Experience — Senan Mele

Experiences

Direct the annual forecasting cycle from source-data acquisition through modeling, market allocation, quality assurance, and stakeholder communication, producing national and market-level outlooks that inform business planning across the local advertising ecosystem.
Set the analytical architecture and methodological standards for forecasts covering television, radio, cable, search, social, digital display, digital video, CTV/OTT, out-of-home, direct mail, and other local media, ensuring that projections are consistent across markets, channels, industries, and forecast years.
Design and refine forecasting models using hierarchical Bayesian frameworks, ARIMA/ARIMAX, regression, time-series, allocation, and scenario-based methods, improving forecast stability while separating macroeconomic, structural, cyclical, market-specific, and event-driven effects.
Lead the modernization of BIA’s forecasting infrastructure, migrating fragmented inputs and outputs into a governed, multi-schema database environment that improves data lineage, reproducibility, accessibility, and the reliability of recurring production cycles.
Build and automate modular ETL pipelines in R, Python, and SQL to ingest, standardize, reconcile, and validate large datasets from Nielsen, AdImpact, the FEC, government economic sources, industry vendors, company filings, internal systems, and market-level reference data, reducing manual processing and accelerating the integration of new information.
Establish automated QA, reconciliation, exception-reporting, and model-monitoring processes across historical data, market allocations, station-level estimates, and forecast outputs, identifying material anomalies before publication and strengthening confidence in client-facing data.
Develop scalable methodologies to estimate and reconcile core and political advertising by market, media, category, and year, enabling stakeholders to distinguish underlying market performance from election-cycle volatility and better understand year-over-year changes.
Translate model results into executive-ready analysis on market growth, channel shifts, competitive positioning, advertising-category trends, and revenue opportunities, supporting senior-leadership decisions, lender presentations, webinars, published research, sales efforts, and high-stakes client conversations.
Lead custom analytical engagements including share-of-wallet, station benchmarking, market dynamics, category opportunity, and competitive-performance studies that connect BIA’s broader forecasting intelligence to specific client strategies and commercial opportunities.
Partner with research, product, content, and market-intelligence leadership to evaluate emerging trends, align forecasting assumptions, explain revisions, and operationalize methodological changes across BIA’s products and production systems.
Guide analysts and engineering resources supporting data inputs, factors tables, database processes, model execution, and forecast refreshes, establishing clearer documentation, ownership, and operating standards for a more scalable forecasting function.

Optimized the allocation of media spend across multiple channels and publishers by assessing the impact of paid TV, Search, Social, and Digital media campaigns and creative elements on conversions using Multiple Linear and Bayesian regression modeling.
Conducted Incrementality Analysis to measure the potential impact of additional media channels before larger-scale implementation using statistical techniques to select proper test and control markets, to ensure adequate statistical power, and to run a counterfactual analysis.
Ensured data relevance and accuracy by working closely with the business science and data engineering teams to use the right metrics for data analysis and verify that those metrics trend as expected.
Created a robust foundation for advanced analysis by developing R and Python scripts in collaboration with the data engineering and business science teams to analyze ad campaigns performance and support clients’ goals.
Supported junior members of the team by training them on how to run existing and ad hoc analyses requested by clients and our business science teams.

Enhanced data integrity and accuracy by setting up and managing an Amazon Redshift database in partnership with the data engineering team and getting data feeds implemented – pulling in data from multiple data sources for our clients.
Reduced hours of analytics work to seconds by using R and Python to build data products and tools for the team, which helped streamline the selection of test and control markets, run causal impact analyses, build multi-touch attribution models, and run multiple linear regressions to evaluate impact of media campaigns and creatives on performance.
Revamped the reporting process by building insightful Tableau and Power BI dashboards using tables and views from the Amazon Redshift database, freeing up the team’s time as they would no longer need to spend dozens of hours each week preparing the data and building the report in Microsoft Excel.
Ensured data availability by working with the strategy and data engineering teams, as well as clients and data partners to agree on the data fields needed, ways to capture and pull them into the database (API, FTP, Amazon S3, and/or Excel File Sharing) as well as cadence of data refreshes.
Enabled seamless deployment and accessibility by containerizing web applications with Docker and orchestrating their publication using Kubernetes, ensuring the Business Strategy team could efficiently access and utilize them.
Conducted training sessions and facilitated knowledge sharing to increase the team’s skills set and efficiency by creating various PowerPoints and documentations on programs such as MS Excel, Knime, Tableau, and statistical models.

