REDESIGNING PRODUCT FOR A SAAS
Rethinking How We Discover, Save, and Revisit Restaurants in Sydney
OVERVIEW
This project was part of a 10-week Product innovation program through Prodigi, where we worked in a multidisciplinary team to explore and redefine Truffle’s core experience. My team competed against 4 other teams to pitch our concepts for a chance at implementation.
Using the Double Diamond framework, we focused heavily on the discovery and define phases understanding how people interact with food content today, and where existing tools fall short. Visit Prodigi to explore how they taught Product Design.
Sydney foodies at the centre
I focused on making it effortless for users to capture and revisit restaurants. The goal was to reduce friction in how people save, organise, and return to dining experiences without overthinking it.
Experience transformation
Working across research and design, I helped reshape Truffle’s core experience and introducing a clearer structure with features that actually support how people discover and remember places.
Services
UX/UI research, concept development, user journey mapping, prototyping
Team
Dylan - Product manager
Me - UI/UX designer
S.W - UI/UX designer
Timeline
Jan 2024 - March 2024 (3 months)
CLIENT BRIEF
Redesign Truffle and provide users with two transformed user experiences. You have the option to improve current features, retain one or add two new ones. Provide reasoning for your decisions
●
●
●
SOLUTION? (DANIEL AND NGHI THIS ONES FOR YOU)
●
●
●
HIGHLIGHTS
A dining experience built around recall: making it effortless to discover, capture, and return to restaurants without losing what mattered.
GEO TAGGING RESTAURANTS WITH ‘DINE DROP’
THEMED FOLDER SYSTEM - EAT STREET
PROBLEM
Through research, we found that while people are constantly discovering new restaurants, the way they document them breaks down over time. Lists become cluttered, navigation becomes tedious, and the places that mattered get lost across social media or note taking apps.
Looking at how people currently save restaurants, across Instagram, note taking, and memory, it became clear that documentation was fragmented and unstructured. What starts as inspiration quickly turns into noise, making it harder to return to anything meaningful.
No clear way to narrow: Users struggled to filter by what actually matters—vibe, location, or dietary needs. Discovery felt broad and overwhelming rather than intentional.
Saving without returning: Restaurants were saved with good intent, but rarely revisited. Over time, lists became long, unstructured, and easy to ignore.
Tools that don’t fit behaviour: Existing tools like Instagram and Notes offered flexibility, but lacked structure. Users were forced to create their own systems, which made recall unreliable.
OUR GOAL
Make capturing effortless, navigating intuitive, and returning rewarding...so documentation evolves from a buried list into a living experience users build, revisit, and share.
The prodigi way
The Double Diamond
Start with behaviours, define the breakdown, and design the loop that fixes it.
Project management
Maintain an agile workflow with constant stand ups and retros. This was my first exposure in a structured team environment and a PM to report to
Vibes
Prodigi encouraged teams to step away from Figma and connect through a social leaderboard with point-based challenges such as taking a team selfie during dinner or making tiktok dances.
APPROACH
Unpacking behaviours, defining breakdowns, and mapping the journey to insight.
We began in the empathy phase by unpacking the brief and forming research questions around behaviours, motivations, and the intersection of food and social media. From there, we immersed ourselves in Truffle through usability evaluation, desktop research, and competitor analysis to understand its target users—variety-seeking individuals driven by exploration and self-expression
Discovery & define
Concept Ideation
Product features
Through workshops and synthesis, we translated research into personas, journeys, and four key insights that defined the problem space.
We translated our insights into personas to ground the problem in real behaviours. Diana represents a variety-seeker, someone constantly discovering new places, but struggling to keep track of them in a way that’s actually useful later. Through stakeholder workshops, we mapped and prioritised the core user goals that shape this experience, aligning them with what the product needs to deliver.
Through this, it became clear there was a heavy emphasis on getting the simple things right.
Let me find things quickly and easily
Favourite restaurants, cuisines, food items, dietary requirements, flavours and ingredients
Keep me updated on what matters
Ensure information is live, accurate, and consistent so I can trust what I’m seeing in the moment.
USER JOURNEY
“The restaurant I want is buried somewhere in a list.”
We broke the experience down into stages to see where things actually start to fall apart. What begins as excitement quickly turns into hesitation, as Diana struggles to narrow options and make sense of what she’s seeing. The lack of visual depth and clear prioritisation creates friction, turning what should be a quick decision into a drawn-out process.
