Deployment
CloudThe app runs on the device, while its backend, AI inference, subscriptions, attribution, and acquisition depend on cloud services.
Case-backedCloudAdvanced
The Paid-Acquisition AI Mobile App Stack Behind ChartDetector’s Reported $56K/Month
A case-backed breakdown of how ChartDetector reportedly moved from a lightweight AI prototype to a subscription mobile app powered by attribution, short-form creative, and paid acquisition.
The revenue, profit, download, and operating-time figures on this page come from a founder interview. AI LinkBase has not independently audited them, and they are not a forecast for another product.
Some tool links may become affiliate links. Affiliate availability does not determine inclusion, Component Status, or Current Primary selection. Disclosure details
Stack type
Case-backed
Workflow
8 stages
Execution map
9 capability groups
Primary category
Marketing & Business
Fixed launch baseline
About $52.25/month
Setup difficulty
Advanced
Prices reviewed July 27, 2026 · Full Stack reviewed July 27, 2026
Decision profile
Architecture Snapshot
These labels describe how the complete workflow operates—not the license or deployment mode of every individual component.
Commercial readiness
ConditionalA commercial product is possible, but AI, app-store, privacy, advertising, subscription, and finance-related requirements need separate review.
Model openness
ProprietaryThe original case points to a hosted multimodal API rather than an open-weight core.
Setup difficulty
AdvancedThe workflow spans mobile development, in-app purchases, event attribution, privacy controls, paid media, and contribution economics.
Source case
Case Snapshot
ChartDetector lets users upload or photograph trading charts and receive pattern, trend, and educational strategy summaries. AI LinkBase does not assess or endorse the accuracy of those outputs.
- Project
- ChartDetector AI
- Founder / team
- Timo Köhler and his brother
- Product
- Subscription iOS chart-analysis app
- Acquisition
- Primarily TikTok paid advertising
- Primary source
- Starter Story, July 23, 2026
- Evidence level
- Founder-reported; not independently audited
Evidence
Reported Results
Every figure keeps its evidence label so the historical case cannot be confused with an independently verified benchmark.
| Metric | Reported value | Context | Evidence |
|---|---|---|---|
| Monthly revenue | About $56,000 | At the interview date | Founder-reported |
| Cumulative revenue | More than $260,000 | First roughly 13 months | Founder-reported |
| Monthly downloads | About 13,000–14,000 | At the interview date | Founder-reported |
| April revenue | About $43,700 | Monthly example | Founder-reported |
| April profit | About $11,500 | The source does not provide a complete audited cost ledger | Founder-reported |
| Profit margin | About 25% | Reported overall | Founder-reported |
| Operating time | About 20 hours/month | At the interview date | Founder-reported |
Cost control
Cost Snapshot
Fixed software is separated from variable usage, advertising, store fees, refunds, taxes, and labor. The largest cost is not necessarily a SaaS subscription.
Minimum Viable Validation
EstimatedNear $0 fixed software
- ✓Validate the core AI action in a bot or lightweight web prototype.
- ✓AI inference remains usage-based.
- ✓Public iOS distribution later requires a $99/year Apple Developer membership.
Not included
Paid advertising · Public app distribution · Human labor
Recommended Launch Baseline
EstimatedAbout $52.25/month — $72.25 with optional Cursor
- ✓Expo Starter: $19/month
- ✓Supabase Pro: from $25/month
- ✓Apple membership: about $8.25/month equivalent, billed annually
- ✓Cursor Pro: optional $20/month
Not included
AI usage · Revenue-linked subscription fees · Attribution usage · Advertising · Store commissions, refunds, taxes, legal review, and labor
Case-reported Scale
Founder-reportedAbout $18,000/month in paid media
- ✓The founder reported this TikTok spend at the interview date.
- ✓A separate $50/day starting suggestion equals about $1,500 over 30 days; it is not a universal platform minimum.
Not included
Complete infrastructure and AI usage · Full labor, tax, legal, refund, and support costs
Operating system
How the Workflow Works
The order matters: validate the task, build the conversion and measurement path, then buy traffic and scale only when contribution economics work.
01
Opportunity
Map a new AI capability to a narrow user problem.
Components
New model capability + use-case definition
Output
A testable product hypothesis
↓
02
Lightweight Validation
Test real usage before building the full mobile product.
Components
Telegram bot or the simplest suitable prototype
Output
Usage feedback and a build-or-stop decision
↓
03
Mobile Product
Turn the validated interaction into an installable app.
Components
React Native + Expo + Supabase + multimodal AI
Output
A working mobile application
↓
04
Conversion System
Turn product value into a measurable subscription.
