BI Implementation Case Study for Steelite International
How to transform metric chaos into analytics
with 99% accuracy
When a business encompasses manufacturing, international distribution, and a large product catalog, it creates a complex operating model with high demands on data accuracy and decision-making speed.
After a series of mergers, a system with multiple ERPs and different accounting approaches was formed within one company, which led to inconsistency in indicators at the business level.
At some point, the problem went beyond the technical and moved to the management level: analytics ceased to be the basis for decision-making.
But we solved these issues for Steelite International, a global manufacturing company in the HoReCa segment.
Steelite International is a leading manufacturer of porcelain and tableware for the HoReCa industry. Its products are manufactured in the UK and sold in 145 countries.
Product quality is at the heart of the brand’s “made to last a lifetime” philosophy, which is why the manufacturer even offers a lifetime warranty against chipping.
We connected with the company through LinkedIn. At first, the request was simple—could we create a dashboard to track metrics?
But as we worked with the client, it became clear that the issue ran much deeper…
They came to us with 160,000 SKUs spread across 8 ERP systems
By the time we first met, the manufacturer had already begun using modern analytics technologies.
The company had accumulated 8 different ERP systems that handled finance, orders, inventory, and production.
These systems emerged following a series of acquisitions and mergers, as the company had been continuously expanding through business acquisitions and consolidations.
- Unsurprisingly, data across the systems remained fragmented.
- Visibility into sales, margins, and inventory at the regional level was limited.
- The accounting systems contained over 160,000 active SKUs.
- There were discrepancies in product specifications, names, and classification rules.
- Overall, the IT infrastructure was complex and inconsistent.
Results achieved: 92% faster with 99% analytical accuracy
Single data repository
A cloud-based data repository was created, serving as the single source of truth for all reporting at the brand and individual SKU levels. Data from various systems is consolidated into a single model without the need for manual consolidation.
Data Cleaning and Harmonization
Prior to integration, we conducted extensive preparatory work, which allowed us to resolve discrepancies and achieve 99% accuracy in BI reports.
Automation of Reporting
Reporting has been transitioned from a manual to an automated process: preparation time has been reduced by 92%, and updates are now performed daily instead of every ten days.
Global Analytics
We have implemented standardized dashboards and KPIs covering the United States, the United Kingdom, Canada, Europe, and Australia. Our analytical coverage has expanded by 65%.
Investment Preparation
To support investment transactions and financial due diligence, an analytical data cube was created. This allows for faster data extraction and ensures consistency across all reports.
Speed of Decision-Making
Centralized data enabled the company to respond more quickly to changes: it shifted from monthly reporting meetings to weekly analytical sessions.
How it all happened…
We Started by Analyzing the Architecture and the Stack
We always start our work by figuring out how the client’s business operates.
- So we asked them about their key processes, the rules of their operational model, and the areas where problems most often arise.
- This helped us understand which metrics are critical for them and where the business lacks transparency in its day-to-day decision-making.
- One of the key challenges was to find the best way to integrate data from 8 ERP systems without compromising the calculation logic or the consistency of metrics.
- After that, we moved on to the technical side: we clarified where the final calculation logic is embedded, how metrics are synchronized across systems, and who is responsible for ensuring their consistency.
Based on this analysis, we determined that an architecture was needed that covers the entire
data lifecycle — from ingestion to analytics.
Get a free consultation to understand how to structure your data and reporting
In 30 minutes, you will get a clear view of your current gaps, key risks, and the fastest way to improve your analytics.
Or text us via Messenger
What we proposed instead of a complex IT structure
When it became clear that the calculation logic was distributed across systems and not managed centrally, we proposed the following architecture:
- Azure Data Factory provides integration and orchestration of data flows, enabling centralized management of data ingestion without the need for additional code.
- dbt handles data transformation and modeling, making business logic transparent and controllable.
- Snowflake serves as a central repository that scales easily and delivers high analytics performance.
- Azure Functions adds flexibility because it doesn’t require a dedicated server and allows you to quickly automate individual steps without complicating the architecture.
