'Good design leadership enables conditions for everyone to do their best work.'
Product Design Leader driving revenue growth, empowering teams and scaling design across multi-product SaaS.
Associated with
Product Design Leader driving revenue growth, empowering teams and scaling design across multi-product SaaS.
Associated with
Projects
Competitor Experience for Mid-Market & Enterprise
From UI Refresh to Enterprise Product Transformation
Driving Product Simplification & Adoption Post-Acquisition
Building & Managing a Chatbot Ecosystem for Career Sites
Designing an End-to-End Marketing Automation Suite
Design leader blending creative intuition with strategic clarity. I lead teams that build user-centered, business-aligned experiences across B2B and B2C domains. Always learning, always designing, driven by curiosity and love for art and storytelling.
Download PDFExperience
Education
Open to senior design leadership roles, advisory work, and conversations about scaling design in SaaS. Always happy to connect.
📆 Case Study · Phenom People
Driving product simplification and adoption post-acquisition. How we closed a critical gap in Phenom's ecosystem and made scheduling interviews feel as easy as sending a calendar invite.
Phenom CRM was feature-rich — capturing qualified leads from career sites, targeting and sending email & SMS campaigns, keeping records of candidate profiles, tracking activities and much more.
On a competitive analysis among all of Phenom's customers, we figured out that one of the prominent gaps in the Phenom ecosystem was that there was no easy way to schedule interviews with candidates. Recruiters needed to go back and forth between candidates and interviewers to finalise a time slot.
I looked at scheduling solutions across the industry to understand prevailing design patterns and the emerging trend of candidate-first scheduling — where the end user chooses when the interview can happen.
Design a scheduling solution within the CRM product so that users don't need to depend on any external tool. Built around three core value propositions:
Common overlapping free slots from interviewers' calendars are surfaced to candidates, who pick what works for them.
If no slot works, candidates can propose an alternative time for the recruiter to review.
Candidates can cancel or reschedule their interviews easily as per their convenience.
Phenom acquired MyAlly — a scheduling platform with AI capabilities for responding to candidate queries via email — adding new capabilities to incorporate into our solution.
The version of our solution post-acquisition didn't fit into our customers' workflows. There was barely any adoption. We conducted research with 5 employees from different customer organisations to investigate.
Once we designed the simplified version, we wanted to make sure it aligned with customer goals. We conducted usability testing with 6 recruiters.
One of the major learnings was the need to build conviction early about how the product would be built. Our assumption was that MyAlly was already an established solution we could simply incorporate. This backfired badly.
Being on the ground during the acquisition into Phenom meant witnessing a lot of chaos, lack of direction and product clarity — fighting through all of that to build something good was invaluable.
After the initial failure, usability testing became extremely crucial to get early validation. Without it, the entire product can become a very expensive affair.
📊 Case Study · Birdeye
From UI refresh to enterprise product transformation. Redesigning Birdeye's Insights module from the ground up to give enterprise customers actionable visibility into their brand performance.
Birdeye had an existing Insights module, but its approach to showing brand sentiment was very limited. The current approach was not enough to offer better value to customers or convince them to choose Birdeye over competitors like Sprinklr and Reputation.
The core question the team faced early: is this a UI revamp or a full product revamp? After exhaustive competitive analysis, the team made a compelling case to do a full product revamp to stand out as a strong alternative for enterprise online reputation management.
"Give our customers a consolidated view of what is not working out for them and what actions they need to take for their location, region and brand."
Imagine an enterprise customer like Dominos — they want to know what people say about their pizza, delivery, ambience, pricing and quality; how they compare against industry averages; and whether all their locations are correctly listed on Google, Zomato, Yext, Apple and more. This is what Insights 2.0 solves.
Sprinklr and Reputation are the primary competitors in the online reputation management space. The goal was to simplify understanding of a customer's brand in the same way these companies do — but with a simpler, more actionable UX.
This project was nearly a ground-up rebuild and required significant justification for why the product needed to exist in a new form. The project underwent at least 4 full end-to-end version reviews, taking stakeholder feedback each time to get to a cleaner, more understandable format. All versions were validated with executive leadership, customer feedback sessions, and incorporated incrementally.
Alongside the design revamp, the team developed new ways of crunching data. The core concept: an 'Experience Score' — a single metric telling an enterprise customer how their overall brand is performing across all locations. It breaks down into three sub-scores:
A numeric value representing sentiment across reviews and surveys, drilled down by location and category.
Whether all locations are listed accurately and keywords are ranking highly as per SEO.
A numeric value derived from average rating, review count and review response rate.
Multiple data visualisations were built: bubble charts, treemaps, and radial views so customers can get different angles on the same data. Each score links to a detailed drill-down by location, category, and time period — nudging users toward concrete action.
