Faster checks. Full fleet visibility. Zero data exposure. A managed AI compliance engine that scales across all markets.
THE CHALLENGE
Manual audits hit a ceiling
As rider fleets grow across markets, patrol-based compliance checks cover only a fraction of active riders on any given day. The rest ride unchecked — a compliance blind spot that scales with fleet size.
10-12 min
Per manual check
~38%
Typical fleet coverage
60%
Patroller time on audits
Hiring alone doesn't solve this
Each additional patroller costs approximately AED 68,500 in year one. Adding 15 patrollers to reach full coverage on a 40K fleet costs over AED 1M annually — and still only reaches ~84% when the fleet grows to 55K. The economics of headcount-based compliance don't scale.
THE SOLUTION
Two modules, one compliance engine
Module A makes your patrollers dramatically faster and more consistent. Module B extends coverage to every rider, every shift — without a single new hire.
MODULE A — AI-ASSISTED T-PATROL
Supercharge existing patrollers
AI integrates directly into the existing Boolanga Compliance Auditor App. Patrollers take a photo — AI instantly evaluates gear, uniform, branding, and safety equipment. Results auto-populate the audit report and sync to OMS.
✓
Check time reduced from 10-12 min to ~6 min
✓
Standardized scoring across all 8 markets
✓
Eliminates subjective variance between patrollers
✓
Direct OMS sync — zero manual data entry
MODULE B — RIDER PRE-SHIFT SELF-CHECK
Every rider visible, every shift
Standalone web-based tool — no app install. Before each shift, riders take a selfie in full gear. AI evaluates compliance in real-time. Patrollers receive prioritized lists: who failed, who skipped, who hasn't been checked in 30+ days.
✓
Coverage extends from patroller-only to entire fleet
✓
Rider images never leave Boolanga infrastructure
✓
Patrollers become targeted enforcement, not random patrol
✓
Full audit trail — every rider, every shift
Why combined is greater than the sum of parts
Module A alone improves the ~38% of riders patrollers already reach. Module B alone gives broad coverage. Together, every rider self-checks before shift (B), and patrollers focus exclusively on flagged riders (A). Coverage goes from ~38% to effectively 100% — without a single additional hire.
HOW IT WORKS
Daily compliance flow
A continuous loop that catches non-compliance before riders start their shift — not hours after.
1
Pre-Shift Self-Check
Rider opens the web-based check-in tool, takes a photo in full gear. Under 60 seconds. No app install.
2
AI Evaluation
Computer vision analyzes uniform, branding, helmet, safety gear, and windshield. Real-time pass/fail with violation codes.
3
OMS Status Update
Compliance status logs automatically. Passed riders proceed. Failed riders get remediation instructions.
4
Targeted Patrol
Patrollers receive a prioritized list: failed self-checks, skipped checks, riders not checked in 30+ days.
5
Full Audit Trail
Every check logged with timestamp, photo evidence, AI score, and pass/fail. Complete visibility across all markets.
PRICING
Flexible pricing, predictable cost
Per-picture if you want costs tied to usage. Unlimited if you want a fixed monthly spend. Bundle both modules for the best economics.
OPTION A · PER PICTURE
T-Patrol AI
€0.008
per check
✓
AI visual compliance scoring
✓
Real-time pass/fail + violation codes
✓
Automated OMS sync
✓
Model retraining & accuracy monitoring
✓
Quarterly business reviews
12-MONTH CONTRACT
OPTION A · UNLIMITED
T-Patrol AI
€4,500
per month — unlimited checks
✓
Everything in per-picture, plus:
✓
No volume cap
✓
Fixed cost regardless of output
✓
Ideal as patroller efficiency grows
✓
Cost certainty for budgeting
24-MONTH CONTRACT
OPTION B · PER PICTURE
T-Patrol + Self-Check
€0.004
per check — 50% lower rate
✓
Everything in Option A, plus:
✓
Rider pre-shift self-check tool
✓
Prioritized patrol lists from AI
✓
Full fleet coverage
✓
€8,000 one-time setup fee
12-MONTH CONTRACT
BEST VALUE
OPTION B · UNLIMITED
T-Patrol + Self-Check
€6,500
per month — unlimited checks
✓
Everything in per-picture B, plus:
✓
Unlimited rider + patroller checks
✓
Fixed cost at any volume
✓
Setup fee included
✓
Best economics at scale
24-MONTH CONTRACT
Monthly cost at different volumes
All in EUR. Highlighted = per-picture exceeds unlimited.