With these priorities defined, we conducted a current state analysis to evaluate how well the experience supported these needs and where it was falling short. By interacting with the current mvp product, we identified usability gaps, missing features, and opportunities that framed the rest of our research.
Alongside stakeholder concerns, we grounded the problem in real user behaviour—looking at how people discover, save, and return to restaurants across platforms.
The current experience wasn’t broken, but it fell short where it mattered. Recall was unreliable, information lacked structure, and navigation made simple tasks feel harder than they should be.
Recall breaks under volume
Users save everywhere—screenshots, Notes, private folders—but when it matters, they can’t find what they’re looking for. The restaurant is buried somewhere in the list.
Discovery lacks structure
Exploration is rich but messy. Social platforms prioritise randomness over relevance, forcing users to rely on memory instead of meaningful filtering.
Information lacks trust
Dietary needs, location context, and accuracy aren’t reliable. Users default to safe, repeated choices instead of exploring something new.
To finish off our discovery, we translated these findings into four key insights
Recall via Recording
Users are already recording everywhere screenshots, Notes, private folders but recall breaks under volume and timing.
Camera eats first
Choice is driven by visuals, vibe, colours, portions, ambience but current platforms scatter this across feeds.
Opinions → Recommendations
Users already rate, caption, and share food experiences with close circles, but that signal gets lost.
Exploration is key
Users want to filter by location, dietary suitability, and trusted recommendations. Social platforms lack precise filters, while maps lack social and visual richness, making it harder to explore with confidence.
Chaos ensued when we spent 3 nights affinity diagram into 1 singular polished problem statement which we had to present back to the Prodigi judges during our Mid session showcase
This slipped through us in a way that felt small at the time but compounded quickly. We were split on which insights mattered most, pulling the problem in different directions, and without strong product ownership in that moment, the decision never properly converged. The result was a problem statement that didn’t fully reflect the depth of what we had uncovered. We paid the price for that and were scored accordingly. Below is the version we refined after feedback.
PROBLEM STATEMENT
How might we elevate Truffle’s visual discovery and make returning to documentation a natural, rewarding habit so users can quickly narrow by vibe/location/dietary needs, capture reliably in the moment, and resurface the right place when it matters?
Discovery & define
Concept Ideation
Product features
DESIGN
CRAZY 8s and SCAMPER time!!
We used Crazy 8s and SCAMPER to rapidly generate concepts grounded in our insights, focusing on visual-first browsing, dietary-trust filters, time and location-based recall, and opinion-sharing. To narrow down, we ran a decision matrix similar to MoSCoW to prioritise the directions that best aligned with the problem.
I was particularly interested in how Apple extends experiences beyond the app through things like widgets, App Clips, and Dynamic Island. I wanted to push Truffle in that direction and create something that lives outside the core app, giving users value in the moment. Out of this was done the ‘Shazam Idea’
Adapting the music identification platform of Shazam, using the unique geo-location of each restaurant, this feature could auto-tag and add this place to a certain list to be later re-organised for documentation
Very cool idea, but we had to come to vote against all other ideas
Honestly, the ranking system didn’t feel entirely fair, but at that point I prioritised momentum over perfection. We were already halfway through, and continued indecision was slowing us down. I pushed for a call to be made so we could move forward, even if it meant working with an idea I didn’t like.
THE REAL CONCEPT DESIGN
Now that we've decided on TWO features to transform the truffle app with...
From ideation, two feature directions emerged that directly responded to our key insights around recall, visual discovery, and social sharing. We focused on concepts that could be tested not just for usability, but for how they fit into real behaviours around saving, exploring, and recommending.
We developed two feature prototypes to bring these ideas to life and validate them in context, each targeting a different part of the journey from discovery to recall.
Eat Street focused on exploration and sharing
Dine Drop was built around the idea that users already record everywhere but struggle to return. Using geolocation, it allows users to quickly capture and save restaurants in the moment, turning passive recording into something structured and retrievable.
Dine Drop capture
A single tap uses geolocation to identify the restaurant and save it instantly. This removes the need to manually document or rely on memory.
Eat Street browsing
A folder-based interface that lets users organise, browse, and revisit restaurants visually. Users can sort into “To-dine” and “Dined” states.
Sharing + resurfacing
Users can explore friends’ Eat Streets and share their own, turning saved restaurants into recommendations. This supports users sharing opinions and relying on close circles for recommendations
Conceptualising dine-drop
Discovery
Concept Ideation
Product features
Using think aloud testing, we were able to validate our concepts directly through real user tasks.
