Dieter Rappold, co-founder and CEO of Context SDK, joins Apptivate to explore how motion sensor data and on-device AI are reshaping mobile marketing strategies. He explains how contextual signals from accelerometers, gyroscopes, and other device inputs can help marketers identify the right moment to engage users, optimize conversion timing, and improve monetization without relying on personal data or tracking permissions. The conversation covers the evolution from event-based to moment-based optimization, practical implementation considerations, privacy implications, fraud detection use cases, and how contextual intelligence could power the next generation of agentic AI and decision-making in mobile growth.

• What types of smartphone sensor signals can marketers use today?
• How does Context SDK turn motion data into actionable growth insights?
• What is the difference between event-based and moment-based optimization?
• Which app categories benefit most from contextual engagement timing?
• How does on-device AI change personalization in a privacy-first world?
• What implementation effort is required to test contextual optimization?
• How can motion signals support fraud detection and ad performance?
• Where could contextual intelligence influence agentic AI and future UX?
• What strategic priorities should mobile marketers focus on next?

• (0:04) — Motion sensor data in mobile marketing
• (0:56) — Accelerometer and gyroscope explained
• (2:06) — Founding story and origins of Context SDK
• (3:10) — Awareness gap among mobile marketers about sensor data
• (4:24) — Airbnb example illustrating real-world user intent
• (5:44) — Physics-based data vs opinion-driven marketing models
• (6:55) — Session duration differences and conversion timing opportunities
• (8:22) — Event-based vs moment-based optimization strategy
• (10:13) — App verticals and use cases for contextual timing
• (11:15) — Additional signals beyond motion sensors
• (12:02) — Model training requirements and data scale needed
• (14:02) — Privacy compliance, ATT and permissionless personalization
• (15:11) — Fraud detection applications using motion behavior
• (15:43) — Ad network integration and performance uplift example
• (17:28) — Context data as signal layer for agentic AI
• (19:18) — Strategic priorities and competitive positioning for marketers
• (20:23) — Limits of sociodemographic targeting frameworks
• (22:06) — How to connect with Context SDK team
• (22:29) — Rapid-fire questions
• (24:58) — Episode wrap

• (4:45) “If you're walking down the street and open Airbnb. You probably look for the key code of the apartment that you have booked. But if you're comfy on the sofa and open Airbnb, you are probably planning your next vacation.”
• (5:03) “Humans constantly move in a three dimensional space while they're using their smartphones and the apps on it… we have different needs, different pain points, different session durations and different likelihoods to convert in different actions.”
• (8:47) “I do believe that we should not get rid of event based, but we should combine it with moment based because both things can tell us something. Event based gives us behavioral context, meaning the behavior in the app. But the question is, when is the timing right?”
• (13:35) “Based on how we built this, our architecture and our approach, we don't need ATT, we don't need any permissions, and we are out of the box GDPR compliant because we don't collect any PII, we don't collect a unique user ID, we don't collect unique device ID.”

• Dieter Rappold on Linkedin
• Context SDK

Bani Malhotra, former head of personalization and site experience for Walmart’s e-commerce platform, joins Apptivate to unpack the real drivers of revenue growth in modern digital products. She explores how recommendation systems, search behavior, ratings and reviews can influence conversion rates across the funnel. She also discusses why many teams misdiagnose the cause of revenue leaks, and how behavioral signals increasingly outperform demographic targeting. Bani also discusses the evolving responsibilities of product leaders, including ownership of go-to-market and revenue outcomes, and shares lessons from building AI-native consumer apps where product systems must handle probabilistic outputs, uncertainty, and the balance between automation and user control.

• Which product levers most reliably drive revenue in e-commerce?
• Where do companies actually lose revenue in the funnel?
• How should teams think about personalization without creating discovery problems?
• What behavioral signals best predict purchase intent?
• How is the product leadership role evolving?
• What changes when building AI-native consumer products?
• How should teams design systems when AI outputs are probabilistic?

• (0:03) — Bani Malhotra’s background and experience leading personalization at Walmart
• (3:17) — Why revenue growth comes from multiple connected drivers
• (4:02) — Recommendation systems and the evolution of personalization
• (5:10) — The impact of ratings and reviews on customer confidence
• (8:37) — Search behavior as a signal of user intent
• (13:35) — Diagnosing revenue leaks across the funnel
• (20:03) — When personalization becomes an echo chamber
• (27:27) — Why upselling can damage customer trust
• (32:35) — Behavioral signals versus demographic targeting
• (37:41) — How the product leadership role has evolved
• (40:54) — Designing AI-native consumer products
• (44:44) — Rapid-fire questions and closing

• (3:30) “Revenue isn't just working on one level. There are multiple revenue drivers that connect to each other, and when they work together in tandem, it compounds.”
• (8:48) “When somebody searches, not only are they starting consideration, they are giving you intent.”
• (13:39) “More often than not I have seen the biggest revenue leaks to be mid-funnel and bottom of the funnel.”

• Bani on Linkedin

Artsiom Kazimirchyk, co-founder and CEO of Campaignswell, joins Apptivate to break down predictive LTV modeling, the critical flaws in how teams measure unit economics, and why today's mobile marketers need unified tools that connect profitability analysis across channels. The conversation covers what's broken in traditional LTV reporting, the technical pain points of fragmented data definitions across platforms, and how accurate cohort analysis can unlock smarter budget allocation.

• What is Campaignswell, and what problem is it solving for mobile marketers?
• What is wrong with traditional LTV reporting?
• What exactly is predictive LTV and how far out can you forecast?
• Which monetization models are easiest versus hardest to predict LTV?
• When teams estimate their own LTV, how accurate are they usually?
• What immediate changes can marketers make if their LTV is poorly defined?
• How does Campaignswell guide budget allocation across different channels?
• What is the elevator pitch for Campaignswell to get teams to adopt it?
• Why is cohort analytics misunderstood by most marketing teams?
• How should marketers think about payback periods when measuring campaign efficiency?

• (0:26) — What Campaignswell is and what problem it solves for marketers
• (1:24) — Building Campaignswell as a single source across all teams
• (5:20) — Why speed matters in campaign decisions
• (7:50) — The hidden costs in LTV
• (9:47) — Predictive LTV and calculating on specific horizons
• (11:19) — Why subscription monetization is easiest to predict
• (16:19) — Client LTV predictions: When teams' numbers are off by 2x or more
• (20:27) — Matching optimization targets to the right LTV metrics across channels
• (26:22) — Why cohort analytics is misunderstood by most marketing teams
• (28:20) — Lightning round: First thing every morning
• (29:49) — Closing: Where Artsiom wants to travel next

• (8:07) "Apple takes 30 percent of your revenue by default. It's really huge. It might be all your margin. And if you cannot calculate it, your comparison with customer acquisition cost might be wrong.”
• (25:20) “Using Campaignswell, you won't be in a situation where one team says they have a CPA of $20 bucks and another team says it’s $30.”
• (29:14) "I noticed that marketing spend was $600,000 per day with really strong performance. ROI was something around 30 percent. So it's a really huge amount of marketing budget. The first thing I thought was there’s probably something wrong."

• Campaignswell
• Artisom on Linkedin