Monthly Checks
A Per-Pic
A Unlimited
B Per-Pic
B Unlimited
400K
€3,200
€4,500
€1,600
€6,500
600K
€4,800
€4,500
€2,400
€6,500
1M
€8,000
€4,500
€4,000
€6,500
1.5M
€12,000
€4,500
€6,000
€6,500
2.5M
€20,000
€4,500
€10,000
€6,500
5M
€40,000
€4,500
€20,000
€6,500
ADDRESSING THE QUESTION
Why a one-time fee doesn't work
This is the most common question we receive on AI pricing. Here's a transparent breakdown of why recurring pricing is the only model that protects both parties.
01
The AI runs on dedicated GPU infrastructure — 24/7
AI visual compliance requires dedicated high-performance GPU compute running continuously. This infrastructure has a fixed monthly cost whether it processes zero images or three million. A one-time fee cannot cover an ongoing hardware commitment.
02
Your gear standards aren't static — the AI can't be either
Talabat changes uniforms, introduces new branding, rolls out seasonal gear, adjusts safety requirements across markets. A one-time model is frozen at delivery. Within 6 months it scores against outdated standards. Ongoing pricing covers continuous retraining across all 8 markets.
03
AI accuracy degrades without active monitoring
Image quality varies by device, lighting, and rider behavior. Without continuous monitoring, false positive and negative rates drift upward. The service includes real-time accuracy tracking, weekly precision reporting, and automatic model correction.
04
Building internally is recurring cost too
Internal build means developer salaries (recurring), GPU compute (recurring), cloud infrastructure (recurring), and maintenance (recurring). The choice isn't one-time vs. recurring — it's your recurring cost vs. ours. Ours is lower because infrastructure is shared across clients.
05
Per-check pricing is lower risk for you
Costs scale proportionally. Low-usage month? Lower cost. Pause a market? No charges. Scale to new countries? Predictable growth. You never overpay for unused capacity.
UNIT ECONOMICS
Cost per audit — the real comparison
Scenario
Monthly Audits
Coverage
Cost / Audit
Manual only (36 patrollers)
~15,000
38%
AED 13.70
AI-assisted patrol (Opt A)
~33,000
82%
~AED 7.00
AI + Self-Check (Opt B)
46,000+
95-100%
~AED 5.50
49%
Cost-per-audit reduction
2.2x
Coverage multiplier
~6 min
Average check time
DATA & SECURITY
Complete data isolation
No API integration
AI operates within Boolanga infrastructure. No connection to Talabat's rider app or systems.
Images stay in Boolanga
Photos never leave our servers. Talabat receives only structured data — pass/fail, violation codes, timestamps.
External security audit
Independent global IT security team engaged. Periodic review access available to Talabat's security team.
Encryption & access control
Data encrypted at rest and in transit. Role-based access. Full audit logging. Regular vulnerability assessments.
FAQ
Common questions
Yes. Option A deploys independently. If you add Module B later, bundle pricing applies from that point. However, launching both together gives the best per-check rate from day one and avoids a second onboarding cycle.
We run a validation pilot in a single city before full rollout. If accuracy doesn't meet the agreed threshold (≥95%), we continue tuning at no additional cost until it does. You don't pay for full service until accuracy is validated.
Module A: 4-6 weeks. Module B adds 3-4 weeks for the self-check tool build. Full deployment across all markets: 8-12 weeks from kickoff.
Visual: uniform condition, branding placement and quality, helmet type and condition, safety gear (vest, gloves), windshield status, and overall presentation. Non-visual checks (hygiene, odor, politeness) remain manual fields in the patroller workflow.
New markets are onboarded by training the AI on that market's specific gear standards and branding guidelines. Typical onboarding: 2-3 weeks. Volume from new markets is included in existing pricing.
NEXT STEPS
From proposal to production
1
Confirm scope and pricing option
Align on Option A or B, per-picture or unlimited. Finalize contract terms and rollout sequence.
2
AI training data and label sets
Define violation types, gear standards, and tolerances per market. Boolanga prepares the AI model using existing audit data.
3
Pilot in one city
Live AI-assisted audits with a limited patroller group in Dubai. Validate accuracy, speed, and OMS integration.
4
Validation and full rollout
Once accuracy ≥95%, expand to remaining markets. Full deployment within 8-12 weeks of kickoff.