Prototype versions were shown to customers active in Birdeye's beta programs. The feedback was largely positive:
This started as a UI revamp. Conversations with customer success managers revealed that a visual update alone wouldn't solve the underlying user need.
Users and decision makers have different expectations. Executive input is critical to understand what buyers need from a product, not just what end-users want.
Customers switching from other tools need the learning curve to be as flat as possible. Feature parity lowers the barrier to adoption and transition.
Developers are partners, not just implementers. Including key dev leads in feedback phases surfaces API, frontend and backend constraints early — a healthy tradeoff is always the goal.
📈 Case Study · Birdeye
Re-imagining the competitor experience for Birdeye — offering better value proposition to mid-market and enterprise customers through smarter setup, richer data visualisation, and deeper integration across Insights, Reviews, and Social.
Birdeye had an existing competitor module, but it had lower engagement and sales compared to other products. The existing product was a single-page layout with dial UI for each competitor's review sentiment and star rating — data rich, but not easily consumable.
On the competitor setup front, the process was tedious and manual, which further impacted adoption. With the recent build of the Insights module and the push of AI across the industry, the opportunity was clear: enhance the competitor module experience within Insights and sell them as a single, higher-value package.
"To re-imagine the competitor experience for Birdeye, offering better value proposition to mid-market and enterprise customers."
Design inspiration was drawn from competitor and competitor ranking products to understand best-in-class patterns for surfacing competitive intelligence.
Competitors fetched automatically via GBP APIs based on customer location. Enabled by default. Brand and location views available.
Competitor data flows into the Insights module — comparative scores, SWOT-style visualisations, strengths & weaknesses, trends over time.
Competitor grid view, leaderboard by source, and head-to-head review comparisons across multiple listing sites.
Competitor Setup: GBP APIs pull competitor location, review sites, ratings, and social links automatically — users can also add competitors manually. Both brand and location views are available, displayed alongside a map.
Insights: Competitor data flows into a Benchmarking section showing AI-generated summaries, Birdeye score comparisons, reputation and sentiment scores side-by-side, a scatter-plot bubble chart for visual ranking, and bar charts for sentiment scores by category across all competitors.
Strengths & Weaknesses: A SWOT-style view compares the customer's brand vs each competitor — showing areas each side is winning or losing, with AI-generated narrative summaries.
Reviews module: Customers compare their reviews with competitors across multiple listing sites — via a competitor grid (satisfaction vs market presence), a leaderboard by source (Google, Healthgrades, Doctors.com etc.), and head-to-head comparison views. A similar pattern is followed in the Social module for cross-channel follower and growth comparisons.
This started as a UI revamp project. Conversations with customer success managers revealed that a visual update alone wouldn't address the underlying user need or drive adoption.
The decision maker buying the product has different expectations than the end user. Internal executive input — from those who are front-facing with buyers — is essential to shape the right product direction.
Developers are partners. Including key dev leads in feedback phases surfaces API, frontend and backend constraints early — before they become blockers. A healthy tradeoff is always the goal.
💬 Case Study · Phenom People
Two connected products: a candidate-facing chatbot that converts career site visitors into quality leads, and the management platform built to let customers configure and own it themselves — without ever filing a support ticket.
Phenom's core product was personalised career sites. But on those sites, web traffic was frequently not converting into leads — contact details and names weren't being captured. Only around 15% of traffic was converting as leads.
The solution needed to feel approachable, provide genuine value to the candidate, and capture leads in the process — at a time when chatbots were trending and many startups were entering this space.
Answer common candidate questions about benefits, culture, and leadership — eliminating the need for call centres and saving recruiter time.
Nudge candidates to share minimal info — job category, title, location, experience — to surface relevant openings immediately.
Capture name, email, and phone number as a natural outcome of helping candidates — a win-win for candidate and recruiter.
Offer related jobs by title and location when primary results are exhausted, ensuring every visit is maximally useful.
When no suitable jobs exist today, candidates can set up email alerts — keeping them in the pipeline for future roles.
The chatbot was continuously tweaked based on data and insights. The screens below represent the most recent design iteration.
The bot initially sat passively on the career site, resulting in just 0.02% usage. Adding a proactive pop-up pushed this to up to 7% of web traffic. Adding sounds gave a marginal further boost — but required careful tuning.
Once the bot had sound and auto-pop behaviour on every page, candidates found it annoying as they scanned through jobs. Conditional triggering rules had to be introduced to limit the behaviour.
Early versions asked for personal details before showing jobs. Flipping the flow to show jobs first and collect leads only after expressing interest — despite expectations of a dip — actually increased the number of leads and produced higher-quality ones.