Saved the company over $100,000 annually by developing a full-service analytics infrastructure using R, SQL, and Power BI, eliminating the need for costly third-party software.
Created interactive Power BI reporting enabling seamless data flow from source to reporting by designing, implementing, and maintaining end-to-end data pipelines—developing R scripts for API data extraction, and loading data into SQL Server.
Established data interface for clients to pull reports through Plumber package by developing a secure web API using R to facilitate data sharing from MHE's database.
Improved advertising campaign performance by analyzing key performance indicators (KPIs) in social media post impressions and identifying optimization opportunities.
Ensured seamless project execution and success by leading cross-functional collaboration between internal teams and external business partners.
Drove successful business outcomes for clients by spearheading and managing multiple projects in a fast-paced, results-oriented environment. Carrying an average of 3 clients each week.

Increased media effectiveness for $2M client campaigns by analyzing both campaign performance metrics and brand lift study results and providing strategic recommendations to planning and activation teams.
Aligned global media measurement strategies with regional goals by collaborating with EA teams across NA, LATAM, EU, and APAC to understand their objectives, coordinate brand lift study setup, and tailor reporting to drive region-specific insights and performance improvements.
Conducted market research to improve brand health and consumer behavior measurement by collaborating with data partners (Kantar, Facebook, Twitter, Google, Snapchat, Twitch) to identify and implement relevant studies such as Context Lab Studies and larger scale Brand Lift Studies depending on client goals.
Trained junior team members in Excel and Power Query, significantly reducing the number of hours needed to prepare the data and report on campaign performance learnings.

Enhanced team expertise and efficiency by training direct reports in R’s advanced statistical capabilities and SQL, equipping them with the skills to analyze campaign performance, troubleshoot reporting issues, and develop data-driven solutions.
Optimized cross-channel performance tracking by automating monthly reporting for TV, Print, Search, Social, Programmatic, and Digital channels, leveraging Amazon Redshift SQL views and Tableau to reduce manual effort and enhance data accessibility.
Implemented a structured QA process across multiple sources, while proactively resolving inconsistencies by diagnosing reporting and API issues. Collaborated with data providers to standardize data taxonomy and integration.
Identified inefficient conversion campaigns where shifting dollar spend to a more efficient campaign drove 35% more conversions.

Saved 600+ hours of labor annually by automating repetitive tasks with R, Macros, and Excel VBA, accelerating data transformation and analysis for media teams eliminating time consuming Excel reporting.
Enhanced attribution modeling accuracy by implementing a Markov Chain-based approach, shifting from heuristic last-touch attribution to a probabilistic model that better mapped customer journeys.
Strengthened media performance measurement by analyzing key KPIs (ratings, impressions, unique visits, and spend) and delivering year-over-year TV network trend reports to support media planning.
Improved proprietary analytics tools by testing updates, identifying gaps, and driving enhancements to optimize functionality and usability.

Optimized media cost analysis and campaign spend tracking by automating weekly dynamic cost deliverables for cross-channel attribution and conducting quarter-over-quarter and year-over-year analyses of GRPs, spend, and impressions, leading to more efficient budget allocation and performance insights.
Streamlined data ingestion and ensured accurate predictive analysis for AT&T’s Marketing Mix Modeling (MMM) by using Microsoft Power Query to process measurement feeds, improving data integration and reliability.
Enhanced competitive intelligence and media planning by compiling engagement reports on AT&T and its competitors’ social media presence, providing actionable insights that informed strategic decisions.
Increased efficiency in media cost analysis and reporting by automating dynamic cost computations and streamlining Local Cable TV spend calculations, significantly reducing manual effort and enhancing data accuracy.

Your browser is outdated!

Senan Mele

Data Scientist, Advanced Analytics Leader

VP, Forecasting and Data Analysis

Manager, Advanced Analytics Modeling

Director, Analytics

Director, Data Performance

Manager, Business Insights

Supervisor, Data & Analytics

Senior Analyst, Performance Insights

Analyst, Data Integration