- Python enables custom data processing and automation where standard tools cannot cover specific business requirements.
- Power BI delivers analytical dashboards that give teams access to unified, up-to-date metrics for daily decision-making.
Together, these tools form a unified data management system: they enable the centralized collection of information from various sources, its conversion into understandable business logic, its storage in a scalable repository, and the automation of individual processes without unnecessary complexity.
DOWNLOAD FREE CHECKLIST
of 25 recommendations for successful BI implementation
Why the client decided to build a centralized data warehouse
One of the reasons was how to centralize business data from different departments and regions to ensure consistent metrics, transparent reporting, and resolve issues:
Siloed Data Sources
The company’s data was scattered across various systems, making it difficult to get a comprehensive view of the business.
Wasted Time
Preparing reports was time-consuming because information had to be collected and reconciled manually.
Outdated Data
The business needed up-to-date analytics, and a centralized data warehouse allows for quick generation of accurate reports.
Lack of History
It was also important to maintain a history of changes to analyze not only current but also past data.
Need for Control
Quality control became a separate task: in the data warehouse, data can be cleaned, verified, and standardized before use in reporting.
Need for Scalability
Finally, the company needed to scale its analytics—a centralized data warehouse provides the foundation for growth in the number of sources, reports, and users.
What We Did as Part of the Project
The data warehouse solution for a company with multiple ERP systems was implemented gradually over the course of six months. However, the first tangible results were already visible in the first month.
Business Needs Analysis
We conducted interviews and research to identify the company’s key objectives and understand what data was needed for centralized analytics. This provided the foundation for further decisions.
Identification of Data Sources
We compiled a list of the main systems and information flows that needed to be integrated. This allowed us to identify which data was critical for business management.
Data Integration
We set up processes for collecting and consolidating information from various sources into a single environment. As a result, the data became accessible and consistent.
Data Warehouse Design and Transformation
We created the data warehouse structure and configured transformations to ensure that metrics were clear and aligned with business logic. This ensured transparency in calculations.
Data Cleansing and Quality Assurance
We performed data cleansing and standardization and verified the data’s accuracy and consistency. This eliminated the risk of errors and ensured the reproducibility of results.
Analytics and Reporting
We developed dashboards and reports that provide a transparent view of finances and operations. The results were validated by reconciliation with official financial statements.
In general, the work was organized so that changes were implemented in parallel with ongoing processes. Thus, business processes did not stop—the company continued to operate as usual. The implementation of the new system was seamless for daily operations and posed no risks to clients or financial reporting.
Examples of dashboards that we made for the customer
How the data cleansing and standardization process worked
We designed the process so that each stage addressed a specific task without
introducing new discrepancies in the data.
Data from various systems was consolidated into a single environment and immediately checked for completeness and basic accuracy to prevent errors from propagating further through the system.
At this stage, duplicates were removed, missing values were filled in or truncated, and obvious errors affecting the final metrics were corrected.
Dates, names, codes, and units of measurement were standardized so that the same information appeared consistent regardless of the source.
Data was reconciled using uniform rules and reference guides to ensure metrics did not vary depending on their source.
After transformation, the results were additionally verified to ensure that only reconciled and verified values were included in the reports.
We performed mapping to combine identical entities from different systems and obtain a unified view.
DOWNLOAD FREE CHECKLIST
of 25 recommendations for successful BI implementation
How has Steelite International’s operations changed as a result of our collaboration?
consolidating financial, operational, and management data into a single system.
Which Teams are Using the Company’s New Analytics?
The new analytics have become a common foundation for several teams. The finance department works with revenue, margin, and inventory metrics and reconciles them with financial statements. The sales department gets a comprehensive view of trends, customer segments, and product categories on a global level. Operations management uses inventory and turnover data for planning and timely response to changes.
How Has the Finance Team’s Workflow Changed Since the System was Launched?
The finance team’s work has become more structured and less reliant on manual report preparation. The time required to collect, reconcile, and consolidate data from various sources—which previously took hours—has been reduced to just a few minutes. Instead of manually preparing data, the team can now focus on analyzing metrics, identifying variances, and supporting management decisions.