After building the chatbot, there was no way for customers to manage it themselves. Every change — styling, interactions, FAQs — required the implementation or development team. Without an interface, adding FAQs alone could take more than 3 weeks, as talent marketers needed time to compile question-answer pairs.
This increased onboarding time, left candidates unserviced, and wasted both customer and internal productive time. The solution: a self-service management platform.
The FAQ section lets customers create and manage question-answer pairs the bot should know. Key design considerations: easy navigation between questions, a recommended answer character limit (soft, not enforced), support for training phrases, and confidence score visibility per FAQ.
When a candidate asks a question, the system tries to match it to an FAQ with 80%+ confidence. Below that threshold, the bot gives a fallback answer and logs the query. Knowledge Gaps shows aggregated patterns with severity and frequency. Unanswered Questions shows individual queries alongside the full candidate conversation — allowing recruiters to reply directly or link the query to an existing FAQ.
Customers can configure the bot's avatar, name, brand colours, and interactions — all with a live preview panel so changes are seen instantly before publishing.
As a growing B2B company, the goal was to satisfy customers while keeping the team focused on roadmap priorities. Building self-service tools gives customers the flexibility to make changes on their own timeline — reducing onboarding time, eliminating support tickets, and resulting in happier customers who feel genuinely in control.
📣 Case Study · Phenom People
Designing an end-to-end marketing automation suite that empowers talent marketers to enrich contacts, map relationships, build campaigns, and orchestrate workflows — all without engineering support.
Talent marketers at enterprise companies deal with fragmented tools — separate systems for managing contacts, understanding relationships, building campaign content, and orchestrating outreach. This project brought all four capabilities into a single, unified marketing automation suite within Phenom's platform.
The suite is designed around the journey of a talent marketer: first understand who your audience is (Contact Enrichment), then map how they connect (Knowledge Graph), then build what you'll send (Template Builder), and finally decide when and how it reaches them (Workflow Builder).
Automatically enrich candidate and contact profiles with data from multiple sources to keep records complete and actionable.
Visualise and explore the relationships between candidates, roles, locations, and skills to uncover hidden connections in your talent pool.
A drag-and-drop email and SMS template builder with out-of-the-box templates to help marketers launch campaigns faster.
A visual, node-based workflow engine to automate multi-step outreach sequences based on triggers, filters, and candidate behaviour.
Talent marketers often work with incomplete contact records — missing phone numbers, outdated job titles, or no location data. Sending campaigns to stale or incomplete data wastes budget and reduces deliverability. The goal was to build a way for teams to automatically enrich profiles using integrated third-party data providers and internal signals.
The enrichment panel allows marketers to preview what data will be filled, choose which fields to update, and configure auto-enrichment rules so new contacts are always processed on entry. A confidence score alongside each enriched field lets users decide whether to accept or override the suggestion.
Recruiters intuitively know that a strong referral or a former colleague's recommendation carries more weight than a cold application. But this relationship data lives in people's heads — not in the ATS. Without a way to surface connections between contacts, roles, and skills, teams miss warm paths into talent pools.
The Knowledge Graph enables recruiters to click on any contact and see their entire network — people they've worked with, skills they share with other candidates, roles they're connected to, and referral paths. The visualisation is interactive: nodes can be expanded, filtered, and used to seed new audience segments for campaigns.
Talent marketers frequently needed to create email campaigns for job alerts, re-engagement, and diversity outreach — but every template required developer HTML or a design ticket. This created a 5–10 day lag between campaign ideation and launch. The Template Builder was designed to give full creative control back to the marketer.
To help teams launch quickly, a library of pre-built templates was included for common use cases — job alerts, diversity outreach, re-engagement, and event invites.
Campaign sequences — send an email, wait 3 days, send an SMS if no reply — were being set up manually via support tickets or spreadsheets. Without a visual tool to build and edit these sequences, talent marketers could only run single-touch campaigns, leaving significant engagement potential untapped.
The Workflow Builder uses a node-based canvas where talent marketers can drag in triggers, actions, wait steps, and conditions. Every node is configurable — change templates, timing, audience filters — without touching code. Workflows can be saved as reusable templates, activated immediately, or scheduled in advance.
Each product in this suite is standalone — but their real power comes from how they connect. A contact enriched in Product 1 feeds the graph in Product 2, seeds segments used in Product 3, and is targeted in a workflow in Product 4. Designing the handoffs between products was as important as designing each product individually.
Users didn't want to learn four separate interfaces. The entire suite had to feel like one surface. Consistent patterns for navigation, settings, and data selection across all four products significantly reduced the learning curve during onboarding.
Out-of-box templates and pre-configured workflow starters drove adoption far faster than open-ended canvases. Getting users to their first successful campaign in under 10 minutes was the key activation metric that shaped design decisions across the suite.