Components
Onboarding + paywall + RevenueCat
Output
A verified subscription event
↓
05
Attribution
Connect paid subscription outcomes to acquisition sources.
Components
AppsFlyer + verified event pipeline
Output
Campaign-level subscription and revenue attribution
↓
06
Creative Production
Continuously create product demonstrations for testing.
Components
CapCut + creative version process
Output
Multiple testable short-form ads
↓
07
Paid Acquisition
Acquire users while optimizing toward a revenue-related event.
Components
TikTok Smart+ App Campaigns
Output
Installs, subscriptions, acquisition cost, and return data
↓
08
Unit Economics & Scale
Increase budget, refresh creative, or stop using contribution economics.
Components
RevenueCat + AppsFlyer + ad data + finance sheet
Output
A sustainable growth loop or a stop decision
Portfolio
Original Tools vs Current Portfolio
Original Tools preserve what the case reported. Current Portfolio reflects present workflow fit. A Required slot can remain empty when evidence is not strong enough to name a Current Primary.
S1OpportunityRecommendedUnder review
AI opportunity source
Original case
ChatGPT image capability
Current Primary
Selection still under review
The reusable capability matters more than preserving a historical product brand.
S2ValidationRequiredSelected
Validation interface
Original case
Telegram bot
Current Primary
Choose the lowest-cost interface that already matches where target users are.
S3Mobile productRequiredSelected
Cross-platform mobile product
Original case
React NativeExpo
Current Primary
A strong cross-platform path with direct integrations into the selected backend and subscription systems.
Important condition: Use an Expo development build for native subscription modules; Expo Go is not enough for production purchase testing.
S4Mobile productRequiredSelected
Backend and data
Original case
Supabase — high-confidence transcript normalization
Current Primary
Database, authentication, storage, and an official Expo React Native quickstart in one system.
Important condition: Configure production security and monitor usage beyond included limits.
S5Mobile productRequiredPending benchmark
Multimodal AI inference
Original case
OpenAI — high-confidence transcript normalization
Current Primary
No Current Primary Yet
OpenAI remains the original baseline candidate, but price alone cannot decide a finance-related image-analysis task.
Important condition: Run OpenAI, Gemini, and Claude against the same 30–50 representative images before selecting a Current Primary or enabling a purchase CTA.
S6Mobile productOptionalSelected
AI coding environment
Original case
Cursor — high-confidence transcript normalization
Current Primary
Useful engineering assistance, but the workflow can succeed without a paid coding tool.
S7ConversionOptionalUnder review
Onboarding research
Original case
Screens Design — founder-recommended, not confirmed actual use
Current Primary
Selection still under review
The source presents it as a recommendation; current product fit and pricing still need verification.
S8ConversionRequiredSelected
Subscription infrastructure
Original case
RevenueCat
Current Primary
Centralizes paywalls, subscription state, entitlements, and revenue events across the mobile workflow.
Important condition: Verify real purchase events and entitlement state before buying traffic.
S9AttributionRequiredSelected
Mobile measurement partner
Original case
AppsFlyer
Current Primary
Supports paid-campaign attribution and subscription-event measurement.
Important condition: Recalculate plan fit when conversion volume, paid channels, or geographic coverage changes.
S10AttributionOptionalSelected
Attribution integration specialist
Original case
Upwork specialist — founder-recommended resource
Current Primary
Qualified mobile attribution specialist
The needed capability is verified MMP implementation experience, not a particular freelance marketplace.
S11CreativeRequiredSelected
Short-form creative production
Original case
CapCut
Current Primary
Closely matches the case’s fast product-demonstration workflow.
Important condition: Verify regional paid pricing and commercial terms before presenting a precise paid-plan cost.
S12Paid acquisitionRequiredSelected
Paid media platform
Original case
TikTok Ads / Smart+
Current Primary
Preserves the documented acquisition path, but only after the conversion and measurement system works.
Important condition: Recheck current platform guidance, finance-related advertising policy, event optimization, and cash-flow readiness before every material campaign change.
S13EconomicsRequiredSelected
Unit-economics monitor
Original case
RevenueCat + AppsFlyer + ad data
Current Primary
RevenueCat + AppsFlyer + finance sheet
Reuse existing subscription, attribution, and advertising data before adding another analytics product.
