Chapter 04 / 04Flagship — the culmination of the journey
Enterprise product leadership
Flagship Case Study · Enterprise Product Design

Title Hub

The operational platform that reorganized a company around the file

Title Hub is not a feature or a dashboard. It is a large operational platform that gradually replaced the disconnected systems an entire title-operations company relied on every day — the workspace where daily work happens, where operational data becomes trustworthy, and where managers run the operation.

Enterprise PlatformProduct StrategyOperational SystemsInformation ArchitectureDesign System
Workflow boardManager viewFile mailboxWorkflow dataDocument intelligence & data extraction

The Operating Model

From a fragmented stack to one operating system

Title Hub did not replace a single tool. Before it, the operation ran on a scattered stack — six broad areas of work, each split across different applications, spreadsheets and human memory. The design task was not to consolidate tools; it was to change the unit around which work is organized.

Before · fragmented operating stack
CommunicationOutlook, internal chats
Task executionMonday, personal follow-ups
DocumentsFolders, attachments, repositories
Operational dataInternal databases, external systems
Status & ownershipBoards, spreadsheets, human memory
Management visibilityManual reports, meetings, follow-ups
After · Title Hub
File Workspacethe operational file at the center
EmailsTasksDocumentsQuestionsStatusDataActivityOwnershipValidationReporting

Title Hub did not simply consolidate tools. It changed the unit around which work was organized — from applications and departments to the operational file.

Project Snapshot

My role
Lead Product Designer — I led product definition, requirements and the design end-to-end, in partnership with a product manager.
Collaboration
Product manager, department managers, operational employees and engineering teams — I drove the product and design decisions.
Research
Interviews with department managers and operational employees, workflow analysis and iterative validation.
Adoption
Live and in daily use — hundreds of users across multiple departments and offices (New Jersey, New York and more), up to manager level.
Complexity
Role-based access, cross-department workflows, external-system integrations, documents, communication and operational data.
Design scope
Research, workflow mapping, architecture, IA, wireframes, prototypes, testing, UI and design-system evolution.

The Challenge

An organizational problem before a UX problem

Employees were not struggling because Outlook was bad. They were struggling because work itself was fragmented — the operational knowledge that kept files moving lived inside people's heads and across a dozen systems, not inside a product. The clearest way to feel this is to follow a single, ordinary task.

One task, many tools — “review a file and move it to the next stage”
1
Monday
Open the task board and identify the task
2
File system
Find the file number the task refers to
3
Outlook
Search the email thread for the latest context
4
Folders
Locate the supporting documents
5
External system
Open an external system to pull operational data
6
External system
Compare the values by hand
7
Chat
Ask a colleague a question to unblock a detail
8
Monday
Go back and update the status
9
Outlook
Report to the manager or hand the file off

Nine steps, six tools, and repeated jumps backward — for one file, many times a day, across every department.

Time lost to searching

Most effort went into finding and re-finding context, not into the judgment only a person could add.

Context lost between systems

Each hop dropped a little context; nothing held the whole story of a file in one place.

Accountability lost between handoffs

With work spread across tools, ownership blurred — steps were duplicated or quietly dropped.

Framed correctly, this was an organizational transformation problem. The interface would only ever be the visible outcome of getting the operating model right.

The Organization

The product started from how the company works — not from screens

Before designing a single screen, the work was to understand the operation itself: its departments, the roles inside them, how a file moves between them, who owns each step, where decisions are made and where information enters and leaves. The product could only ever be as good as this understanding.

Departments & teams

Operations are organized by jurisdiction and product line — each department runs its own boards and queues, yet they share files, data and clients. The product had to respect this structure while making the whole legible.

National Title

Commercial
  • Commercial readers & queues
  • National orders across states
  • Multi-site coordination

NY Title

Jurisdiction team
  • State-specific requirements
  • Local proofing & questions
  • County-level records

NJ Title

Jurisdiction team
  • State-specific requirements
  • Local proofing & questions
  • County-level records

Proforma & Support

Cross-cutting
  • Proforma & multi-site tasks
  • Shared operational data
  • Client-facing coordination
Department relationships

Departments and roles don't work in isolation — a file passes between them, and information, data and decisions flow across the boundaries. Mapping these relationships showed why point solutions couldn't work, and why the operation needed a shared operating layer rather than better-integrated tools.

completed file delivered back to the clientroles & permissions govern allorderassignreviewapprovereports statusdata pull & validationClient / Ordersubmits & receivesTeam Managerassigns · approvesReaderexecutes workProofing / QAreviewsDeliverysent to clientExternal Systemsoperational dataAdministratorroles & permissions
A file's journey across departments

A single file crosses hands and departments through a defined lifecycle. Above the line is the role that owns each stage; below it, the information and artifacts the stage depends on. Mapping this became the backbone the product was later organized around.

