Case study · No. 06 · FoodTech · marketplace · 2023
Home Chef: neighbourhood cooking, delivered still warm
A platform for home-cooked meals made by the cooks of your neighbourhood: order today's tagine, follow it to your door, and, kitchen side, make a living from what you already cook.
- Web UX/UI
- Marketplace
- Design system
- Role
- Product designer, from research to prototype
- Duration
- 8 weeks
- Team
- Concept project, solo
- Platform
- Web, desktop
- Tools
Figma 
Miro 
Google Meet 
Google Docs- Deliverables
- Benchmark, Proto-personas, Journey map, User flows, Wireframes, Design system, 25 screens, Prototype
The project in numbers
What was delivered, counted in the Figma file. A fictional product: no usage results.
25
screens
16 for the customer, 7 for the cook, 2 for the team
3
roles served
The one who eats, the one who cooks, the team that visits kitchens
337
prototype links
Four linked journeys, from the order to Friday's payout
01 · The project
A kitchen that already feeds the street
In Morocco many women cook every day for a whole family, often for the neighbours on Fridays, without earning from it: in 2023 only 19 % of women were in the labour force, against 69 % of men.
Across the street, busy families and offices want a real lunch, not fast food. Home-made dishes already sell, but on Instagram and WhatsApp: no payment, no tracking, no checked hygiene.
Home Chef links the two: dishes cooked the same day, in small batches, delivered still warm; a kitchen visited before opening; and a space where the cook sets her prices and days and sees her earnings, paid every Friday.
Eight weeks, from the kitchen to the door
01 · Weeks 1–2
Discover
Four ways to eat “home-made”, from delivered restaurants to the cook on Instagram.
- Benchmark
- Desk research
02 · Weeks 2–3
Define
The customer and the cook, the experience map of an ordered lunch.
- Proto-personas
- Journey map
- HMW
03 · Weeks 3–4
Ideate
Prioritisation, the ordering flow, wireframes.
- Impact × effort
- User flow
- Wireframes
04 · Weeks 4–7
Prototype
Design system, 25 screens, the hand-drawn logo, a linked prototype.
- Design system
- High-fidelity UI
- Figma prototype
05 · Weeks 7–8
Check
A review against UX laws, sold-out dishes and delays, a test plan.
- Heuristic review
- Test plan
02 · Discovery
A skill without a market
12 – 2 pm
the lunch window
It all plays out in two hours: orders by 10, delivery at noon.
Hypothesis
≈ 70 DH
the price of a portion
Between a snack bar and a restaurant.
Hypothesis
< 40 min
from kitchen to door
To arrive still warm.
Hypothesis
Four ways to eat “home-made”
Named by family: restaurant delivery, the cook on social media, the caterer.
| Criterion | Home Chef · the projectThe project | Restaurant deliveryPlatform | Cook on social mediaIndirect | CatererOn order |
|---|---|---|---|---|
| Who cooks, and where | YesThe cook, her neighbourhood | PartlyThe restaurant | Yes | Partly |
| Today's dishes, in small batches | YesPortions left | NoFixed menu | PartlyDaily posts | No |
| Payment and delivery built in | Yes | Yes | NoCash, delivery to arrange | Partly |
| Checked hygiene | YesKitchen visited | Partly | No | Partly |
| The cook's income in view | YesThe week's earnings | Not applicable | No | Not applicable |
The table scrolls sideways.
A desk review, verdicts simplified by family.
Takeaway
Platforms know delivery, social media knows emotion. Home Chef keeps the cook's face and adds what is missing: payment, tracking and trust.
03 · Definition
The one who orders, the one who cooks
Proto-personas: hypotheses to validate in interviews; their quotes are illustrative.
P1 · the customer
Yasmine, 36
Works in Maarif, two children, orders lunch three times a week
- Goal
- A home-made dish for the family, delivered warm and on time
- What gets in the way
- Not knowing who cooks or how, cold deliveries
« I want to know who cooked what my children eat. »
P2 · the cook
Fatima Zahra, 52
In Derb Ghallef, cooks for twenty every Friday, without pay
- Goal
- Get paid for what she already does, at her pace
- What gets in the way
- Can't set a price, fears paperwork and a complicated app
« I already cook for the street: just tell me how much and when. »
Yasmine's lunch, before
Yasmine
01Get hungry
02Search
03Order
04Pay
05Wait
06Eat
Does
Thinks
Feels
Opportunity
The map scrolls sideways.
The problem to solve
Busy families need a home-made meal delivered warm at noon, because fast food does not replace neighbourhood cooking. Today, these dishes are found on social media, with no payment, no tracking and no checked hygiene.
The questions the design answers
Q1
How might we make an unknown cook trustworthy?
Why this one · Yasmine feeds her children with it.
Q2
How might we sell today's dishes in small batches?
Why this one · A home kitchen has no fixed menu.
Q3
How might we deliver still warm, on time?
Why this one · The low point of the experience map.
Q4
How might we pay the cook without paperwork?
Why this one · What holds Fatima Zahra back.
