A mobile app that helps people quickly understand an air raid
alert and reach the nearest shelter.
Role
Product Designer
Scope
Research, UX, UI Design
Platform
Mobile App
Industry
Safety Tech
Timeline
1 month
01
3:17 AM
The siren goes off. Notification appears. It tells you almost
nothing about what's actually happening.
Is it a drone or a missile? Is the threat nearby? Do you need to get
to a shelter now or can you wait?
To answer those questions, people leave the alert and start checking
other sources. Telegram. Maps. Local channels. It takes another
three to five minutes before they know whether to act or wait.
The Sequence of a Typical Response
!Siren
Sound triggers alert
2Push Alert
Notification arrives
3What type?
Missile? Drone?
4How far?
Direction unclear
5Switch apps
Multiple sources
✓Decision
Act or wait?
Alert← avg. 3-5 min of uncertainty →Action
02
A Simple Flow
During every alert, people follow the same sequence: understand
what is happening, decide whether it affects them, then act if
necessary.
Core Flow
01
Alert
The official alert becomes the entry point
Region and time from official sources.
›
02
Context
Assess the threat
Threat type, direction and estimated area.
›
03
Shelter
Find the nearest available shelter
Distance and map.
›
04
Route
Get there with one tap
Estimated time and directions.
TriggerAction
Alert -> Context -> Shelter -> Route.
03
Research
4
Methods
3
Weeks of Observation
12+
Sources Analyzed
Research combined competitor analysis, user observation, informal
interviews and firsthand experience using alert apps during air
raids.
Together, these methods revealed what people actually do after an
alert, not just what existing apps were designed to support.
Key Observations
01
What happens after the alert?
The notification was rarely the end of the journey. It was the starting point.
People immediately turned to other sources to understand whether the threat affected them.
02
How do people decide what to do?
No single source answered the whole question.
Telegram explained the threat. Maps showed nearby shelters. Local chats confirmed whether they were actually open. People had to piece the answer together themselves.
03
What slows people down?
Finding shelter almost always meant switching between multiple sources. Practical details like accessibility, conditions, or availability were often missing.
04
From alerts to decisions
The product wasn't designed to explain the alert. It was designed
to support the next decision.
Before
Identify the threat
Focus on the threat
→After
Support the response
Focus on the next action
05
Designing for urgency
During an alert, attention is limited. The interface was designed
around one question: what needs to be visible first?
Critical
Always visible. Drives immediate action.
Threat Type
Missile, drone, aircraft
Threat Direction
To which region
Shelter Access
Nearest open shelter
Important
Available one tap away. Supports verification.
Threat Map
Visual spread overview
Trusted Sources
Air Force, regional channels
Secondary
Accessible but deprioritized. Not time-critical.
Community
Peer support
Mini-game
Stress relief
Core Principle
Alert → Context → Action
◎
01
Push Alert
Signals that attention is needed. Entry point into the experience.
›⚡
02
Threat Context
Reduce uncertainty before action. Official context before making a decision.
›⊕
03
Shelter Access
Show the nearest available shelter.
›→
04
Route
Remove friction when every second matters. The fastest path to safety.
06
The Product Experience
Primary Flow
Alert State.
Answers the first question immediately: is something happening
right now?
Threat Context.
Explains what is happening and whether it affects the user.
Shelter Access.
Removes the extra step between recognizing the threat and finding
shelter.
Supporting Tools
Supporting tools add context without interrupting the core flow. The
threat map shows how the situation is evolving. Trusted sources allow
people to verify information without leaving the app. Alert history helps
identify patterns beyond a single event.
Personalization
People follow different regions and live different routines. Region
selection keeps multiple important locations visible at once, while
notification settings adapt alerts to everyday life without changing the
core experience.
Beyond the Core Flow
Community. Gives people a way to share local context
that official sources often can't provide.
Mini-Game. Offers a distraction while waiting. Every
coin earned can be turned into a donation.
07
What the product intentionally doesn't do.
Not every problem is solved by a new feature.
The product stays
focused on a single job: helping people understand the situation
and decide what to do next.
Does not generate alerts
It adds context to alerts that already exist.
Does not make decisions
It provides information. The final decision is up to the person.
Does not amplify fear
The interface stays calm because the situation is already stressful enough.
Does not replace official sources
It complements official sources by gathering essential information
in one place.
08
The hardest decision in this project wasn't what to add. It was
what to remove.
At first, showing more felt like the obvious answer:
a complete
threat map, alert history and details from every available
source. But every new layer of information required more time to
read before people could decide.
Eventually, I realized the product didn't need to explain
everything. It needed to answer one question as quickly as
possible: should I act now or can I wait? Everything else exists
to support that decision,
not compete for
attention.
Primary InsightThe first answer matters more than every answer.Core DecisionSupport the decision. Never make it.Design PrincipleClarity before completeness.
Next Project
Idea School
Website redesign for a private school that helps parents figure out
if it's the right fit for their child.
