Merkle’s Data Optimization Lab allows the marketer to make an informed purchase by understanding the value and ROI associated with data through three key components:
Source Optimization
- Dry List Testing - Analytic approach to determine whether a list or audience will perform prior to deploying a campaign
- Data Evaluation - Modular process for evaluating the utility of data across four dimensions: source quality, descriptive power, predictive power and universe reach/expansion
Data Optimization
- Derived/composite variables - Merkle employs advanced analytics to create new industry-specific and client-specific variables to predict response, profitability, interest, or intent by combining existing attributes from the base data package.
- Data augmentation/white space infill - Gap analysis and enrichment of contact information and demographic/firmographic/behavioral data variables
Optimized Targeting
- LSO (List Source Optimization) - Scientific way to drive optimal data purchase decisions by leveraging Advanced Statistical and Optimization Techniques
- Lead Generation & Scoring - Custom processes to identify, evaluate, and develop high quality sales leads
- Custom Audiences - Merkle can identify groups of individuals (delivered as mailing lists, email lists, online cookie pools, etc) that fit a pre-defined targeting criterion
We’ve created and maintain high-performing propensity variables for multiple industries and can rapidly create custom industry and client specific variables combining the syndicated Simmons survey and DataSourceTM . Other than derived data, we also create custom online audiences for high-conversion display targeting by using campaign history and combine with CRM data to target based on higher-level metrics such as lifetime value and profitability.
Figure A: Modeled Audiences
The Merkle Content Lab is an infrastructure that combines robust quantitative methodologies with best-practice operational database marketing procedures into one synergistic data content optimization platform. Merkle defines data content as data that is determined to have predictive, descriptive, or business value and drives improved marketing results. Under this context, Merkle created the Content Lab Environment, which leverages our industry-recognized leadership in analytical discipline and database operations to continuously evaluate and optimize our clients’ marketing universe. This approach to marketing universe optimization is unique to the marketplace, as it employs rigorous quantitative methodologies to the evaluation and integration of data content.
Figure B: Merkle Content Lab
The Content Lab Environment consists of two major components: Research and Development and Content Management. Data content enters the Merkle Content Lab in the Research and Development component where Content Lab Applications execute a series of analytical procedures that extract quantitative insights about the data. These insights then inform the Content Management component of the Merkle Content Lab, which updates and manages the business rules and database integration logic to optimize the Merkle Marketing Universe. This universe then feeds campaign production and analytics teams. The Content Management component integrates learnings and marketing results from production and analytics back into the Research and Development component, and the cycle of data optimization through the Content Lab Environment begins again. The Content Lab Environment thereby ensures that the optimal mix of data content exists in the Merkle Marketing Universe.
Central to the Merkle Content Lab is a unique feature-set called the Content Lab Analytical Engine. The Analytical Engine consists of four key modules. Each module is a suite of rigorous analytic procedures that ensure data evaluations are conducted in a consistent and unbiased fashion.
The methodologies employed in each module quantify the impact a data asset has along a specific dimension. These dimensions are of particular interest to database marketers and include assessments of Predictive Power, Universe Expansion, Source Quality, and Descriptive Power.
Analysts can leverage a single module for a focused analysis along the dimension of interest, or leverage Content Lab applications to combine the outputs across any combination of the four dimensions to arrive at a composite score. Data content is passed to the modules in the Content Lab Analytical Engine after it has run through the KnowledgeLink CDI engine.
Importantly, the modules in the Content Lab Analytic Engine are fueled by several libraries in the Research and Development component:
- Predictive Model Bank, consisting of predictive targeting and segmentation models
- Sample Campaign Repository, which stores sample campaign response data
- Metadata and Business Rules
- Marketing Universe Sample Extract
Each of these libraries is continuously updated from the Content Management component of the Content Lab Environment, as models are developed or refined, campaigns are executed, and offer, promotion, and campaign history is captured.
Merkle has refined this process over time to determine the best sources of data in the marketplace driven by our analytical evaluation process. This enables Merkle to provide the best source of data for each client’s demands.
