My Role
UX/UI Design, User Testing, Prototyping
Team
3 Product Designers
Timeline
2 weeks, 2025
Device
iOS
Context
About ornikar
Today, the time spent searching for an order is too long, leading to a significant drop-off.
37%: The percentage of users who performed a search but did not place an order.
9 min: The average session duration to complete an order.
problem
This leads to the following challenge:
How might we make the search experience for restaurants and meals faster and more relevant?
Research
Due to time constraints, we did not conduct primary user research for this project.
Instead, we worked with the data we had and made informed assumptions based on clear usage signals.
We observed that:
37% of users who searched did not complete an order.
The average session duration was 9 minutes — unusually long for a food delivery flow.
From this, we hypothesized that:
Users may be experiencing choice overload.
The search experience might not be surfacing relevant results quickly enough.
While research would have added depth, we chose to move forward with lean design iterations, and aimed to validate our hypotheses through usage data post-release.
IDEATION
inspiration
We explored a large variety of apps in two main areas:
Stats and dashboards from finance, fitness, and e-learning apps
Email summaries and digests
Uber Eats introduced a new “Lists” feature in the US, offering curated "best of" collections created by foodie influencers and local users.
Glovo, Spain’s leading delivery app, features short meal videos on the home page to entice users and encourage faster meal selection.
Queue is a fun and interactive app for tracking movies and shows. It features a Tinder-like swipe system to quickly mark what you've watched or want to see.
Mapstr adds a strong social layer, helping users discover restaurants their friends have visited and recommended.
selected ideas
We explored various ideas to address and selected 2 key solutions based on their perceived low cost and high impact.
Magic Wand
A quick, smart recommendation feature to streamline the search process.
Curated Lits
Restaurant and meal recommendations created by influencers to inspire users and simplify decision-making.
Design
Easy Access to the Magic Wand
The Magic Wand is accessible from two high-intent zones: the header for proactive users, and at the end of strategic pages to support those still undecided.
A Fast, Frictionless Flow
The Magic Wand delivers tailored restaurant suggestions based on just three simple preferences — cutting through choice overload and helping users decide in seconds.
Curated Lists that Inspire Confidence
We showcase influencer-curated lists to guide discovery with social proof, making decisions easier, faster, and more exciting.
User Testing
To validate our hypotheses, we conducted user tests with 5 participants:
Users: French, age 28-34yo, all living in big urban areas
Format: 40-minute sessions
objectives
1.
Ensure users notice the magic wand and find it engaging.
2.
Validate that the magic wand flow is intuitive, efficient, and frictionless.
3.
Confirm that users see the Curated Lists and find them helpful in simplifying their search.
Key insights
5/5
Find the magic wand is really helpful
Super easy and quick to use and it saves me the hassle. – Jean
2/5
Would like to notice sooner the magic wand.
I noticed it a bit late, which is a shame. Would be nice to see it sooner. – Wadii
5/5
Want the ability to adjust Magic Wand settings.
Navigation feels restrictive—I wanted to go back. – Paul
3/5
Didn’t find the list useful.
It feels like I have to spend too much time exploring. – Mathilde
5/5
Would be interested in friend-generated lists.
It would be so cool to have lists from friends, like on Mapstr. – Jean
final iteration
Before
After
Enhancing Magic wand visibility
Introduced a permanent banner below the categories to highlight the Magic Wand feature.
Improving Magic wand navigation
Added a back icon to allow users to easily edit their choices and refine their recommendations.
Before
After
Before
After
Introducing Friends' Lists
Replaced influencer-curated lists with personalised lists created by friends for a more relevant and engaging experience.
reflection
summary
To make the search experience for restaurants and meals faster and more relevant, we designed two key features:
Magic Wand – a fun, engaging, and efficient way to simplify decision-making by providing quick, tailored recommendations.
Friends’ Lists – fostering a sense of community and enhancing the ordering experience through personalised recommendations from friends.
Impact
To monitor the impact of those features, I would have focus on the following KPIs:
Northstar KPI:
Average time-to-order: Average session duration required to complete an order.
Secondary KPIs:
Magic wand
adoption rate: Percentage of users who engage with the Magic Wand feature.
success rate: Percentage of users who use the Magic Wand and proceed to place an order.
drop-off rate: Percentage of users who interact with the Magic Wand but do not complete a purchase.
Curated list
conversion rate: Percentage of meals purchased through a curated list.
next steps
To stay aligned with business goals and cost efficiency, we recommend launching the Magic Wand.
The Friends’ Lists feature requires further design work meaning additional development to deliver a complete experience. Some key missing user stories include:
Following other users
Viewing lists created by friends
Creating, editing, and deleting lists
Given the difference in development effort, we recommend prioritising the Magic Wand, which received highly positive feedback from users.
If I Had More Time…
Run additional user tests on the final iteration: To validate usability and perceived value of the Magic Wand
Explore deeper personalization: Test how preferences (e.g. dietary, cuisine, past orders) could improve the relevance of recommendations.
Measure long-term impact: Set up follow-up metrics to evaluate how the magic wand features influence retention, time-to-order, and decision confidence over time.