AI INteraction labs

Whatfix Quickread

Context not just search - Whatfix Quickread helps end users reduce information search time and cognitive load by Enabling contextual information within users workflow

Timeline
2025 - 2 quarters
Focus
AI agents
Impact
379
Enterprise IDs enabled
19%
M-o-m growth by mid Q4
Evaluated through 2 quarters in GA
My Role
Advocated and owned end-to-end design.
Collaborated with PM, 1 researcher, 2 engineers, 2 customer design partners from insurance domain.

QuickRead launched as a beta in 2023 with select customers and reached GA in 2024. The goal was to help users who are are non-technical professionals across Sales, Support, Operations, HR, Finance, and similar teams fetch inforamtion from connected organizational repositories without leaving their workflow.

By early 2025, customer feedback revealed growing issues around accuracy, lack of context, overly long responses, and poor in-flow usability. While adoption initially spiked post launch through sales driven rollout, adoption declined to just 2% month-on-month, with several enterprise accounts eventually disabling the experience altogether.

Early Quickread experience

On the left - Looker data of Quickread through 2023 to 2025 , on the right - customer testimonials via NPS and productboard tickets

Challenges

Through 30+ conversations with internal CSMs and end users, and by analysing behavioural patterns, productboard tickets, and recurring support themes, a deeper issue was uncovered , in how users accessed information within their workflows. Two core problems emerged.

⛓️‍💥
Information not in the flow of work
Users already had access to documentation, knowledge bases, and search tools; but none of them worked contextually within the flow of work. Finding the right answer required leaving the task, framing structured queries, and navigating dense one-shot responses
🔔
Experience was transactional
The experience functioned like a one-time Q&A flow; ask, get an answer, start over. It lacked exploration, transparency, and context, making it difficult for users to refine inputs,

Jobs to be done

🎯
Help me understand without leaving my task.
When I’m stuck mid-task, I want a quick, relevant explanation from my own tools, so I can keep working without context switching.
Current workaround
Open documentation Ask teammates Search internal repositories

Why current solutions failed
AI tools required context re-creation and felt disconnected from the context of the task
❤️
Help me trust the answer or information.
When an AI gives me an answer, I want to know where it came from and whether it applies to my situation, and have a same output each time so I can confidently act on it.
Trust signals users look for:
Source references Recency of information Ability to drill deeper or ask follow-ups
Why current solutions failed
Generic answers and different information for similar questions led to users abandoning the feature after initial usage
✅️
Don’t make me figure out how to ask.
When I’m in my workflow, I want relevant information without having to craft or refine my search query.
Current workaround
Rephrase the same question multiple times Copy internal terms from documentation, tickets, or teammates Ask colleagues instead
Why current solutions failed
Users assume internal company language, acronyms, or team-specific terms will be understood When results feel off, users blame themselves for “asking it wrong” and disengage

Strategy

Beyond solving gaps, I realized the challenge wasn’t just about improving search ; it was about designing trust, context, and confidence into AI assisted decision making.

⏰
Beyond summarization
If QuickRead remained a summarisation widget, it would compete with search. It needed to evolve into a contextual, agentic layer to help redefine how enterprise users consume knowledge. The opportunity wasn’t just faster answers.
 It was about reducing the time between understanding context and taking confident action.

Form follows intent

A user’s primary goal within any platform is to complete their workflow efficiently. The experience, interaction model, and functionality therefore needed to align with user intent at the right moment within the flow of work.

Key design interventions

Form - Evolving Quickread into a flexible system of forms tailored to varying user needs, organisational contexts.and application code base

🤗
Built for users reality
Quickread works across legacy and modern applications, accessible directly within the users workflow, via in-app floating modals, HUB widget, or as Chrome extension, delivering enterprise knowledge without disruption.

Intent defined not intrusive - focusing on in-context, on-demand assistance, especially in frequently used or high-friction areas, supported by conversational and suggestive follow-ups.

♾️
Omnipresent, not intrusive
Information remained accessible within the flow of work, allowing users to retrieve what they needed at the right moment without interrupting their primary task or overwhelming the interface.

Introducing the all new Quickread

📌
Designed to be read not decoded
Breaking complex information into clean, glanceable pieces that reduce friction, lower mental fatigue and builds trust.

Validation

While most platforms focused on building more powerful conversational AI, our hypothesis was different: the challenge wasn’t access to information, but access to the right information in context and within the flow of work.

We validated this with a design partner in the insurance domain, where underwriters relied on fragmented SharePoint documents and Excel sheets mapped to IBC codes; forcing them to leave their workflow, search manually, and interpret dense information.

🏆
Small success big impact
At a design partner workshop, we achieved an early success by surfacing contextual information directly within underwriting workflows and replacing query-heavy search behavior with in-flow, one-click access to relevant information. This immediately improved decision-making and reduced a process that previously took 2–3 business days to under a day, with higher confidence and fewer errors.

Beyond Quickread - Whatfix AI

Quickread evolved into a core component of Whatfix AI, forming the conversational layer for premium users of Whatfix Seek — the agentic AI platform.</div>

Follow on

Impact

Following the evolution of Quickread into a contextual, AI tool, enablement reached an all-time high of 379 enterprise tenants across 233 customers. Adoption steadily increased to 19% month-over-month by mid-Q4, reflecting stronger engagement within the flow of work.

🙌
Customer voice
There were clear wins and important learnings. Embedding Quickread directly within the flow of work , with a seamless pathway to Whatfix’s agentic AI capabilities, significantly improved user confidence in both the AI experience and the broader product ecosystem. Customers validated this shift, with one noting, “I can definitely 100% validate that right now in all our conversations and engagement - this is spot on.”

Takeaways

🧭
Simplify integrations
Reducing setup fatigue required seamless repository integrations, easier onboarding, and lower governance overhead for enterprise teams.
🚀
Design for AI readiness
Admin controls and system intelligence needed to evolve to improve contextual understanding, response accuracy, and scalability.
✨
AI should feel invisible
Enterprise AI succeeds not by feeling more powerful, but by integrating seamlessly into existing workflows and systems.