Are you looking for smarter insights delivered straight to your inbox? Subscribe to our weekly newsletters and receive essential updates tailored for enterprise AI, data, and security leaders.
The Rise and Fall of Intuit Assist
In the competitive landscape of generative AI that erupted after the launch of ChatGPT, Intuit’s CEO issued a clear directive: to roll out the company’s most significant AI-driven product by September 2023. In response, the $200 billion company, known for QuickBooks, TurboTax, and Mailchimp, introduced Intuit Assist. This initial attempt featured a chat-style assistant integrated into its applications, aimed at showcasing Intuit’s innovative edge. However, the outcome was disappointing.
“When you take a beautifully designed user interface and simply add human-like chat to the side, it doesn’t necessarily enhance the experience,” stated Alex Balazs, Intuit’s Chief Technology Officer, in an interview with VentureBeat.
Challenges in AI Scaling
The limitations of AI scaling, including power constraints, increasing token costs, and inference delays, are reshaping the enterprise AI landscape. To explore how leading teams are addressing these challenges, join our exclusive salon where we will discuss strategies for:
– Transforming energy into a strategic advantage
– Architecting efficient inference for substantial throughput gains
– Unlocking competitive ROI through sustainable AI systems
Secure your spot to stay ahead: https://bit.ly/4mwGngO
The Trough of Disillusionment
The unsuccessful launch left Intuit in what Dave Talach, Senior Vice President of the QuickBooks team, referred to as the “trough of disillusionment.” The chatbot occupied valuable screen real estate and caused confusion among users. “There was a blinking cursor. We inadvertently placed a cognitive burden on users, prompting questions like, ‘What can it do? Can I trust it?’” Talach reflected. The pressure mounted as he prepared to present to Intuit’s Board of Directors, explaining the missteps and lessons learned.
What ensued was not a minor adjustment, but a rigorous nine-month effort to “burn the boats” and reinvent the product development approach of the 40-year-old company. This narrative reveals how Intuit developed a practical playbook for enterprise AI that other leaders can emulate.
A New Direction
The shift away from the chatbot began with a keen observation of customer behavior. Talach recalls a pivotal moment when his team noticed QuickBooks users manually transcribing invoices, using a split-screen setup with an email on one side and QuickBooks on the other. This insight led to a new mission: rather than attempting to create new user behaviors through chat, the focus shifted to eliminating “manual toil” within existing workflows.
Recognizing this momentum, CTO Alex Balazs and Marianna Tessel, General Manager of the business group, took decisive action. “We need to make a declaration together,” Balazs remembers Tessel saying. The only way forward was a complete commitment to an AI-native future. “It’s burning the boats; we’re moving forward solely with the AI approach.”
Implementing Change
To realize this vision, management reassigned a key technology leader, Clarence Huang, from the core tech team to the heart of the QuickBooks business. His mission was to foster a “builder-centric mindset” focused on rapid, customer-oriented prototyping. Embracing this new model required dismantling the previous one. To empower smaller, more agile teams, the company made the tough decision to eliminate layers of middle management, resulting in the layoff of 1,800 employees in 2024 from roles misaligned with new priorities. However, the company pledged to hire approximately 1,800 new employees with expertise in engineering, product management, and other customer-facing roles.
Intuit’s transformation hinged on a new operational model built upon three fundamental changes: empowering its workforce, re-engineering its processes, and developing a technology engine designed for speed. Successful execution of the pivot necessitated placing the right people in the right structure and enabling them to work in entirely new ways.
Talent Acquisition and Team Structure
To bolster its core AI team, Intuit undertook aggressive hiring, expanding from just 30 employees in 2017 to several hundred today, with a significant acceleration in recruitment over the past two years. This included attracting top-tier AI leaders from major companies such as Uber, Twitter, and Bytedance.
The foundation of the new model consisted of small, empowered, cross-functional teams. These groups, often comprising members from up to ten different units—including data science, research, product, design, and engineering—focused exclusively on delivering specific user experiences. To facilitate this, managers prioritized ruthlessly, eliminating any tasks that did not rank among the top three priorities. “That ruthless prioritization was absolutely crucial,” Huang emphasized.
Empowered Working Environment
Within these teams, traditional job descriptions were blurred, creating what Huang describes as a “smearing” of roles. Every team member was expected to engage with customers. Huang maintained his own list of 30 customer contacts whom he called regularly to gather feedback and insights.
Through these transformative efforts, Intuit is paving the way for a more efficient and effective approach to enterprise AI, setting a benchmark for others in the industry.