Logo-black

Scaling multi-touch attribution to optimize pharmaceutical marketing impact

  • Healthcare & Life Sciences

Scaling multi-touch attribution to optimize pharmaceutical marketing impact

OneSix developed a scalable multi-touch attribution solution for a biopharmaceutical company, enabling precise measurement of marketing impact across channels, optimizing budget allocation, and accelerating data-driven insights for increased healthcare provider engagement.
AI & Machine Learning
AI-Driven Marketing

Overview

Improving multi-touch attribution for targeted biopharma marketing

A leading biopharmaceutical company, known for its breakthroughs in innovative treatments, sought to improve its understanding of multi-channel marketing impacts on healthcare providers (HCPs), specifically in driving new-to-brand prescriptions (NBRx). With a vast marketing ecosystem, the company employed multiple touchpoints—including email, digital ads, and in-person events—to reach providers across various stages of the decision journey.

Although they had a proof-of-concept model for multi-touch attribution (MTA), it needed to be scaled and fine-tuned to operate effectively in a production environment. Additionally, the company needed a parameterized solution capable of segmenting MTA results by brand, franchise, and indication. The ultimate objective was to develop a robust and flexible MTA model that could accurately attribute marketing impact and optimize budget allocation to maximize engagement with HCPs.

Effective multi-channel marketing in the pharmaceutical industry is challenging, as each channel and publisher varies in its reach, engagement, and effectiveness. Unlike a single-channel approach, multi-touch attribution must capture how touchpoints interact within complex user journeys. An ideal solution would involve controlled experiments to precisely isolate channel impacts; however, the cost and frequency requirements of such experiments make them impractical for real-world applications. The client needed a more scalable approach that leveraged existing data to measure past performance and generate actionable insights for future marketing decisions.

Our Solution

Designing a scalable and adaptive MTA pipeline

OneSix built a highly parameterized, unit-tested Python package to perform MTA on the client’s diverse marketing initiatives, focusing on measuring individual touchpoint effectiveness across brands and indications. The model’s core function was to predict the probability of an NBRx occurring, based on a combination of control and independent variables derived from the various marketing channels. To further refine the model, OneSix introduced an advanced explainer model that could assign a partial contribution to each control and independent variable, providing a breakdown of the factors driving NBRx outcomes.

The MTA model was designed to address key technical challenges, including calibration to adjust for the sigmoid distortion often seen in probability densities from predictive models. This adjustment was achieved through a custom calibration scheme, which corrected probability distortions to ensure that all variables received a positive partial contribution. The parameterized structure of the model allowed users to modify factors such as study period lengths, feature sets, and segment parameters (e.g., brand or indication) with ease. The package was controlled by a single configuration file, providing a centralized interface for rapid experimentation and model adjustments via a command-line interface.

As a result, the pipeline offered flexibility for experimentation across different market baskets and feature combinations, empowering the client’s data science team to iterate quickly and test various configurations. By providing a modular, scalable, and flexible solution, OneSix’s MTA model allowed for high adaptability, enabling the client to execute MTA analyses on demand and derive actionable insights at a pace previously not possible.

Results

Accelerated data-driven insights and improved marketing allocation

The implementation of this comprehensive MTA pipeline enabled the client to gain a deeper understanding of how different marketing touchpoints contributed to NBRx conversions and overall engagement with HCPs. With OneSix’s solution in place, the company was able to assess the individual and combined impacts of each marketing channel, allowing them to identify high-performing channels and optimize spend allocation with confidence. By analyzing the contributions of different touchpoints within the customer journey, the company could now tailor its marketing strategies to maximize engagement and ROI on specific channels.