Replacement choices
Alternatives & Trade-offs
These are comparison directions, not automatic recommendations. Replacing one component can change integrations, operating burden, attribution quality, or the economics of the whole workflow.
| Slot | Lower-cost / direct path | Specialized path | Replacement trade-off |
|---|---|---|---|
| Mobile framework | Flutter | Native SwiftUI for iOS | Replacing React Native changes the development stack; native may improve platform control but removes shared cross-platform code. |
| Backend | Firebase free tier | Appwrite or a managed cloud architecture | Firebase changes the data model and vendor dependency; deployment-focused alternatives add operational decisions. |
| AI inference | Gemini API test tier where appropriate | Claude API or a task-specific model | Every model must be tested on identical representative images; price alone cannot decide this slot. |
| Subscription | Native StoreKit and Play Billing | Adapty | Native code removes an extra SaaS layer but increases engineering and cross-platform maintenance. |
| Attribution | Platform-native reporting | Adjust or Singular | Lower-cost reporting may not provide equivalent cross-channel attribution or subscription-event context. |
| Creative | Canva free tools | Adobe Premiere | Simpler tools speed iteration; specialist tools add control while increasing production time and complexity. |
| Paid media | Smaller controlled tests on a suitable channel | Meta Ads or Apple Search Ads | Audience intent, creative format, learning behavior, and attribution differ; these are not drop-in replacements. |
| Economics | Spreadsheet | Product analytics or a data warehouse | A spreadsheet is sufficient early but becomes fragile with higher volume, multiple channels, and complex cohorts. |
Transferability
What to Copy—and What Not to Assume
What is transferable
- ✓Validate the core AI action in a lightweight interface before building the complete mobile product.
- ✓Optimize acquisition toward a real purchase event rather than downloads alone.
- ✓Treat onboarding, paywall, subscription state, attribution, and advertising as one connected conversion system.
- ✓Maintain multiple creative variants and a replacement pipeline.
- ✓Scale according to contribution profit rather than surface ROAS.
- ✓Preserve Original Tools as history while allowing Current Recommendations to change.
What may not transfer
- !The founder had software-engineering experience and worked with his brother.
- !The product entered during an earlier image-understanding capability window.
- !Finance-related AI products carry higher policy, legal, and user-protection risk.
- !A hard paywall without a trial can reduce trust, retention, or refund performance in another market.
- !Paid acquisition requires enough cash to survive learning periods and failed creatives.
- !The reported margin may change materially after complete refunds, taxes, AI usage, support, and labor costs.
Guardrails
Risks and Stop Conditions
Paid acquisition should amplify a verified system, not hide a broken product, unsafe claim, or incomplete cost model.
Critical
Users may interpret AI-generated chart analysis as financial advice.
Mitigation: Use explicit educational positioning, uncertainty handling, product-policy review, and independent legal review.
High
Founder-reported revenue may be mistaken for a reproducible result.
Mitigation: Bind the Founder-reported label to every result and state that AI LinkBase did not audit the figures.
High
Attribution errors can train the ad platform on the wrong event.
Mitigation: Test subscription and revenue-event delivery before paid traffic; use a qualified specialist when needed.
High
Surface ROAS can hide refunds, store fees, AI usage, backend overages, creative cost, and tax.
Mitigation: Scale only from a complete contribution-profit view.
High
Creative fatigue or platform-policy changes can break acquisition performance.
Mitigation: Maintain a creative pipeline and review platform guidance before material campaign changes.
Stop when: product value is unproven, financial or income claims would be misleading, subscription events cannot be attributed reliably, contribution profit stays negative after material variable costs, or the team cannot fund the learning period safely.
Frequently Asked Questions
Is this a proven way to make $56,000 per month?+
No. It documents one founder’s account. AI LinkBase has not audited the financial data, and another product may produce very different results.
Do I need every tool used in the original case?+
No. Required capabilities matter more than brands. Original tools are preserved for historical accuracy; current tools should fit your product, country, skills, and budget.
Should I start with TikTok Ads immediately?+
No. Validate product value, subscription events, attribution, creative capacity, and contribution economics first.
Is a hard paywall always better?+
No. It was the founder’s strategy in this case and should be tested against trial, freemium, or preview-based alternatives.
How much does the Stack cost?+
The known recommended fixed baseline is about $52.25/month, excluding optional Cursor. AI usage, attribution, advertising, store fees, refunds, taxes, and labor are variable and may dominate the total.
Transparency
Sources & Methodology
AI LinkBase separates founder statements, publicly observable product facts, official current tool facts, editorial workflow reconstruction, current recommendations, and unresolved information.
Starter Story founder interview ↗ChartDetector official website ↗ChartDetector U.S. App Store listing ↗React Native ↗Expo pricing ↗Supabase pricing ↗Cursor pricing ↗OpenAI model pricing ↗RevenueCat pricing ↗AppsFlyer pricing ↗TikTok Smart+ App Campaigns ↗Apple Developer membership ↗
Source case: I Make $50K Per Month Working 5 Hours A Week, Starter Story, July 23, 2026. Short excerpts are used only where necessary; this page is an original workflow analysis.