This is a low fidelity flow of the dine drop feature being accessed via the operating system’s control centre
We tested our prototypes through 5 think aloud sessions, guiding users through key flows like capturing a restaurant with Dine Drop, organising it into folders, and revisiting it later. This allowed us to observe how users navigated the experience in real time, where they hesitated, and what broke down.
After testing, we used the System Usability Scale (SUS) to measure how intuitive and usable our designs were across participants. Scores ranged from 70 to 85, indicating a generally strong level of usability while highlighting areas that still required refinement.
USABILITY TESTING LO-FI
Users struggled most at moments where the features required interpretation rather than recognition.
Interactions were not immediately legible. Controls that required sliding were treated as tap targets, and toggles or multi-state elements lacked clear behavioural cues. This created hesitation and incorrect actions early in flows.
Meaning was also unclear. Icons carried too much weight without labels, leading users to guess functionality or confuse similar actions such as bookmarking versus marking as visited. Feature naming further compounded this, with users trying to rationalise intent rather than intuitively understanding it.
Finally, the product assumed familiarity it hadn’t earned. With no onboarding or guidance, users were left to learn the system through trial and error.
1.
“I don’t usually find app features on an iPhone control centre”
User triggers Dine Drop from the control centre and immediately questions if it’s even allowed. They hesitate before continuing.
This breaks the user’s mental model and platform conventions, causing a loss of trust before the feature is even understood.
2.
“Which one do you think it’s… the ribbon or the crown?”
While trying to save a restaurant, the user is forced to guess between multiple unclear icons with no labels.
This violates recognition over recall, increasing cognitive load as users must interpret meaning instead of instantly understanding actions.
3.
“A crown usually means premium… like locked or paid”.
The user misreads the priority feature as a paid or restricted function due to the crown icon.
This reflects expectation bias and misaligned signifiers, where strong learned conventions override intended meaning.
4.
“I don’t know what to do now”
After entering the app from Dine Drop, the user stalls with no clear next step or guidance.
This exposes a missing feedback loop and lack of feedforward, leaving the user without direction in the flow.
OUTCOME
A clear signal that the product was usable, but not immediately understandable. Users were able to complete key tasks, reflected in above-average usability scores, but consistently had to pause and interpret interactions, navigation, and meaning before progressing. This gave us a focused direction going into hi-fi—prioritising clarity over complexity by reducing ambiguity in interactions, strengthening visual hierarchy, and making core actions and states immediately legible.
SOLUTION
Final Designs
A product that works with the way people already behave.
Dine Drop was designed as a quick capture system that also is accessible outside the app, using geo-location to identify nearby restaurants and let users save in the moment without breaking their flow. Instead of relying on memory or screenshots, it turns “I’ll come back to this” into something immediate and reliable, making capture feel natural rather than effortful.
Eat Street brings structure to everything that comes after. It reframes saved restaurants into organised, personal collections where users can browse, prioritise, and share through folders and friend networks. What was previously scattered across notes, screenshots, and social media is now a single, returnable thread that users can actually act on.
Users can quickly narrow by vibe, location, and dietary needs without overthinking filters or navigation
Eat Street organises saved places into clear, personal collections that are easy to revisit and share
Dine Drop enables fast, geo-based saving so users don’t rely on memory or scattered screenshots
The experience supports the full loop from discovery to decision, without requiring users to think about how to use it.
Together, these features directly respond to what we saw in testing. Users record everywhere, but recall fails under volume and timing. By making capture seamless and recall structured.
We presented to a panel of product design judges and placed second. It was a strong validation of both the clarity of our ideas and the level of fidelity we achieved.
RETROSPECTIVE
Some key takeaways
Team Collaboration: Working closely with a multidisciplinary team showed me how much smoother the process becomes when everyone is aligned early. We built a steady rhythm of communication, shared thinking, and small iterative decisions that helped the work feel cohesive rather than fragmented. It reminded me how valuable it is when a team genuinely collaborates instead of working in silos.
Personal Growth: I also grew a lot in my own craft. I became much more comfortable experimenting in Figma especially with vector graphics, interaction flows, and building cleaner component structures. It helped me translate ideas more clearly and gave me confidence to take on more of the visual and interaction detail without overthinking it.
UX/UI Strategy: From a strategy perspective, this project reinforced how important it is to let research anchor the design direction. Understanding the real gaps between how people discover restaurants, how they document them, and why they forget them shaped every decision we made. It strengthened my ability to balance the broader UX intent with the smaller interface details that support it.
Thanks for reading!!!