The streamlined configuration and command-line interface allowed the client’s data science team to rapidly test hypotheses and iterate on model features, reducing the research cycle and enhancing their agility in responding to market dynamics. Continued collaboration with OneSix provided the company with regular updates and enhancements to the MTA model, enabling ongoing improvements and refinements to its methodology. As a result, the biopharmaceutical company was able to achieve a more precise, data-driven approach to marketing attribution, laying a scalable foundation for sustainable growth and optimized channel investment across its brands and franchises.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Building a forecasting engine and media mix modeling pipeline for a FinTech firm

  • Financial Services

Building a forecasting engine and media mix modeling pipeline for a FinTech firm

OneSix implemented machine learning models for a financial services client to optimize marketing spend, resulting in a strategic reallocation that improved ROAS from 0.5-0.8x to 1.5x and achieved the client’s first quarter of positive marketing ROI.
Data Science
AI & Machine Learning
AI-Driven Marketing

Overview

Improving marketing efficiency in a fast-growing financial services firm

A rapidly expanding firm in the consumer financial services industry, offering both traditional and cryptocurrency brokerage solutions, faced challenges with low and declining Return on Ad Spend (ROAS), estimated at 0.5-0.8x.

Despite impressive growth driven by a surge in interest in stock and alternative assets since 2020, the company’s marketing spend had consistently outpaced revenue. OneSix was tasked with implementing machine learning and data science solutions to enhance marketing efficiency by accurately measuring spend effectiveness and building an automated pipeline for optimized media allocation.

Our Solution

Building a marketing efficiency and optimization platform

To tackle the client’s challenge, OneSix developed two custom models designed to provide insights into current marketing spend efficiency and inform future optimization strategies:

LifeTime Value (LTV) Model

OneSix created a predictive LTV model capable of forecasting each newly acquired user’s value within 12 hours of signup. This model offered near real-time insights into customer acquisition health by forecasting future revenue over multiple time horizons for users and cohorts. Integrating this model with direct attribution data from the client’s Mobile Measurement Provider (MMP) and custom attribution logic enabled precise calculations of Customer Acquisition Costs (CAC) at both user and cohort levels. The model decomposed LTV predictions into key metrics like time-to-convert, time-to-churn, subscription revenue, and non-subscription revenue. This breakdown highlighted specific channel performance issues, revealing, for instance, that some channels suffered from retention issues while others had low conversion rates.

Media Mix Model (MMM)

OneSix also developed a Media Mix Model (MMM) that used historical LTV estimates and spend data to calculate the LTV/Spend (ROAS) ratio for each marketing channel. The MMM accounted for marketing and non-marketing factors, including market sentiment, seasonality, holidays, and product/pricing changes. By optimizing for aggregate forecasted LTV rather than short-term metrics like new users or first-month revenue, the model avoided common pitfalls and focused on maximizing long-term marketing ROI. Both models were deployed and automated using Flyte on Kubernetes, enabling weekly retraining with fresh data and pushing results to a data lake for real-time reporting.

Our marketing/media mix model predicted total aggregate LTV acquired on a daily basis through a combination of marketing and non-marketing drivers.

Results

Improvement in ROAS and a strategic pivot in marketing allocation

The new insights provided by these models led to a major shift in marketing spend allocation. The analysis revealed that certain channels previously deemed effective were attracting low-value, high-churn customers, while others seen as saturated actually delivered higher customer value.

Following MMM’s spend recommendations, the client projected an increase in ROAS from 0.5-0.8x to 1.4x. Within two months of adopting the optimized spending recommendations, the client achieved a 1.5x ROAS, doubling historical returns and achieving their first quarter of positive marketing ROI. This marked the beginning of a new era of accelerated growth and customer value for the company.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Optimizing personalized marketing campaigns at scale in real estate

  • Technology

Optimizing personalized marketing campaigns at scale in real estate

OneSix developed a personalized, data-driven marketing platform for a real estate client, driving increased revenue, customer engagement, and establishing a scalable foundation for long-term, automated personalization.
Data Science
AI & Machine Learning
AI-Driven Marketing

Overview

Adapting marketing strategies to achieve personalized engagement

Our client, a leading consumer-facing real estate company, had achieved substantial market penetration with near-total product awareness. However, this market saturation led to diminishing returns from traditional, broad-based marketing campaigns. A one-size-fits-all approach no longer captured customer attention effectively, nor did it foster meaningful engagement.

