This transcript documents a meeting at the start of summer where a team discusses their data reporting infrastructure and tools. The core group includes regular attendees, though some members may join intermittently based on availability. The team acknowledges that their data comes from multiple sources of varying quality, including outdated SSIS jobs that require careful handling to avoid consuming manipulated data.
The discussion centers on Guice, a reporting engine similar to Power BI that enables product team members (primarily Jacob) to build client-facing reports through a drag-and-drop interface without engineering involvement. However, Guice has significant limitations—most notably, it cannot perform database joins, requiring the team to consolidate multi-source data in their data warehouse reporting database (DWRPT) at the SQL level before presenting it to Guice.
The team is not committed to Guice long-term and is already exploring alternatives. While Guice itself won't be accessible to new team members, access to the underlying data warehouse reporting database—which contains all tables and views powering the analytics—can be provided. The speaker demonstrates a sample Guice report from their 3Birds/Dash CDXP site, showing how users can filter data by various parameters like campaign methods and communication channels.
Core team established for the project A core group of team members identified, with some flexibility for others to join based on availability
Data quality issues from multiple sources Reporting data comes from multiple sources with varying quality; some SSIS jobs are outdated and not ideal for CDP consumption
Raw data preferred over manipulated data Preference to consume raw data from SSIS processes rather than end results to allow selective use of data pieces
Guice is a reporting engine, not a long-term priority Guice functions like Power BI for drag-and-drop report building; not critical for long-term roadmap and alternatives are being explored
Guice has technical limitations with joins Guice cannot perform joins across multiple data sources; data must be joined at SQL level in DWRPT before presenting to Guice
Data quality improvements implemented Common Client ID implemented to improve table joins; discovered data cleanliness issues
Guice reports are client-facing Reports generated in Guice are used for client-facing purposes; built by internal product team (primarily Jacob), non-technical users
Data warehouse contains multiple schemas Data organized by application schemas (e.g., SL for Social Logic, ML for MediaLogic) to prevent data conflation
Guice embedded via iframe with security controls Guice reports embedded in company portal using iframe with login requirements and specific security rules
Power BI available but resource constraints Power BI available but lack of internal Power BI developers; offshore team utilized to avoid taking engineering resources
graph LR
Ron["Ron"]:::person
CDXP["CDXP"]:::technology
Guice["Guice"]:::technology
Javis["Javis"]:::technology
Views["Views"]:::concept
DWRPT["DWRPT"]:::technology
Reports["Reports"]:::concept
AWS["AWS"]:::technology
Power_BI["Power BI"]:::technology
SSIS["SSIS"]:::technology
data_warehouse["data warehouse"]:::technology
reporting["reporting"]:::concept
Negative_View_Analysis["Negative View Analysis"]:::technology
Jarvis["Jarvis"]:::technology
CIM["CIM"]:::technology
historical_data["historical data"]:::concept
iframe["iframe"]:::technology
Portal["Portal"]:::concept
Juice["Juice"]:::technology
ADF["ADF"]:::technology
AI["AI"]:::concept
vehicle["vehicle"]:::concept
service_dates["service dates"]:::concept
Tables["Tables"]:::concept
usage_patterns["usage patterns"]:::concept
Ron -- "can provide information about" --> reporting
SSIS -- "provides data to" --> reporting
Guice -- "is similar to" --> Power_BI
Ron -- "has knowledge of" --> Guice
DWRPT -- "feeds data to" --> Guice
Guice -- "generates" --> Reports
Guice -- "similar functionality to" --> Power_BI
Reports -- "contains" --> Views
Portal -- "embeds" --> iframe
Ron -- "assigned to work on" --> Power_BI
CDXP -- "hosted in" --> AWS
SSIS -- "manages data flow for" --> CDXP
Juice -- "consumes data from" --> Views
Javis -- "built over" --> AI
DWRPT -- "feeds data to" --> AI
Javis -- "technically sits on" --> AWS
Jarvis -- "related to data migrations in" --> AWS
Ron -- "has access to/involved with" --> Jarvis
CDXP -- "tracks" --> vehicle
reporting -- "has problems originating from" --> CDXP
vehicle -- "has associated" --> service_dates
CIM -- "holds" --> historical_data
Ron -- "makes point about not actively monitoring" --> data_warehouse
usage_patterns -- "describes traffic on" --> data_warehouse
classDef person fill:#f9d5e5,stroke:#333,stroke-width:1px
classDef concept fill:#eeeeee,stroke:#333,stroke-width:1px
classDef diagram fill:#d5f9e5,stroke:#333,stroke-width:1px
classDef time fill:#e5d5f9,stroke:#333,stroke-width:1px