Intake
Order received
  • Order details
  • Client email
Team Manager
Assigned
  • Task
  • Ownership
Reader · Search
Research / Search
  • External data
  • Documents
Reader
Review
  • Reading
  • Status
Reader ↔ Manager
Questions / Exceptions
  • Questions
  • Emails
Reader
Data validation
  • Checklist
  • External data
Manager
Approval
  • Decision
  • Approval
Reader · Client
Completion
  • Sent to client
  • History
Organization & responsibility map

The same file is touched by several actors with different responsibilities, visibility and moments of action. Making this explicit is what later justified role-based views and a permissions layer rather than one screen for everyone.

ActorResponsible forSeesCommunicates withActs at stage
Operational employee
Reader
Reading, data entry, resolving questionsOwn files & queuesManager, colleagues, client (via manager)Research → Validation
Team managerAssignment, workload, proofing, approvalAll team files & statusEmployees, other departments, clientAssign · Review · Approve
Manager-contributorExecutes and oversees at onceOwn files + team overviewEmployees, managersAcross all stages
AdministratorRoles, permissions, configurationSystem-wide (governance)ManagersCross-cutting
External systemsSource of operational dataPlatform, via syncIntake · Validation

Research

The product was built only after understanding the work

This was product discovery, not UX research in isolation. I met with department managers and operational employees, observed real workflows, mapped how work moved across departments, documented bottlenecks and ownership, identified repeated manual work, studied communication patterns, and learned which information each role actually needed.

Layer 1 · The questions I went in with
With department managers

How the operation is run

  • How is work assigned?
  • How do you know where a file is stuck?
  • How do you understand each employee's workload?
  • Which events require escalation?
  • Where do handoffs fail?
  • What do you need to see but currently can't?
With operational employees

How the work actually happens

  • What starts your day?
  • How do you know what to do next?
  • Which systems do you open for a single task?
  • What information do you repeatedly search for?
  • What do you track outside the official systems?
  • Where do errors or delays happen?
Layer 2 · What observation surfaced

Recurring patterns across interviews and observation — reconstructed here, not quoted verbatim.

“Monday is kept open to know what to do — but the information itself lives in the emails.”

Observed pattern

“Understanding one file means searching in several places.”

Observed pattern

“A delay is usually discovered only after someone asks about it.”

Observed pattern

“The status doesn't always say what is actually blocking the file.”

Observed pattern

User groups — same file, different jobs

Operational Employee

Goal: complete assigned work efficiently. Needs action clarity — what to do next, the information at hand, and status without noise or switching.

Team Manager

Goal: keep the operation moving. Needs visibility across files and people, workload balance, and the ability to spot and clear bottlenecks.

Manager-Contributor

Goal: execute and oversee at once. Needs to do personal work and manage a team without losing context between the two.

Layer 3 · Synthesis — from behavior to product opportunity

Each observed behavior was traced to the operational problem underneath it, and then to the product opportunity it pointed at. This is the bridge from research to strategy.

Observed behavior

Searching across tools to finish one task

Operational problem

Context is fragmented across systems

Product opportunity

Bring communication, data and actions onto the file

Observed behavior

Managers ask employees for status updates

Operational problem

Progress is invisible until someone asks

Product opportunity

Build role-based operational visibility

Observed behavior

Cross-checking values by hand against external systems

Operational problem

Validation is slow and error-prone

Product opportunity

Surface comparisons; preserve human approval

Observed behavior

Knowledge kept in inboxes, spreadsheets and memory

Operational problem

The operation depends on individuals, not the product

Product opportunity

Encode workflow, ownership and history into the file

From Research to Product Strategy

The bridge most case studies skip

Findings on their own don't build a product. The real work is the translation: each observed problem becomes a product decision, justified by a reason, and delivered as a platform capability. This decision framework is where the product strategy is actually defined — before a single screen exists.