To drive continued growth, the company needed to transition to a highly personalized, data-driven marketing strategy. However, several obstacles stood in the way: the company lacked the infrastructure to analyze customer data effectively, tailor outreach strategies based on customer behaviors, and automate campaigns at scale. Additionally, they needed a reliable way to measure campaign impact to ensure continuous optimization based on real-world performance.

Our Solution

Building a scalable marketing platform for automated personalization

To address these challenges, OneSix partnered with the client to design and implement a production-grade marketing platform that would support scalable, automated, personalized campaigns. This platform was engineered to ingest real-time customer data across various business lines, using insights from behavioral data to drive micro-targeted marketing efforts. Key elements of the solution included:

  • Automated Micro-targeted Campaigns: The custom platform empowered the client to run personalized, data-driven campaigns by automating the ingestion and processing of user-level behavioral data. This allowed marketing teams to create campaigns tailored to individual preferences, targeting customers with high relevance across the best-fit channels.
  • Incrementality Measurement: The platform incorporated robust A/B testing to rigorously measure the incrementality of campaigns across audience segments, content types, and distribution channels. By comparing control groups and treatment groups, the client could accurately gauge the impact of campaigns, optimizing messaging and channel strategies based on real-world data.
  • Strategic Campaign Verticalization: We developed distinct campaign “verticals,” each focused on different customer segments within specific business lines. This approach allowed us to deploy highly specific campaigns with unique creative content and messaging, driving relevance and engagement across diverse segments.
  • Audience Segmentation and Experimentation: The platform automatically segmented the client’s customer base according to affinity with specific campaign verticals, using autonomous experimentation to test various content types and channels. The platform maintained matched control groups for each action, fueling ongoing analysis and optimization.
  • Next-Best-Action Optimization: Leveraging uplift modeling and next-best-action recommendations, we analyzed experimental results to understand the incremental impact of campaigns across different audience segments. This enabled the client to optimize future campaigns with content and channels proven to resonate with each segment.

Results

Increased revenue and customer engagement

The implementation of a highly targeted, personalized marketing platform delivered significant business impact for the client. The integrated approach of automation, real-time data analytics, and strategic segmentation resulted in millions of dollars in incremental revenue. Customer engagement improved substantially, with personalized messaging leading to higher rates of acquisition, retention, and loyalty.

The project established an evergreen marketing framework that will continue to serve the client well into the future. Insights and best practices gained from this engagement are now embedded across the organization, providing a scalable foundation for future personalization efforts. The client now possesses the tools to continuously adapt to evolving customer preferences and market trends, positioning them for sustained growth and competitive advantage.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Enhancing a visual crop detection model for an agricultural firm

  • Agriculture

Enhancing a visual crop detection model for an agricultural firm

OneSix developed a high-speed, accurate crop detection model for an agricultural firm’s autonomous weeding robots, enhancing operational efficiency and sustainability through adaptable, AI-driven solutions for multi-crop environments.
AI & Machine Learning
Computer Vision

Overview

Improving accuracy in crop detection for autonomous robots

A major agricultural firm, a leader in innovative farming technologies, sought OneSix’s expertise to enhance its autonomous robotic weeding solution. The company’s robots, designed to selectively remove weeds while preserving valuable crops, required a highly accurate vision system that could function effectively on low-powered edge hardware in the field. While the initial crop detection model could distinguish between crops and weeds, it struggled to meet performance demands, particularly in terms of speed and accuracy. This created limitations in the field, where both real-time processing and accuracy are critical for efficient operations and reducing reliance on costly manual labor and environmentally harmful pesticides.

Our Solution

Developing an advanced ML model optimized for edge processing

OneSix collaborated closely with the agricultural firm’s internal teams to understand the unique challenges and nuances of autonomous weeding, particularly the need to balance high-speed processing with accuracy on edge devices.

To address these requirements, we designed an optimized machine learning model that leverages state-of-the-art techniques in object detection, enabling it to run quickly on edge hardware without sacrificing accuracy. Through careful tuning, we achieved a model that could process images faster than real-time and maintain, or even exceed, the detection accuracy of the original model.

