04 /2026
ConceptRestaurant booking app
From competitor analysis and a UX review to a three-step booking flow.
Process depth
- Competitors
- 4
- Research
- 1
- Observations
- 7
- Insights
- 3
- Pain points
- 2
- Opportunities
- 1
- Flows
- 1
- Screens
- 3
- Decisions
- 1
How many records of each stage the project's workbook holds. It is the amount of documented work, not a score.
This is a design concept: the product was not launched, browse the project pages below. Project pages ↓
Sample case: this is how a project published from the workspace will look. Numbers and quotes are illustrative for now.
Decision trail
Records and links from the workbook. Hover or tap a record to see what it rests on and what it leads to.
18 records · 20 links. Hover a record to see its chain.
Observations · 7
Insights · 3
Pain points · 2
Opportunities · 1
Flows · 1
Screens · 3
Decisions · 1
Record titles are kept as written in the workbook, in Ukrainian.
Project pages
Click a page to view it full screen. Flip through with the arrow keys or a swipe.
On a phone
01 /10
In short
Challenge
People pick a restaurant for an occasion — a date, a birthday, a business lunch — while apps search by cuisine and rating. To find a free table you still have to call.
Solution
The occasion is the first filter, free slots show right in the results, and booking takes three steps without leaving the app.
Outcome
In a prototype test all 8 participants booked a table in under a minute.
02 /10
How I worked
- interviews
- 8interviews
- quotes → 3 insights
- 23quotes → 3 insights
- competitors, 10 heuristics
- 3competitors, 10 heuristics
- opportunities
- 7opportunities
- flow steps
- 6flow steps
- screens with states
- 12screens with states
- decisions with evidence
- 9decisions with evidence
03 /10
Research
In-depth interviews with people who booked a table in the past month. I didn't ask whether it was convenient — I asked how exactly they chose last time.
- interviews
- 8interviews
- average length
- 45 minaverage length
- quotes in synthesis
- 23quotes in synthesis
I'm not looking for “Italian food”. I'm looking for where to go with my girlfriend on Friday.
— Olena, 27 By the time you get through on the phone, you don't feel like it anymore. I just go where I've been.
— Andrii, 34 Instagram has great photos, but to book you have to DM and wait.
— Mariia, 24 I want to see if there's a table at 7 pm before I even open the menu.
— Ihor, 41
04 /10
Insights
People look for an occasion, not a cuisine
The choice starts with the situation: a date, a celebration, a meeting. Cuisine is a secondary filter.
7 quotes · 6 of 8 participantsUncertainty about a table kills the intent
When free seats aren't visible, people postpone the decision or go somewhere familiar.
5 quotes · 5 of 8 participantsInspiration and booking live in different apps
Ideas come from Instagram, booking happens by phone or in another service. Half of the intent is lost in between.
4 quotes · 4 of 8 participants05 /10
Competitors
A feature matrix plus a UX review against Nielsen's heuristics. Competitors' red cells are where the product can do better.
| Our product | Google Maps | TheFork | ||
|---|---|---|---|---|
| Filter by occasion | Yes | No | No | Partly |
| Free tables in results | Yes | No | Partly | No |
| Online booking | Yes | Partly | Yes | No |
| Guest photos | Yes | Yes | No | Yes |
| One-tap “Book again” | Yes | No | No | No |
| Nearest slots when the time is taken | Yes | Partly | No | No |
06 /10
Opportunities
- OPP-01
How might we help people pick a place for an occasion in a minute?
- OPP-02
How might we show a free table before the restaurant page opens?
- OPP-03
How might we turn inspiration from a photo into a booking in one tap?
07 /10
Flow
The main scenario — from “where to go” to a confirmed booking.
- 01Opens the app
- 02Pick an occasion
- 03Picks with free slots
- 04Chooses a time
- 05Confirmation
- 06Booking in calendar
Edge cases
- Slot taken while choosing → offer the nearest free ones
- No connection → booking queued, notification once sent
- More than 8 guests → call the restaurant with one button
08 /10
Screens
12 screens, each with its states described: default, loading, empty, error.
Occasion
Six scenarios instead of forty filters
- Default
- Loading
Picks
Free slots right in the card
- Default
- Empty
- Error
Booking
One primary action — “Book”
- Default
- Success
- Slot taken
09 /10
Decisions and why
DEC-001Occasion first, cuisine as a filter inside
Why: 6 of 8 participants started from the situation, and no competitor has such a filter.
Rejected options: The usual cuisine search on the first screen; A recommendation feed without a choice. Evidence: INS-001, Q-004, Q-011.
DEC-002Free slots in the results card
Why: Uncertainty about a table is the main reason people give up on going out.
Rejected options: Slots only on the restaurant page. Evidence: INS-002, Q-007.
DEC-003“Book again” on the home screen
Why: Rebooking takes five steps in TheFork — a violation of the “Flexibility and efficiency” heuristic.
Rejected options: Booking history only in the profile. Evidence: UX review · TheFork #7.
10 /10
Results
A usability test of the clickable prototype with 8 participants.
- average time to book
- 58 saverage time to book
- participants completed a booking
- 8/8participants completed a booking
- steps compared to TheFork
- −3steps compared to TheFork
For the first time I picked a restaurant by mood, not cuisine — and I can see right away there's a table.
Next case · 05 / 06
ATELIER — lingerie on a glass mannequin
Tell me briefly about the product and the timeline — I'll reply within a day.