Research findingProduct decisionWhyPlatform capability
Employees switch between Outlook and Monday to work one fileKeep all context on the fileWork must not scatter across toolsFile Workspace
Managers don't know who owns work or where it's stuckMake ownership & status always visibleYou can't manage what you can't seeAssignments & operational visibility
Employees compare values across systems by handSurface differences; keep human judgmentAutomation without oversight risks accuracyValidation Center
Operational knowledge lives in people, not the productEncode workflow, history & roles into the fileThe operation shouldn't depend on individualsPlatform services

The strategic choice — how should work be organized?

The findings pointed at a fork. Three approaches were genuinely on the table, and the whole platform depended on which one we committed to.

Option A

Improve each tool

Make Outlook, Monday and the rest work better and integrate them. Lowest disruption — but keeps work fragmented and the mental model unchanged.

SelectedOption B

A file-centered platform

Reorganize everything around the operational file. Highest ambition — but it fixes the root cause: the unit work is organized around.

Option C

A task-centered hub

Put tasks at the center, as Monday does. Familiar — but tasks are only one facet of a file, so context would still scatter.

We selected Option B. Tasks, emails and documents are all just facets of a file — organizing around any single facet would have left the others fragmented. Only the file could hold the whole.

Why the File?

The one decision the entire product rests on

Everything else in Title Hub follows from a single choice about the unit of work. Rather than asking people to move between systems, we moved the systems around the file: every capability the operation needed became an attribute of the file itself.

Before · scattered tools
Outlook
Monday
Chat
Documents
External data
Reports
Status
Questions
Assignments
FILEthe unit of work
After · attributes of the file
Workspace
Timeline
Validation
Tasks
Comments
Notifications
Ownership
Search
History
Data

Instead of asking employees to move between systems, we moved the systems around the file.

Should emails remain a separate inbox?
No — every communication belongs to the file. A message about a file that lives in a personal inbox is context the next person can't see.
Trade-off Less inbox flexibility, far more shared context and history.
Should tasks be a global task board?
No — tasks belong to the operational context of the file. A task detached from its file just becomes another thing to reconcile.
Result Less searching; a task always carries what it's about.

The Architecture

Understand the platform before seeing an interface

With the file as the organizing unit, the platform was structured as three layers: core navigation that orients any role, file-workspace modules that carry a file's context, and platform services that operate across every file. This architecture is what let dozens of capabilities feel like one product — and what made role-based access and permissions coherent rather than bolted on.

Three-layer architecture — Layer 1 core navigation · Layer 2 file-workspace modules · Layer 3 platform services
Three-layer architecture — Layer 1 core navigation · Layer 2 file-workspace modules · Layer 3 platform services
Layer 1

Core navigation

Dashboard/Workspace, File Workspace, Manager View and Search — the orientation layer every role starts from.

Layer 2

File-workspace modules

Emails, Tasks, Chat, Docs, Data, Status and History — the context that travels with each file.

Layer 3

Platform services

Notifications, Sync & Integrations, Permissions & Roles and Activity Tracking — services that run across all files.

What's shared vs what's domain-specific

Separating the cross-cutting foundation from the product domains was a deliberate architectural decision: it let each domain evolve independently while permissions, search, notifications and activity stayed consistent everywhere.

Shared foundation
Core navigationSearchPermissions & rolesNotificationsActivity trackingSync & integrations
Product domains
Operational WorkspaceData Intelligence & ValidationOperational Visibility

Which surfaces the most important truth about Title Hub: it is not one product. On the shared foundation sit three distinct products, each serving a different job.

Domain 01

Operational Workspace

Problem

Finishing one file meant touching five tools.

Users

Operational readers & manager-contributors.

Goal

Do the whole job from one place, on the file.

Capabilities

Board, queues, tasks, communication, status, search.

Domain 02

Data Intelligence & Validation

Problem

Operational data was cross-checked by hand.

Users

Readers validating; managers approving.

Goal

Make data trustworthy without removing judgment.

Capabilities

Extraction, comparison, exceptions, approval, sync.

Domain 03

Operational Visibility

Problem

Managers couldn't see the operation.

Users

Team managers & administrators.

Goal

See the whole; assign, rebalance, unblock.

Capabilities

Dashboards, workload, roles & permissions, analytics.

Everything Connects to the File

The whole system in one picture

If there is one diagram to remember, it is this. Every capability the operation needs — communication, documents and data, workflow and governance — is an attribute of the file. And because it all lives on one model, both products are just different views onto it: the internal HUB executes the work, the portal exposes a controlled read of it.