In addition to building the core model, OneSix broadened the project’s scope to introduce additional features that further increased the robustness and versatility of the robotic weeding solution:

  • Multi-Crop Detection: The team transitioned from a single-crop detection model to a multi-crop detection framework, allowing the same model to be deployed across fields with different crops. This innovation simplifies deployment for end users by enabling robots to operate in diverse environments without needing separate configurations for each crop type. As a result, operators can now deploy the robots across fields, regardless of the crop variety, significantly streamlining operations and reducing setup times.
  • Unannotated Image Search: To optimize training, OneSix implemented an unannotated image search feature, allowing the model to automatically explore large datasets of untagged images. This feature enabled the identification of complex crop and weed combinations, which helps improve detection accuracy by training on a wider variety of real-world data, all without requiring extensive manual annotation efforts.
  • Generative Image Training: Adapting machine learning models to the variable conditions of agricultural environments—such as differences in lighting, soil texture, and camera angles—was essential for real-world deployment. To meet this challenge, OneSix introduced generative image training using advanced AI-driven image translation techniques. This approach simulates a variety of field conditions in a controlled setting, producing synthetic images that allow the model to learn from diverse scenarios without needing new data collection. By leveraging generative AI, the model could continuously improve its adaptability, providing reliable performance in various field conditions.

Results

Increased speed, accuracy, and adaptability

The optimized crop detection model delivered substantial improvements in both speed and accuracy, meeting the demands of real-time, edge-based processing. The integration of multi-crop detection, unannotated image search, and generative image training provided significant benefits, allowing the autonomous robots to perform reliably across different crops and environments. By automating complex weeding tasks, the model enables the agricultural firm to reduce manual labor costs and minimize pesticide use, supporting sustainable farming practices.

With this advanced, adaptable machine learning platform, the agricultural firm is now equipped to deploy its robotic weeding solution at scale, enhancing operational efficiency and promoting environmentally friendly practices in the agricultural sector. The success of this project highlights the transformative potential of machine learning in addressing complex agricultural challenges, showcasing how AI-driven innovation can drive meaningful impact in real-world applications.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Optimizing restaurant inventory using demand forecasting for Relish Works

  • Travel & Hospitality

Optimizing restaurant inventory using demand forecasting for Relish Works

OneSix developed a machine learning platform for a food distributor, enabling restaurants to optimize ingredient ordering, reduce waste, and enhance profitability through data-driven inventory management.
Data Science
AI & Machine Learning
Forecasting & Prediction

Overview

Maximizing restaurant profitability by optimizing ingredient ordering

A leading food distributor sought OneSix’s expertise to create a data-driven platform that could support restaurants in managing ingredient inventory more efficiently. While many restaurants relied on manual processes and intuition to determine restocking needs, the client wanted to introduce a solution that would allow for optimal ordering strategies to balance the risks of missed demand and food wastage.

Our Solution

Building an ML platform for dynamic ingredient forecasting

OneSix developed a machine learning-powered tool to generate ingredient-level forecasts based on diners’ orders, point-of-sale data, and menu items. The platform used these forecasts to predict future inventory needs, providing optimal restocking recommendations.

By factoring in each ingredient’s shelf life, seasonality, demand patterns, and role in various menu items, the platform automated order generation, allowing restaurant operators to submit optimized orders with just two clicks. Additionally, the tool aggregated ingredient and dish-level trends, offering insights into seasonal shifts and evolving consumer preferences to further guide menu planning.

Results

Enhanced profitability through data-driven inventory management

The machine learning solution enabled restaurants to make informed restocking decisions, minimizing waste while ensuring ingredient availability for high-margin items. By automating inventory management, the platform significantly reduced manual efforts and improved restaurant profitability. The aggregated trend insights also empowered operators to adapt to seasonal preferences and make data-backed decisions in menu engineering, providing ongoing value to restaurant businesses.

