Key Takeaways
- Established role as a digital intermediary
- Cut data refresh times from 53 minutes to six
- Improved frequency and quality of reports
- Supported acquisition strategy with fast data transfer
Data Stack
- Pipeline: Fivetran
- Source: SQL Server, Microsoft Dynamics, PostgreSQL
- Destination: Snowflake
- Business Intelligence: Power BI
- SI Partner: Biztory
I don't think we'd be able to deliver our data strategy at the pace this business needs without Fivetran.
- Tom Hocking, Head of IT, Troy
Troy, a buying group for independent businesses in the engineering and industrial supply sector, represents 350 member companies across 500 outlets in the UK. The business provides economies of scale in purchasing from over 400 suppliers, including huge brands like Stanley Black & Decker and Bosch with vast catalogues of over 60,000 products. “We are an extended team for our members and support them in their trading,” explained Tom Hocking, Head of IT at Troy.
The Challenge: Digital ambitions obstructed by manual processes
The engineering sector ranks as the second most digitally immature in the UK after agriculture, so when Troy began investing in omnichannel technology and data analytics, it was pursuing digital transformation on behalf of a market segment as much as itself. An early obstacle was finding more efficient ways to access data in a paper-based supply chain.
The business receives 650,000 invoices each year that provide the main source of purchasing data. After successfully completing a two-year project to replace paper with electronic invoicing, the next challenge was accessing multiple lines of stored data for insights. Jenny Bicket, Senior Data Governance and Strategy Lead stated:
There was a lot of manual deduplication and cleaning data that was trapped in silos. The major issue was the time it took to manually load the data – three times a day, 53 minutes each time.
The company had got to the point where it had outgrown its architecture and needed to automate processes, pursue ‘cloud-first’ ambitions, and scale.
The Solution: Fast and cost-effective connections
After working with Biztory as their systems integrator, Troy decided to move away from downloading data directly from SQL Server and migrate to cloud-first architecture with Snowflake as the data warehouse. The need for Fivetran connectors quickly became evident as the advantage of having “technology built-in” promised a fast and cost effective alternative to trying to connect to multiple systems manually.
Once set up, Fivetran has run seamlessly ever since, according to Jenny Bicket. “It’s ticking along in the background; we know it’s working because we keep seeing the data coming through,” she said. The data is turned into insights more quickly and presented to the Sales and Purchasing Departments, Directors and the company CEO in a number of formats: Power BI dashboards, downloadable reports and excel spreadsheets.
Fivetran also supports the company’s acquisition strategy. There used to be huge technical challenges and business disruption as whole ERP systems were migrated onto its platform. Now Troy uses Fivetran connectors to tap into the data it needs and leaves the rest in situ and makes migration decisions for the right reasons.
The Results: Evolving towards a data led organisation
The most immediate benefit from Fivetran is the dramatic reduction in data refresh times – cut from 53 minutes to six. There is now the capacity to refresh more often than three times a day as well as the option of ad-hoc reports if the business urgently needs new information. “We bought in Fivetran to fix an immediate issue on loading data and were soon looking at what else we could do with it,” said Jenny Bicket.
Besides aiding the acquisition strategy, Fivetran helps detect when items fall short on quality and alleviate confusion around different suppliers offering the same products. “Before, when all that data hit our ERP system there was nowhere to clean and fix it. Now Fivetran lets us pull it out and go through the cycle of tidying it up,” said Tom Hocking.
Fivetran has also improved the frequency and quality of reports shared with management. The plan over time is to move from descriptive analytics, the ‘why, how, when’ of the business to deeper insight into member company buying choices – online purchasing, for example, or buying via mobile apps.
About Fivetran: Our fully managed pipelines and automated connectors for databases, applications, events, files and functions help extract value more quickly from data, accelerating the path to becoming a data-driven business and achieving competitive advantage through insight-based decision making. Find out more at Fivetran.com
.svg)
.svg)
.svg)