VETERAN PRO stops asking veterans to dig their own rights out of a catalogue. The app walks them through it: confirms status, points to the next step, opens benefits and partner discounts, and keeps a direct line to a support specialist.
A live, clickable prototype is on the right → try signing in with BankID
1.3M+
veterans and families in Ukraine
10
service categories in the catalogue
4
tabs, one coherent app
The difference
A catalogue answers questions. A companion asks them first.
The Veteran PRO portal, e-Veteran, Zahyst+ are reference directories. You log in, search, read an instruction. VETERAN PRO flips that around.
How it works today
A passive catalogue
The veteran has to already know what they are entitled to
Search and filters instead of guidance
An instruction, if you find the right section
No feedback on progress at all
VETERAN PRO
A proactive companion
Surfaces the next step and why it matters
Guides a 21-step route from status to a business grant
Reminds, rewards with points, opens nearby discounts
One tap to your own support specialist
Product
Six screens, one journey
Everything below is a real screen from the prototype. Every action works: login, onboarding, completing missions, discount QR codes, chat, request statuses.
Home — your route
Level and XP, a 6/21 progress bar, the next step, active missions and the support-specialist card.
The UBD route
21 steps across 5 blocks: from confirming status through documents and benefits to running your own business.
Opportunities
Partner discounts with filters and a QR code at the till. Some offers unlock as your level grows.
AI-bro
An assistant that explains state services in plain language. In production it is the existing Veteran PRO AI assistant.
My services
A 10-category catalogue and the status of requests in progress. It mirrors the CRM — the app never creates them.
Support specialist
A real person from the CRM. Call or message, with no new paperwork inside the app.
XP is earned for steps that actually change a life: confirming status, claiming a benefit, booking rehabilitation. No loot boxes, no pressure, and respect for an adult audience.
XP per action
Confirm status, +150. Claim the utilities benefit, +150. First partner discount, +75. Every step is explained openly.
Five levels, no demotion
From "Recruit in the system" to "Veteran ambassador". A level unlocks exclusive offers and never drops back down.
Proactive rules
The loyalty engine decides who sees an offer and when, with frequency caps and quiet hours so it never turns into spam.
A separate loyalty module
Partners, offers, QR and segmentation are configured outside the app. The same module can plug into the portal or Diia.
Levels
1Recruit in the system
2In touch
3One of us
4Brothers’ backbone
5Veteran ambassador
Architecture
Modular, done properly, by one team
The client asked for this the way large marketing platforms are built: the data layer, loyalty and the app are separate systems you can swap without breaking the rest. One vendor across all parts removes the risk of drawn-out API negotiations between teams.
01 · Data sources
Veterans registry (EDRVV)
CRM (Odoo / OSKAR)
02 · Data layer, logic, auth
CDM · data layer
Loyalty system
Thin backend (BFF)
IdP · BankID
03 · Channels & consumers
Mobile app
Veteran PRO portal
AI chatbot (AWS)
SuperSet · reports
IdP and BankID
Single sign-on on Keycloak with a custom connector to BankID NBU (which is OAuth 2.0, not OIDC). It returns full name, tax ID and phone. The tax ID is isolated behind an internal subject_id and every access is written to an audit log.
Thin backend
Stores app state and the extended profile, assembles the next-best step, serves GET /home. It stays thin and talks to subsystems through clear API contracts.
Customer Data Management
The client is considering Apache Unomi. Our honest recommendation: for the MVP, a light custom CDM on PostgreSQL + REST that consolidates the registry, Odoo CRM, app and loyalty by tax ID. Unomi is a phase-two option, since it is a heavy headless web-tracking CDP.
Loyalty
A standalone service: better to adopt open source (Open Loyalty) than build from scratch. Partner admin, offers, QR redemption, segmentation by status, region and cohort, plus gamification settings.
AI chatbot
The existing AWS bot is wired in at the API level only, with no in-house AI work, exactly as it already runs on the portal.
Integrations
The EDRVV registry through Trembita. Request statuses are a read-only mirror of the specialist CRM (Odoo or OSKAR — source pending client confirmation). Anonymised analytics go to SuperSet in phase two. Push runs on FCM and APNs.
Pulling personal data from the registry and CRM only works once the system is entered into the security profile (deadline ~10 Aug 2026) and a BankID certificate is issued. Until then that data is mocked in the prototype. Deployment lives in the ministry’s managed infrastructure. We make no claim about data residency beyond what a legal review confirms.
Competitive field
Straight talk about the neighbours
Ukraine already has analogues of individual modules. Nobody has the full combination — an official route, verified identity, loyalty, support and AI in one place.
Zahyst+
Discounts and QR via Diia. It owns the loyalty niche but has no gamification and no proactivity.
Army+ "Pluses"
Ministry of Defence: 504M UAH saved, 16 partners. Proof that the "offer → QR → redemption" model works.
Reserve+
6M users. A model for a thin app over a state registry with familiar login.
Diia
It carries the veteran e-ID. We are not competing with Diia — we build a separate proactive app for the forum and for the outreach Diia cannot do.
Our edge is a single vendor across every subsystem: no long contract negotiations between teams under a tight deadline.
Scope & timeline
What ships in time for the forum
MVP by 22 Aug 2026
BankID sign-in + verified profile
Onboarding and extended questionnaire
Home: route, missions, specialist
Service catalogue + read-only "in progress" statuses
Opportunities + discount QR codes
AI-bro (integration of the existing bot)
Push notifications
Rule-based gamification + basic analytics
Deliberately phase two
Creating CRM requests from the app
ML personalisation of offers
SMS and email automation
Partner self-service portal
A full BI layer in SuperSet
≈ 9–10 weeks
Timeline
4–5 people
Team
from $50k
Demo MVP
Team: 2 Flutter, 1–1.5 backend (BFF), 0.5 DevOps/Keycloak, 0.5 design, 0.5–1 QA, 0.25 security/architecture. Figures are an estimate for the demo MVP; a pilot that clears the full security profile widens both scope and budget. Final numbers follow a locked scope.
Risks, honestly
What sits outside our control
The BankID certificate
The ministry obtains the contract and certificate (free for state bodies). If it is not ready in time, the demo runs on the test environment.
Security profile by 10 Aug
Without it, EDRVV and CRM data stay mocked. The deadline holds; the scope narrows.
App Store / Play accounts
Create them early, or risk a month of review right before the forum.
Benefit figures
Amounts change by Cabinet resolution. In production we pull them from a managed source rather than hard-coding.
Try the app right now
A fully clickable prototype. Sign in with BankID, go through onboarding, complete a mission, open a discount QR code.