What Decisions Were the Company Able to Make More Quickly After the Implementation?
First and foremost, the company began to respond more quickly to deviations in metrics and changes in the business, without waiting for the end of the reporting period. Next, the evaluation of the performance of customer segments and product categories accelerated, enabling the company to adjust its priorities promptly. Finally, inventory management and sales analysis became simpler, as data became available without additional reconciliation or preparation.
How Has the Preparation of Management or Financial Reports Changed?
Report preparation is no longer a process consisting of dozens of manual steps. Whereas it used to take up to 10 days to generate reports and required collecting and reconciling data from multiple systems, data loading, cleansing, and processing are now performed automatically. As a result, report preparation time has been reduced by 92%, and updates occur daily.
Has the Speed of Management Decision-making Changed?
Yes, the speed of management decision-making has increased. Previously, some decisions were delayed while waiting for reports and the need to verify data further. This created a lag between an event and the response to it. Now, management works with real-time metrics and sees changes in sales, inventory, and financial results without delay. Decisions are made the moment deviations occur, allowing for quicker course corrections and no wasted time.
The company’s CFO and СІО comments on the results
The solution has become an enterprise analytics platform handling large-scale SKU-level data management, providing consistent metrics across different teams and regions. Following the implementation of the solution, changes were felt not only in reporting but also in day-to-day data management.
The company transitioned from manual preparation and multi-day reporting cycles to daily updates of metrics, gained a single, consolidated data source, and expanded its analytics from the regional to the global level. This has impacted the pace of the teams’ work and the approach to decision-making—shifting from periodic reports to regular work with up-to-date metrics.
Here’s How CFO Stan Zolotarev Describes His Experience
"Most importantly, the pace of decision-making has changed. We’ve moved from working with data at the regional level to a comprehensive view of all key markets: the U.S., the U.K., Canada, Europe, and Australia. This has expanded our analytical coverage by 65%. As a result, decisions are made about four times faster, and we can respond to changes during the process rather than after the reporting period has ended.”
Here’s How СІО Josh Rammel Describes His Experience
“For us, the key change was how we started working with data. First, report preparation no longer takes days: previously, it took up to 10 days with manual data collection and reconciliation across multiple systems, but now the process is fully automated, metrics are updated daily, and preparation time has been reduced by 92%. Second, we now have a single source of data that ensures nearly complete reporting accuracy—at 99%. This has eliminated discrepancies between teams and allowed us to work with the same metrics across the entire company."
You can also get your data in order and switch to daily analytics
We examined a case study on the implementation of a data warehouse for a company that was working with multiple ERP systems and needed centralized analytics. Most companies face a similar situation: data is scattered across multiple systems, reports are generated manually, metrics don’t align across teams, and decisions are delayed due to verification processes. But you don’t have to put up with this.
Contact us—we’ll help you consolidate your data, set up automated reporting, and ensure transparent metrics for every team. Get daily analytics that let you make decisions quickly and confidently, without delays or discrepancies.
Get a free consultation to understand how to structure your data and reporting
In 30 minutes, you will get a clear view of your current gaps, key risks, and the fastest way to improve your analytics.
Or text us via Messenger
We also offer you to familiarize yourself with our other projects
Healthcare
“Dobrobut” clinics network
Result:
- Clinic's directors got daily updated P&L report;
- The management uses daily dashboard with budget/actual analysis on operational indicators;
- Dashboards combine data from ERP and the Doktor Eleks medical system.
Logistics
Meest China
Result:
- Dashboards combine data from ERP and CRM;
- Routes margin is automaticaly calculated in dashboards and monitored every day by management;
- Daily generated Profit & Loss report;
Manufacturing
Steelite International USA
Result:
- Automated reporting, reducing turnaround from 10 days to overnight.
- Scaled visibility from regional to global (US, UK, Canada, Europe, Australia).
- Centralized data enabled data-driven decisions on a weekly cadence instead of monthly.
Industry-focused case studies designed for your needs
