Back

Designing Contextual Support Across Investor Subscriptions

Help Desk interface embedded within the subscription workflow
Role
Lead Product Designer
Company
Vega
Timeline
2024
Platform
Desktop
What I did
  • Conducted stakeholder and investor research
  • Designed v1 and v2 interfaces and user flows
  • Mapped the investor journey and defined UX pain points
  • Partnered with engineering to ship and QA new DS components
  • Validated impact with analytics and user testing
Collaborators
  • Wei Han (Engineering)

Overview

Completing fund subscription questionnaires can be overwhelming for investors due to dense legal and financial terminology. The Help Desk was designed to embed contextual, human support directly into the subscription workflow — enabling users to ask precise, section-specific questions without leaving the form.

After the first prototype, research and feedback revealed usability gaps. The second iteration introduced richer text editing, expandable input, and searchable conversation threads to make the experience more discoverable, readable, and scalable.

The Problem

The subscription questionnaire spans multiple stages, each loaded with compliance-heavy language. Without guidance, investors often paused progress or sought help externally, disrupting flow and increasing time-to-completion.

Goal

Create an embedded, contextual help system that:

  • Allows investors to ask questions without leaving the workflow
  • Gives Investor Services clear section-level context
  • Reduces support volume and manual email management
  • Builds investor confidence in form completion

Outcome

  • Measurable lift in subscription completion
  • Significant reduction in email-based support volume
  • Faster resolution through contextual, in-platform conversations

Research

Stakeholder interviews

Interviewed 4 members of the Investor Services team. This revealed several recurring friction points in how investors sought help during the subscription process:

Investors had to leave the form to ask for clarification via email or phone, breaking their flow and increasing turnaround time.

Investor Services teams lacked context around which specific question or section the user was referring to, often needing to open the original document manually to locate the issue.

Email queries overwhelmed the team's inboxes, disrupting their workflow and making it difficult to track conversations, ensure accountability, or maintain an auditable record of support interactions.

User observation

Due to data privacy (PII) constraints, direct observation of live investor subscriptions was not permitted. Instead, we conducted simulated walkthroughs with five internal users representing real investor personas and reviewed anonymized feedback from past support tickets.

"I wish I could ask this question right here instead of switching to email."

Users paused for an average of 90 seconds on sections containing legal or tax terms.

4 of 5 users left the platform to send clarification emails.

Progress was frequently halted while waiting for replies.

Journey map

Existing Journey Map

Key insight

Investors experienced cognitive fatigue and loss of context; Investor Services faced inefficiency and tracking issues. Both sides needed a single contextual channel within the platform.

Design Constraints

Regulatory & compliance

  • All conversations had to stay inside altOS to satisfy SEC/FCA audit rules.
  • No external chat tools or attachments allowed.
  • Every message required timestamped storage for compliance logging.

Data privacy & PII restrictions

  • Live investor observation was prohibited; research used anonymized simulations.
  • All messages encrypted and stored per fund entity.

Organizational structure

  • Each fund had its own dedicated Investor Services team, with unique operating hours, workflows, and internal response templates.
  • The Help Desk needed to route messages to the correct team based on fund ID and jurisdiction.
  • Cross-fund visibility was restricted, limiting opportunities for shared knowledge or central dashboards.

Limited support hours

  • Investor Services did not operate 24/7, with coverage aligned to business hours in their respective time zones.
  • The Help Desk had to be designed as an asynchronous communication system, setting realistic expectations through status indicators like "Response expected within business hours."

Solution

Contextual help triggers

Inline help triggers were placed in each section. Clicking opened a Support Drawer pre-tagged with the relevant section and question ID. Investors could ask questions instantly, while Investor Services received precise context.

Clicking 'Help me with this question' opens the support drawer pre-tagged with the relevant section.

Support panel (v1)

The first version introduced:

  • A right-aligned conversation drawer
  • "Mark as Resolved" and "Reopen" conversation states
  • Persistent thread history tied to each document
Support panel showing conversation thread
V1 support panel with conversation drawer, resolve/reopen states, and persistent thread history.

Integrated notifications

Given that Investor Services teams operated during limited hours and across regions, clear notification design became essential.

In-app notifications — Investors receive a real-time badge update and banner when a response is added to their thread. Notifications display concise previews and timestamped activity logs allow investors to review the full context at any time.

Email summaries — If a user is offline, a digest email is triggered summarizing unread replies. Emails include deep links that open directly within the relevant fund's Help Desk panel.

Notification system — in-app and email

Usability Testing

After the initial rollout, we conducted targeted usability testing to evaluate how well investors and agents could compose, read, and retrieve messages within the subscription workflow.

Investors found the narrow input box uncomfortable for writing long or structured questions.

The Investor Services team highlighted that users typically use rich text formatting when sending in queries, so it would be helpful to have that as well.

Testers struggled to read long replies within the constrained panel width, leading to scanning errors.

Testers lacked an efficient way to search or sort through previous messages.

Iteration (v2)

Wider support panel

We increased the panel width and restructured message hierarchy for better text legibility.

Wider support panel comparison
Before and after: increasing panel width and restructuring message hierarchy.
Improved support panel design
Refined message layout with improved text legibility and visual hierarchy.

Rich text editor

Investors needed to clearly articulate complex financial questions, often involving multi-part scenarios or references to specific sections. A plain text box proved limiting.

We replaced the plain input with a markdown-enabled editor supporting bold, italic, bullet points, and inline references — with a formatting toolbar for investors unfamiliar with markdown syntax.

Rich text editor with formatting toolbar for structured investor queries.

Expandable compose area

The compose area became resizable, letting users expand it when composing complex or multi-step queries.

Resizable compose area for complex multi-step queries.

Search & sorting

Introduced global search and filter tools allowing both investors and Investor Services to retrieve past conversations by section, status, or date.

Global search and filter tools for retrieving past conversations by section, status, or date.

Impact

For investors
  • +25% subscription completion rate
  • Questions answered without leaving the workflow
  • 90% positive feedback from user testing sessions
For Investor Services
  • -65% support ticket volume
  • +38% faster resolution time
  • Full context on every query without manual lookup

What I Learned

Context eliminates friction. Embedding help at the point of confusion — rather than routing users to a separate channel — preserved focus and reduced resolution time.

Async systems need clear expectations. Without 24/7 support, status indicators and email summaries became critical to preventing user anxiety about response times.

Iteration validated by testing. The shift from v1 to v2 was entirely driven by usability findings. Rich text, wider panels, and search weren't assumptions — they were observed needs.