards to better reflect real user behavior
Partnered closely with Engineering to ensure feasibility at each phase
How it came together
Rather than arriving fully formed, the structure emerged through the work.
Phase 1 and Phase 2 focused on stabilizing and clarifying the legacy flows — surfacing context, reducing friction, introducing early practice patterns, and adding Roleplay AI — even though the underlying structure couldn’t yet change.
Instead of pursuing a risky backend rebuild, we expressed structure through design: reshaping surfaces, states, and transitions so the workflow felt more predictable within legacy constraints.
As friction decreased, a clearer sequence began to surface:
Understand expectations → attempt → review → iterate
By Phase 3, we were able to define that sequence explicitly. It became the organizing principle for what needed to be visible upfront, what belonged together, and how different roles should experience the same activity.
With targeted backend changes, Coaching shifted from rigid, state-driven flows into a system people could actually reason about.
A unified flow with predictable steps for participants and reviewers.
Unified Activity Page
A single, stable surface where the coaching lifecycle became clear
Instructions, criteria, takes, submissions, reviewers, & feedback lived in one predictable place, giving all roles a shared point of reference. Reps could understand what was expected next without guessing or jumping between flows
What this enabled:
Clear navigation, fewer accidental submissions, and context that stayed intact across the activity
One surface for the lifecycle
instructions, takes, reviews, and feedback finally lived together.
Your submissions
iteration became visible and compare-able.
Role clarity in action
reviewers instantly saw who was stuck, who submitted, and who needed nudging.
Dashboards that matched real behavior
Different roles, different needs — one coherent framework
Participants
To-dos, pending reviews, and completed activities were consolidated into one dashboard, making next steps explicit
Reviewers
Pending reviews, criteria, takes, and context were aligned in a single, consistent pattern, reducing context-hunting and review errors
Managers
Managers could monitor progress and readiness directly within Coaching, instead of relying on Scorecards and fragmented metadata
What this enabled:
Clearer ownership, reduced manual tracking, and more reliable visibility across roles
Participant Dashboard
to-dos, pending reviews, and completed work merged into one predictable flow.
Reviewer Dashboard
video, transcript, criteria, and scoring aligned into a single reviewing surface.
Manager Dashboard — cross-team readiness, bottlenecks, and progress finally became visible
Recording & takes redesigned around iteration
Practice became intentional, not fragile
Recording was redesigned as a focused, full-page experience with visible takes, save/delete controls, and instant AI feedback where technically feasible. Iteration was supported explicitly before submission, rather than hidden inside the flow
What this enabled:
Higher-quality takes, lower cognitive load, and more consistent review context
Practice became safe — reps iterated without fear of submitting the wrong take
Choice before saving — a lightweight checkpoint that reduced anxiety and improved quality
Transcript, video, and AI feedback combined into one learning loop
so reps improved faster with clearer context
Submission switching
Iteration made visible
Reps and reviewers could compare submissions side-by-side, making progress visible and review context explicit instead of hidden
What this enabled:
Clearer progression, safer iteration, and more consistent reviews
Iteration became visible and comparable — a major shift from the opaque, “submit once” legacy model
Phase 1 — Stabilization & Practice Validation
2022–2023
Cleaner flows, transcript editing within feedback, reduced UI drift, and early practice support within legacy limits.
Outcome
CS reported reps practicing roughly 2–3× more before submitting.
Phase 2 — RoleplayAI
2023–2024
Scenario-based practice with AI scoring and inline feedback, strengthening Coaching’s differentiation.
Outcome
Positioned Coaching as an AI-forward differentiator during merger planning.
Phase 3 — Unified Lifecycle
2025
Unified Activity Page, clearer role-aware dashboards, and more explicit workflow states.
Status at transition
Designs were built and validated with CS; rollout timing shifted due to platform integration priorities.
Delivered impact — Phases 1 & 2
Behavioral changes (validated by the CS team)
Improved clarity and reliability within legacy flows
Reduced UI inconsistency and accidental errors
Stronger engagement with practice through Roleplay AI
Clear validation that users wanted safer iteration and a more predictable flow
Strategic outcomes
Created the foundation for AI-supported coaching
Positioned Coaching as a differentiator during acquisition planning
Validated impact — Phase 3
Unified Activity Page resolved fragmented context and unpredictable navigation in CS validation
Role-aware dashboards closed long-standing visibility gaps within Coaching
Iteration became explicit through visible takes and submission switching
Targeted backend changes supported clearer lifecycle states
Internal user feedback
“I finally understand where I am in coaching — it’s not a mystery anymore.” — Sales Rep
“This cut my review time in half.” — Enterprise CS Manager
“My reps finally feel safe to practice.” — Sales Director
Strategic
Established a durable lifecycle model that positioned Coaching as platform-ready
Created a coherent foundation for AI-supported coaching
Produced reusable lifecycle patterns that informed adjacent platform work
Where it stalled
The design worked — the system wasn’t ready.
True practice-first cycles required backend rebuilding
Legacy architecture limited how far fragmentation could be removed
Multi-role complexity stretched existing system boundaries
Acquisition timelines delayed Phase 3 rollout
The real lesson
What I'd do differently
This work reinforced how much clarity depends on structure. Many usability issues weren’t about missing features, but about missing sequence.
In hindsight, earlier alignment with Engineering on backend evolution would have allowed the system to support the workflow more directly. Even so, establishing a clearer structure created a foundation that extended beyond Coaching into other parts of the platform.
