Data Model, Segmentation, and Hygiene
Core Concepts
The data model is the foundation of Marketo. Architects must understand standard objects (Lead, Company, Opportunity, etc.), custom objects, field management, data types, and relationships. Segmentation allows grouping leads based on shared traits (e.g., industry, region) for dynamic content and reporting. Data hygiene involves deduplication, normalization, and validation (e.g., email verification, picklist management). Key is designing a data model that supports business processes without unnecessary complexity.
Common Traps
- Overusing custom objects: Custom objects add complexity and can impact performance; use only when standard objects don't suffice.
- Poor field naming: Inconsistent naming (e.g., "Industry" vs. "industry") causes confusion and reporting errors.
- Neglecting data normalization: Free-text fields lead to segmentation issues; use picklists and standardization.
- Ignoring deduplication: Duplicate leads skew scoring and reporting; implement deduplication rules.
- Misunderstanding segmentation vs. smart lists: Segmentation is for dynamic content and is limited in number; smart lists are for targeting.
Worked Example
A company wants to segment leads by industry for email personalization. Trap: using a free-text industry field. Better: create a picklist field with standardized values, then create a segmentation based on that field. For data hygiene, implement a smart campaign to standardize country values and flag duplicates. Rule of thumb: Segment on fields that are standardized and stable; avoid segmenting on frequently changing fields.
Rule of Thumb
Design fields with clear naming conventions and use picklists for categorical data; limit segmentations to 20 per instance.