04 · Ideation
What the first version keeps
Impact ↑
Quick wins · Impact high · Effort low
Portions left and ready timeIn V1
Q2
Orders by 10 amIn V1
Q2, Q3
Earnings simulatorIn V1
To convince cooks
Big bets · Impact high · Effort high
Tracking and handover codeIn V1
Q3
Kitchen visit before openingIn V1
Q1
Cook space, Friday earningsIn V1
Q4
Nice to have · Impact low · Effort low
Ramadan menusLater
In V2
Later · Impact low · Effort high
Meal subscriptionLater
Effort →
Ordering lunch
Start
Home
What, where, when
Screen
Dishes near me
Ready by 12:30, under 70 DH
Screen
Dish
The cook, portions left
Screen
Basket
Delivery slot, notes
Decision
Signed in?
No: phone number and SMS code, then back to the basket
Screen
Payment
System
The cook accepts
Before 10 am
Screen
Tracking
Cooking, on the way, handover code
End
Delivered, rate the meal
Low-fidelity wireframes
01 · Home
02 · Dish
03 · Tracking
04 · Cook space
05 · Design
Cream, coral and a kitchen hand
The hand-drawn logo, Fraunces for headings, Montserrat for the interface, and a coral for action on a cream background.
Colours
Cream
#FBF6F0
Background
Surface
#FFFFFF
Cards
Coral
#F35151
Main action
Strong coral
#D63A3A
Accent text
Soft coral
#FFF3F1
Banners
Ink
#1F1A17
Text
Secondary text
#4A3F38
Captions
Olive
#4F7A3A
Success, ready
Saffron
#8A5A0C
Almost gone
Danger
#B42318
Sold out, problem
Type
Display
Fraunces SemiBold 72
Home cooking from your neighbourhood
Heading 1
Fraunces SemiBold 48
What’s cooking this week
Dish
Montserrat SemiBold 16
Kefta tagine with eggs
Body
Montserrat Regular 16
Cooked the same day, in small batches.
Price
Montserrat Bold 20
65 DH
25 components
- Logo
- Logo mark
- Button
- Badge
- Input
- Checkbox
- Radio
- Toggle
- Chip
- Avatar
- Rating
- Tab
- Quantity
- Navbar
- Footer
- Search
- Dish card
- Cook card
- Category
- Review
- Order tracker
- Basket line
- Cook sidebar
- Cook top bar
- Order ticket
Home: say who cooks
First iteration
Final version
What changes
The headline says who cooks and where: the women of your neighbourhood.
UX law · Aesthetic-Usability EffectKept
What, where, when: the search replaces “Explore”, area and time prefilled.
UX law · Tesler’s LawKept
Live proof: how many are cooking now, who, and for when.
UX law · Serial Position EffectKept
The customer's journey
Today's dishes
Dish (Q2)
The cook (Q1)
Basket
Checkout
Tracking (Q3)
Rate the meal
Kitchen and team side
Q4 and Q1: becoming a cook, running the day and the earnings, and the visit that opens the kitchen.
Become a cook
Application
Today
Add a dish
Friday's earnings
Applications, team side
Kitchen visit
06 · Validation
UX laws at work
- Kept
Scarcity Effect
Rare feels valuable, until the pressure feels fake.
“4 portions left” is true: the cook makes small batches.
Dish
- Kept
Tesler’s Law
Complexity never disappears, someone carries it: let it be you.
Area and time by default: search in one click.
Search
- Kept
Doherty Threshold
Under 400 ms, the system feels like a conversation.
Each step of the order shows as it happens.
Tracking
- Kept
Aesthetic-Usability Effect
What looks good is forgiven more: beauty buys patience.
Real photos and named cooks: trust goes through a face.
Everywhere
- Kept
Peak-End Rule
People judge the peak and the end, not the average.
The cook's note arrives with the dish.
Delivery
- Kept
Anchoring Effect
The first number sets the scale for every next one.
The simulator starts from a median income, never the best case.
Become a cook
Expected results
What a test and a launch should show. Nothing has been measured.
| Expected result | Indicator | Indicative target |
|---|---|---|
| An order placed with trust | Testers who name the kitchen visit as a reason to order | 3 out of 5 in testing |
| Dishes still warm | Deliveries within 40 minutes of leaving the kitchen | 9 out of 10 (hypothesis) |
| Cooks who stay | Cooks still active after three months | Above 70 % (hypothesis) |
The test that would validate the platform
ParticipantsFive customers who order lunch at least once a week, and three women who already cook for others, in 40-minute sessions.
01
TaskOrder two tagines for 12:45, no olives.
Success criterionOrdered, note passed on, time understood.
02
TaskSay who cooked and whether her kitchen was visited.
Success criterionAnswer found on the listing.
03
TaskTrack the order and find the handover code.
Success criterionCode given to the rider without searching.
04
TaskCook: add a dish and see the week's earnings.
Success criterionDish published, Friday's amount read.
Interview questions
- How do you find a home-made dish today?
- What would make you trust the kitchen of someone you don't know?
- Cooks: what would stop you from selling your dishes?
What I take from it
01
Trust has a face
Naming the cook, her neighbourhood and her visit did more for trust than any written guarantee.
02
A constraint becomes a promise
Small batches, a flaw for a platform, became “cooked the same day”.
03
Design for the one who cooks
The earnings simulator and Friday's payout answer the real question: how much, and when.
Got a project in mind?
Describe the screen or the workflow that gets in the way: I reply with a first reading of the problem.