FILECommunicationEmailsQuestionsStatusNotificationsDocuments & DataDocumentsOCRHistoryReportsWorkflow & GovernanceWorkflowValidationAssignmentsPermissionsInternal HUBexecutionPortalcontrolled visibility

One file, every capability, two products — this is why the whole platform holds together.

Extending the Platform Beyond Internal Users

The Portal — an extension of the ecosystem, not a separate product

While the HUB served internal operational teams, I also designed a lightweight portal for external departments, management and business stakeholders. Instead of exposing the full operational workspace, the portal offers controlled, read-oriented visibility into file progress, communication and outstanding issues — built on the same underlying data model with role-based permissions. Same data, a different audience, far fewer actions.

Internal HUBfull operational workspace — execution
shared data · role-based permissions
Portalread-oriented · controlled visibility
ManagementExternal departmentsBusiness stakeholders
Portal landing — a separate, lightweight entry point for people who don't work inside the full system; it opens on search, not on a workload.
Portal landing — a separate, lightweight entry point for people who don't work inside the full system; it opens on search, not on a workload.
Advanced search
Advanced search — external users discover files by criteria, rather than entering through a personal “My Files” queue
Search results
Search results — reach any file quickly, with just enough status to orient
The file overview — the portal's most important screen: emails, questions, On-Hold history, general information and workflow status about a file, all in one read-only place
The file overview — the portal's most important screen: emails, questions, On-Hold history, general information and workflow status about a file, all in one read-only place
Communication, read-only — stakeholders can follow the email history on a file without entering the operational workspace or touching the work
Communication, read-only — stakeholders can follow the email history on a file without entering the operational workspace or touching the work
Product principle

Same data model, different audience. The portal proves the platform's architecture was right: because everything already lived on the file with role-based permissions, exposing a controlled, read-oriented view to outside stakeholders was an extension — not a second product to build from scratch.

Three Product Domains

This is not one product — it is three product stories

Read separately, each domain is a complete product with its own users, problems and design story. What follows is the deeper story of each — how work gets done, how data becomes trustworthy, and how the operation is managed.

01
Product Domain

Operational Workspace

Where daily work happens

The workspace is where readers execute work. The design problem was to give each person a single place to see what they own, work a file end-to-end, and communicate — without leaving the file. The board organizes every file by priority, state, questions, proofing and flow status; the role-based home surfaces exactly the queues a person is responsible for.

The workflow board — every file with priority, state, questions, proofing status, ownership, flow status and outstanding emails in one operational view
The workflow board — every file with priority, state, questions, proofing status, ownership, flow status and outstanding emails in one operational view
Mailbox on the file
The mailbox on the file — inbox, drafts, preview and attachments; email moved out of Outlook and into the operational context
Workflow data on the file
Workflow Data — status, a live checklist, time tracking and notes: the whole state of the file in one panel
Decision: tasks belong to files, not to departments.
Ownership follows the work, not the org chart. A task attached to a department drifts from the file it is about; a task on the file always carries its context.
Trade-off Less rigid departmental control, far less searching.
Trade-off accepted

Putting communication, tasks and status on the file adds density to one object. We accepted that and paid for it with strong grouping and progressive disclosure, because the alternative — work scattered across tools — was far more expensive.

Domain 01

Operational Workspace — full story

Research, journey, information architecture, flows and the decisions behind the workspace.

Explore the domain
02
Product Domain

Data Intelligence & Validation

Making operational data trustworthy

Operational data is only useful if it can be trusted. This domain turns manual cross-checking into a review workflow: the system surfaces and compares data on the file, flags what doesn't match or is unknown, and keeps a human in the loop for the decision. The checklist on the file makes each validation step — and its state — explicit.

01

Document

The source document is attached to the file and previewed in context.

02

Surface

Relevant data is pulled onto the file from internal and external systems.

03

Compare

The system cross-checks values and flags matches, mismatches and unknowns.

04

Resolve

Exceptions and conflicts are worked on the file, with history preserved.

05

Approve

The reviewer approves; the file's data becomes the trusted record.

