SAP data in BigQuery: demand planning and inventory in technical wholesale
A technical wholesaler with around 650 employees and SAP ECC replicates sales, purchasing and inventory data to Google BigQuery. On this basis, demand forecasts, inventory recommendations and analyses are created, which sales and purchasing query in plain language using Gemini. The project plan allows around three months.
- Companies on SAP ECC or S/4HANA whose reports arrive too late
- Purchasing and sales teams that want to manage inventory and demand better
- IT teams that want to make SAP data usable without a new data warehouse
Why does a wholesaler bring its SAP data into BigQuery?
The wholesaler manages purchasing, warehousing and sales with SAP ECC. Reports are produced monthly in Excel, compiled from several SAP reports. By the time they are finished, they are often already out of date. Management wants to see inventory and demand earlier, not just at the end of the month. At the same time, the move to S/4HANA is still a few years away, and nothing should wait until then.
Inventory ties up capital.
With around 80,000 items, some warehouses are too full, while other items are missing at the crucial moment. There is no reliable demand forecast.
Questions end up with controlling.
Anyone who needs a figure writes an email and waits. Sales and purchasing should be able to answer their own questions, with the same figures as management.
Project profile
How does the SAP data get into BigQuery and into everyday work?
As the wholesaler uses SAP ECC without Business Data Cloud, the data reaches BigQuery by replication: the BigQuery Connector for SAP transfers sales, purchasing and inventory data via the SAP LT Replication Server. In version 7, the Google Cloud Cortex Framework provides ready-made data products for SAP ECC, so analyses don't start from zero; custom Z fields are carried over automatically.
Dashboards for management, sales and purchasing are built in Looker. Demand forecasts run directly in BigQuery. Sales and purchasing ask Gemini their questions in plain language, for example about items with falling demand, and get answers based on the same figures.
Concept and architecture
Target picture for data and AI: which SAP data, route to BigQuery, permissions, Frankfurt region, cost framework. Two weeks.
Use cases and KPIs
Define the most important questions and KPIs with sales, purchasing and controlling. One week.
Connect SAP data
Install the BigQuery Connector for SAP, set up replication for sales, purchasing and inventory, together with the SAP team. Three weeks.
Data model and dashboards
Adapt data products from Cortex Framework v7, build dashboards in Looker. Three weeks.
Forecasts and Gemini
Demand forecast in BigQuery, inventory recommendations, plain-language questions for sales and purchasing. Three weeks.
Operations and expansion
Monitoring, cost control and further areas, such as finance.
Which technology is used?
BigQuery Connector for SAP
Replicates data from SAP ECC to BigQuery via the SAP LT Replication Server.
BigQuery
Central data platform for analyses and forecasts.
Google Cloud Cortex Framework v7
Ready-made data products for SAP data in BigQuery; custom Z fields are carried over automatically.
Looker
Dashboards for management, sales and purchasing.
Gemini in BigQuery and Looker
Questions about KPIs in plain language.
The data is stored in the Google Cloud region Frankfurt.
How can the benefit be measured after three months?
Starting point
- Monthly reports in Excel from several SAP reports
- Too much stock for some items, shortages for others
- No reliable demand forecast
- Every question about figures goes through controlling
Measures after four months
- Time until a KPI is available for sales or purchasing
- Development of stock levels and shortages for the most important items
- Share of questions answered without controlling
Find out in 30 minutes what your SAP data can do
- Assessment of your starting point
- Rough estimate of scope and duration
- Next steps
