Service Design: Intuit Client Referral Ecosystem

A strategic service design initiative to reduce customer churn and redefine the professional referral experience through trust-building features.

Role: Lead UX Researcher & Service Designer

Timeline: June 2025 – Present

Tools: Figma, Miro, HeyMarvin, Notebook LM, Great Question, Qualtrics, R, Python

Methods: Mixed-Methods Research, Service Blueprinting, Root Cause Analysis

Team: Product Design & Research, Data Science, Market Research, Sales, Marketing, Customer Success

Executive Summary

The Challenge:The accounting industry is facing a massive shift, with 44% of single-owner CPA firms set to retire in the next five years. Intuit aimed to capture this transition by building a referral ecosystem to help these professionals offload clients. However, internal data revealed a significant conversion gap: matched clients were dropping off at an alarming rate before engagement began

The Approach I led a mixed-methods research initiative to uncover the root cause of this churn. Using surveys (n=468) and in-depth interviews (n=32), I analyzed the complex relationship between Tax Professionals and their clients.

The Outcome The research revealed that the drop-off wasn’t a UI issue—it was a service failure. I delivered a foundational “Ideal State” Service Blueprint that redefined the 7-stage customer journey. This strategy shifted the product roadmap from “lead generation” to “relationship management,” prioritizing features that build trust and facilitate warm handoffs.

Research Strategy (Mixed-Methods)

To deconstruct this two-sided marketplace (B2B2C), I moved beyond surface-level usability to understand behavioral drivers.

  1. Qualitative Empathy: I performed deep-dive interviews with 11 Tax Pros and 9 Taxpayers to uncover the emotional weight of “breaking up” with a client.
  2. Quantitative Validation (N=468): I conducted large-scale surveys with 316 Tax Pros and 152 Taxpayers to quantify behavioral trends and identifying pain points.
  3. Competitor Benchmarking: analyzed 7 key competitors to benchmark incentive structures and matching algorithms against market standards.
  4. Data Analysis: Utilized R and Python to analyze the survey data, identifying significant correlations between specific “trust signals” and the likelihood of a successful referral.

Customer Segmentation (The “Who”)

Previous internal research indicated that our user base was more complex than originally thought. To address this, we moved beyond simple demographics and categorized users based on their specific offloading behaviors and relationship expectations.

The Supply Side: Tax Pros (Categorized by Intent) We identified three distinct archetypes based on their motivation for offloading:

  • Reactive Pros (The “Lifeline” Seekers): They are overwhelmed by capacity issues and have no clear handoff plan. They need immediate relief.
  • Adaptive Pros (The Testers): They are testing offloading with a flexible plan but haven’t fully committed.
  • Proactive Pros (The Strategists): They are intentionally restructuring their business and require structured, high-control handoffs.

The Demand Side: Taxpayers (Categorized by Expectations) We segmented clients based on what they need from a professional:

  • Low-Expectation Clients: Transactional users who simply need an “Executor” to get the job done.
  • Critical Thinkers: Users with intermediate tax situations who seek a “Strategic Advisor”.
  • Reliant Clients: Users who view their pro as a “True Partner” and require a warm, high-touch transition to feel safe.

Key Takeaway: The platform treated everyone like a “Low-Expectation” transactional user. However, our most valuable users—the Proactive Pros and Reliant Clients—required a relationship-based model that the current system did not support.

Key Insights

My analysis uncovered three critical ‘human’ friction points that were not fully addressed by the current digital workflow.

1. The “Trust Gap”: Know the Pro, Earn the Go

Tax professionals are protective of their reputation. Data showed that 42% of pros lacked confidence in external referral sources, preferring to default to their own network (41%).

  • The Insight: Every referral puts their reputation on the line. They won’t use a platform that feels like a “black box”.
  • The Fix: The platform needed transparent “Trust Signals”—visible vetting criteria and peer testimonials—to replace blind trust.
2. Hard News Needs a Human Voice

The digital product relied on automated emails to notify clients they were being offloaded. This eroded trust immediately.

  • The Insight: 64% of pros prefer to use a phone call to initiate a “breakup” to ensure a soft landing. Clients perceived automated emails as impersonal or even “phishing”.
  • The Fix: We proposed a “Warm Handoff” protocol that integrated human connection points (calls/introductions) before the digital transfer occurred.
3. The “Decision Paralysis” Trap

When clients were offloaded without a specific recommendation, they froze.

  • The Insight: “Offloading without a referral equals paralysis.”. However, 89.6% of clients followed a recommendation when it came from a trusted source.
  • The Fix: Shift from a “directory” model (too many choices) to a “Concierge” model—providing a single, vouched recommendation to reduce cognitive load.

The Solution (Service Blueprint)

Synthesizing these insights, I created the “Ideal State” Service Blueprint.

This artifact mapped the end-to-end lifecycle across 7 unique stages, from “Decide to Offload” to “Evaluate New Service”. It highlighted exactly where the “Current Experience” failed to meet “Ideal Expectations,” specifically identifying:

  • Deal Breakers: Lack of credentials (48%) or slow communication (50%).
  • Risk Transfer: The need for a liability waiver to protect the referring pro’s reputation.

Business Impact

This research fundamentally shifted the product strategy at Intuit.

  1. Roadmap Redirection: The “Ideal State” blueprint became the framework for the new product roadmap, prioritizing “Trust” features over “Speed” features.
  2. Validated “Concierge” Features: The data proved that high-touch handoffs result in nearly 90% acceptance rates, validating the investment in concierge-style support.
  3. Reputation Protection: New features were prioritized to address the “supply side” fear, including liability waivers and transparent vetting badges.

Scroll to Top