Save to Select — the document viewer (with OCR and annotation) beside a field-selection panel: highlight a value in the source and save it straight into the file's structured data
Save to Select — the document viewer (with OCR and annotation) beside a field-selection panel: highlight a value in the source and save it straight into the file's structured data
Validation decision model

Rather than automate blindly, the interface encodes a simple decision model so a reviewer always knows what a value needs.

Data stateWhat the system doesWhat the person does
MatchMarks confirmed, low-noiseNothing — trust and move on
MismatchFlags the conflict & shows both sourcesReviews evidence and resolves
UnknownMarks as missing, requests the valueSources and enters, or asks
ExceptionRoutes to the right ownerDecides and records the reason
Product principle

Automation earns speed; human oversight earns trust. The system surfaces and compares — the person decides. In a process where accuracy is critical, removing the human is a worse risk than a moment of review.

Decision: automate the comparison, never the approval.
The system surfaces and compares; the person decides. Full automation would trade a slow risk for a worse, silent one in a process where accuracy is critical.
Trade-off Validation is not instant — a moment of review is the price of trust.
Domain 02

Data Intelligence & Validation — full story

The data lifecycle, validation workflow, decision matrices and the design of trust.

Explore the domain
03
Product Domain

Operational Visibility & Management

How managers see and steer the operation

Managers do a fundamentally different job than readers, so they need a fundamentally different view of the same data. This domain is about seeing the whole: workload across people, where files are stuck, what needs attention, and the health of the operation — with the controls to assign, rebalance and unblock.

Manager view — all files, assignments and flow status across the team, grouped by stage (In Progress, Awaiting Search, On Hold), with priority and attachments visible
Manager view — all files, assignments and flow status across the team, grouped by stage (In Progress, Awaiting Search, On Hold), with priority and attachments visible
Employee needs

Action clarity

  • What do I own right now?
  • What is the next step on this file?
  • Everything I need, on the file
  • Low noise, no switching
Manager needs

Operational visibility

  • Where is every file, across the team?
  • Who is overloaded; what is stuck?
  • What needs attention or approval?
  • Assign, rebalance and unblock
Product principle

The interface changes with responsibility. Rather than one view compromised for everyone, the same underlying data is framed for two different jobs — action for the employee, oversight for the manager.

Decision: give managers a separate product, not the employee view with more columns.
Manager needs differ in kind, not degree. The same data is reframed for oversight — grouped by stage, led by ownership and flow status.
Result Managers read the operation at a glance instead of row by row.
Domain 03

Operational Visibility & Management — full story

Dashboards, roles & permissions, workload and the design of oversight.

Explore the domain

The Design System

How did dozens of modules feel like one product?

That is the only question this section answers. The design system was less about color and type than about a small set of operational patterns — each defined once, then reused across every domain. The payoff was threefold: consistency (a table behaves the same everywhere), scale (a new module is assembled from known parts, not designed from scratch), and maintainability (fix a pattern once and every module inherits it).

PatternWhy it exists — consistency · scale · maintainability
Data tablesEvery domain is table-heavy; one table anatomy (sort, flag, group, density) means a reader learns it once and reads any module.
Filters & searchThe same find-and-narrow model across boards, files and reports — no relearning per screen.
Status systemA single vocabulary of states and colors so status means the same thing platform-wide.
Cards & formsShared anatomy and input behavior so new capabilities ship without redesigning the basics.
Permissions & rolesOne access model applied everywhere — the reason role-based views stayed coherent as the platform grew.
Tokens & spacingColour, type and spacing tokens change in one place and propagate to every module.
Brand green
Ink navy
Text
Surface tint
Attention
Alert
Product principle

Consistency is not decoration — it is what makes scale affordable. Shared patterns turn ‘another module’ into an assembly of known parts, so the platform can keep growing without fragmenting again.

Where It Stands

From fragmented tools to one operational home — in production

Title Hub shipped and is in daily use. It replaced the scattered mix of tools with a single operational home, now relied on by hundreds of people across multiple departments and offices — including New Jersey and New York — up to manager level. It was adopted well across the teams it was built for, and it keeps growing as new operational needs are added.

Product principle

The measure of an operational platform is not a launch — it is whether people choose it for the real work, every day. Title Hub became that default.

The through-line

What this chapter taught me

Designing the operational platform where enterprise teams execute their whole workday — department research, enterprise workflows, permissions and role-based interfaces, automations, communication, task management, data synchronization, external integrations, manager views and a client-portal extension — all architected around the file as the unit of work.

The through-line See the full four-chapter journey