Consolidation is done manually. Between business units and controlling there are discrepancies in the definition and value of the same metric.
Solution area 02 · Data foundation and business intelligence
Single Point of Truth: a consolidated data foundation for AI & reporting
Consolidating distributed data sources, building a single point of truth and modernizing reporting. Establishing the technical and organizational basis on which strategy and AI initiatives can reliably build.
Typical triggers
Three constellations
Three constellations
in which a data-foundation phase is indicated
Data-foundation initiatives make sense when the existing data landscape structurally limits strategic steering, reporting or planned AI initiatives.
Adjustments to reports have long turnaround times. Self-service analytics is not established. Statements are not consistently traceable.
Training data is incomplete, contradictory or not sufficiently documented. Productive use fails due to data quality, not the model.
Approach
Five phases,
Five phases,
iterative delivery logic
The build-up is iterative. A usable layer emerges within eight to fourteen weeks. Further domains follow in successive iterations that build on one another.
1
Inventory of the data landscape
Structured capture of all relevant data sources: system, ownership, timeliness, quality, access rights. Consideration of both the central source systems and the decentrally maintained spreadsheets.
2
Target architecture
Decision between cloud, on-premise and hybrid architecture. Definition of the data model (data warehouse or lakehouse), the layer model, the naming conventions and the governance framework. Documentation of the architecture decisions.
3
Data pipelines
Implementation of the ETL or ELT pipelines per source system with extraction, cleansing, transformation and loading. Orchestration via Airflow, Dagster or dbt. Integrated data-quality tests per pipeline.
4
Semantic layer and BI
Consolidation of the metric definitions in the semantic layer (dbt, LookML, Power BI Semantic Model). Building the dashboards on a unified data definition for central business figures.
5
Governance and handover
Definition of data ownership per domain, definition of the change process for new metrics, documentation. Structured handover to the internal team. Support during the stabilization phase over 30 to 90 days.
Delivery format
Productive layer in 8 to 14 weeks. Iterative expansion.
The first productive data foundation for one to two business domains emerges within eight to fourteen weeks. Further domains are developed in successive iterations that build on one another.
View reference case
Impact
The impact of a
The impact of a
consolidated data foundation
Three impact dimensions that are regularly achieved in the data-foundation phase of our engagements.
Automated reporting processes
Standard reports and period-end closings run automatically. The capacity freed up in controlling shifts from data preparation to specialist analysis.
Unified metric definitions
The central semantic layer ensures a unified definition of business-relevant metrics. Management, controlling and business units work on the same data foundation.
A reliable data basis for AI
Models for forecasting, anomaly detection and classification build on a documented and quality-assured data foundation. The preparation phase for AI initiatives is significantly shortened.
Scope of services
Scope of services and
Scope of services and
boundaries
We are responsible for architecture, implementation and handover. Topics outside this scope we address transparently through our partner network.
In scope
- Audit of data sources with assessment of quality and timeliness
- Target architecture (data warehouse or lakehouse, cloud setup, layer model)
- ETL and ELT pipelines including data-quality tests
- Semantic layer with consolidated metric definitions
- Dashboards for central use cases (Power BI, Looker, Tableau)
- Governance model with ownership, change process and documentation
- Structured handover and support during the stabilization phase
Out of scope
- License resale for platform providers
- Procurement and operation of hardware and servers
- Permanent platform operation after handover
- SAP and ERP customizing or ABAP development
- Lift-and-shift migrations without a functional architecture concept
For permanent platform operation we work with established implementation partners.
Technology selection
Sovereignty as a
Sovereignty as a
selection criterion
The selection is guided by data residency within the European Union, by sovereignty requirements and by the existing IT landscape. European providers and open standards take priority. International platforms are used exclusively in EU regions with documented data processing agreements.
OVHcloud (EU)
Scaleway (EU)
STACKIT (EU)
PostgreSQL
dbt
Airflow
Dagster
Airbyte
Great Expectations
Metabase
Azure (EU region)
AWS (EU region)
Snowflake (EU region)
Databricks (EU region)
Power BI
Reference
Single Point of Truth
Single Point of Truth
in 14 weeks
Consolidation of eight data sources into one central data foundation at a mid-sized food manufacturer. Significant reduction of the reporting effort.
Engagement · Food production · 600 employees
Consolidation of eight source systems into one central data foundation with an automated period-end close
Starting point: ERP, CRM, MES, four decentralized spreadsheets, manual consolidation. Target state: single point of truth in Snowflake, dbt models, Power BI for management and business units. A reliable basis for downstream AI initiatives.
Read the full case
8 → 1
Data sources
Significantly lower
Reporting effort
14 wks
Time to production
Frequently asked questions
Questions from the
Questions from the
preparation phase
Specific questions about the data foundation and business intelligence solution area. General questions about working with LYN Intelligence can be found on the homepage.
Is a data warehouse strictly necessary, or is a BI tool on the ERP enough?
Direct access from a BI tool to the ERP is suitable for individual reports and a single source system. As soon as three or more sources are integrated, historization is required or several business units need to be served, requirements for performance, maintainability and data quality arise that make a central data foundation necessary.
How long until the first reliable metrics emerge?
A first productive data foundation for one business domain regularly emerges within eight to twelve weeks. The complete build-up across all relevant areas is designed as an iterative process over six to eighteen months, with a productively usable layer available after each iteration.
How are existing reports and spreadsheets handled?
Migration is done step by step. The new layer is built in parallel, critical reports are gradually replaced. Spreadsheets remain available as an ad-hoc analysis tool, but as a consumer of the consolidated data and not as a parallel data source.
Who is responsible for operating the pipelines after implementation?
The standard case foresees a handover to the internal team (IT or a dedicated data-engineering role). Architecture, pipelines and governance are fully documented, training is provided as part of the handover. During the stabilization phase of 30 to 90 days, functional support is provided. Where internal capacity is lacking, we arrange an implementation partner.
Cloud, on-premise or hybrid architecture?
The decision is made based on the existing infrastructure, the regulatory requirements and the scaling expectations. Cloud architectures offer advantages in scalability and implementation speed, on-premise and hybrid solutions where there are strict requirements for data storage and sovereignty. For cloud setups we consistently ensure data residency within the European Union.
How does the data foundation interlock with the AI strategy?
The data foundation and AI strategy are methodically closely linked. The usual entry point is the strategy phase with use-case prioritization and assessment of data maturity. After the data foundation is established, the first AI applications are implemented on a reliable basis.
Solution areas at a glance
Strategy, data foundation and AI
Strategy, data foundation and AI
as connected phases
The three solution areas form phases that build on one another. The data foundation is the prerequisite for strategy work with reliable statements and for productive AI applications.
AI and data strategy
Prioritization of relevant use cases, assessment of the data situation, building a reliable roadmap and a viable business case.
View solution →Data foundation and business intelligence
Consolidation of data sources, building the single point of truth and modernizing reporting as the basis for AI.
Current pageAI agents and delivery
Business-oriented AI applications on enterprise data, from piloting through rollout to productive operation.
View solution →
Arrange a
current-state assessment
In a 60-minute exploratory conversation we clarify whether a data-foundation phase is the suitable entry point or whether another solution area offers the greater leverage.