Meet SplitMetrics AI

Unlock your 10X marketing power

Advanced AI models for app marketing live here.

SplitMetrics AI is a research and development center, committed to driving innovation and building leading AI technologies designed to solve marketing performance challenges for mobile-first companies.

AI that helps advertisers with their ROI challenges

From accelerating productivity to scaling and measuring incrementality, we help our customers to increase their business revenues with AI.

  • 10M+ bids changed & 850M+ keywords analyzed daily
  • Improved campaign performance through bid & budget optimization
  • 10Y of expertise in mobile app marketing
  • Recommendation system for simplified user acquisition
  • 97M data points processed daily

Max Kamenkov

CEO & co-founder of SplitMetrics

“ Established in 2022, SplitMetrics AI is powered by a team of highly qualified experts in machine learning and data engineering, including PhD holders in mathematics and champions of Kaggle competitions. This exceptional team has developed Samba, a family of AI models designed for performance optimization. As the foundation for our core solutions, Samba enables continuous advancements in bid, budget, and CPP optimization, with even more innovations on the horizon.

Discover Samba—our AI model family

Samba is our first set of AI models for marketing performance optimization. It can understand and perform bidding and budget related tasks involving thousands of keywords and campaigns.

Advanced Forecasting Precision

Trained on over 97 million data points daily and 850M+ unique keywords, our models deliver exceptional accuracy in spend and conversion predictions.

Performance Excellence

Engineered with 60+ signals from diverse data sources, our models analyze conversion rates, seasonality, and more, surpassing traditional campaign management techniques.

The most capable AI models we've ever built

Built on 10 years of field-defining research, Samba is a model family optimizing marketing campaigns to hit your ROAS targets.

  • 10+ million successful bid changes each month.

Data gathering

We gather data from various sources and combine it with our market intelligence, ensuring a full view of ad performance and user behavior.

Data preparation

We prepare the data as a unified dataset by removing duplicates, addressing missing values, and correcting metric discrepancies.

Feature engineering

Then we engineer features to enhance the model’s predictive power by converting raw data into meaningful inputs.

Bringing AI to our products, supporting $2B in ad spend.

More revenue from Apple Ads

Scale Apple Ads more efficiently. AI predicts performance and adjusts keyword bids and budgets daily.

Scale with AI predictive systems

Our AI makes 10 times more bid and budget adjustments than you can, ensuring optimal results every day for each keyword.

Analyze and forecast

Accurate daily forecasts based on historical and market data, with performance results like spend, conversion volume and CPA for each bid.

Innovation backed by Technical Reports

SplitMetrics AI is at the frontier of exploring new opportunities to drive business impact.

Capability

  • Standard Bid Management
  • AI bid and budget optimization with Samba 3.0

Time to set up

Varies by campaign complexity

Metrics

– CPA, CPI, ROAS & Cohort ROAS

Optimization modes

– Single and Multi-campaign

Campaigns eligible

– All campaigns

Data processing

Based on available datasets

Scalability

Foundational support

Handling of seasonal trends

Configurable adjustments Dynamic and incorporates trends into optimization logic

Performance

Varies based on setup and oversight

Adaptability

Depends on timely UA manager’s reaction

Our AI Journey & Roadmap

As AI technology continues to evolve, so do our model capabilities and their impact on advertisers. Learn what we achieve and what’s coming next on our roadmap.

Implemented:

  • AI Bid Optimization
  • AI Bid Simulator
  • AI Cohort ROAS Optimization
  • AI Budget Forecasting and Optimization

In development:

  • AI Brand Incrementality Detection

Our AI Policies & Governance

Our models undergo extensive safety tests and training in sandbox environments before deployment. As technologies evolve, we’ll keep refining our process, always committed to the highest standards of data privacy and security.

Certifications

We are certified under SOC 2 and ISO 27001 standards, which validate our commitment to rigorous data security and privacy practices.

Data Types

  1. Campaign Data

We train our AI models using aggregated data from Apple Ads Management API.

  1. MMP Data

We utilize aggregated data, such as the sum of specific types of events per day.

  1. Market Intelligence

We use aggregated market intelligence data from third-party open sources to enhance our services.