“OneSix partnered with us to design a new, machine learning-powered tool for restaurant operators to manage their business. They guided us through the process of framing the problem and determining what was possible with state-of-the-art ML/AI. The solution they developed transformed a key operational challenge into an automated solution that drives new value for restaurants.”

Mandy Tahvonen, Managing Director

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Keeping children safe online with machine learning

  • Technology

Keeping children safe online with machine learning

OneSix developed a scalable, adaptive machine learning platform to detect abusive content in real time, enabling a tech company to enhance online safety for children while continuously adapting to their evolving communication styles.
Data Science
AI & Machine Learning
AI Agents & Chatbots
Forecasting & Prediction

Overview

Protecting children from online abuse

As social media use continues to grow, children are increasingly exposed to potentially harmful content. OneSix partnered with a fast-growing technology company to address this challenge, aiming to monitor and detect abusive content effectively while respecting the unique style and nuances of children’s online communication. This required developing a solution that could handle the vast scale of daily content, including millions of interactions, and adapt to the evolving ways children communicate online.

Our Solution

Designing a scalable, adaptive machine learning platform

OneSix built a custom, end-to-end platform with multiple machine learning models specifically trained to recognize abusive language, emojis, and media in children’s online communication. The solution included independent autoscaling pipelines for triaging content types, analyzing web and media content, and processing natural language text to detect abuse as it occurs.

To adapt to the unique and changing nature of children’s online language, the platform included a real-time feedback loop for continuous model updates. This loop enabled strategic selection of content for expert annotation, feeding those insights back into model training to keep pace with evolving trends. Additionally, a suite of performance monitoring tools was developed to track model accuracy and responsiveness, ensuring both effective monitoring and high-quality data for ongoing improvements.

Results

Enhanced online safety through dynamic abusive content detection

The solution has enabled near real-time identification of abusive content across millions of daily social media interactions, providing a powerful safeguard against cyberbullying and harmful language exposure. The adaptive model and feedback loop ensure that the platform remains effective as communication styles shift, delivering timely and accurate detection to keep children safe online. The collaboration with OneSix has equipped the client with a robust, scalable system that continuously learns and improves, meeting the demands of online safety in an ever-changing digital landscape.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us

Leveraging AI to increase provider engagement for a top pharmaceutical company

  • Healthcare & Life Sciences

Leveraging AI to increase provider engagement for a top pharmaceutical company

OneSix developed an AI-driven, multi-channel outreach engine for a global pharmaceutical client, enabling automated, personalized provider engagement across brand portfolios with scalable, data-informed insights.
AI & Machine Learning
AI-Driven Marketing

Overview

Leveraging AI for a coordinated, personalized marketing strategy

Our client, a top 5 global pharmaceutical company, sought to build and deploy an AI-powered multi-channel outreach engine to coordinate provider engagement across multiple brand portfolios. Their manual engagement process lacked the ability to personalize outreach at the individual provider level, relying instead on aggregate customer segments. This approach created substantial inefficiencies, limited scalability, and missed opportunities to tailor interactions based on individual provider traits and behaviors.

Our Solution

Developing an AI-driven multi-channel outreach engine

OneSix built a custom AI solution leveraging enterprise reinforcement learning technology to automate and optimize provider engagement. Using real-time data on provider traits, behaviors, and patient populations, the solution integrates multi-brand and multi-channel outreach into a single system. The platform, powered by Strong RL, Apache Spark, and Tensorflow, is designed to scale within the client’s on-premise infrastructure to accommodate vast data volumes and diverse brand requirements.

Results

Enhanced efficiency and personalized engagement at scale

With this AI-driven outreach engine, our client achieved a highly efficient, data-driven provider engagement strategy, enabling coordinated, personalized interactions across brands and channels. The system improved engagement outcomes by predicting provider response probabilities and suggesting targeted interventions. As a result, the pharmaceutical company can engage providers in a more tailored, impactful way, leveraging individual insights at scale and significantly reducing manual efforts.

Ready to unlock the full potential of data and AI?

Book a free consultation to learn how OneSix can help drive meaningful business outcomes.

Contact Us