Promote Consulting - Improve the Quality of your Spend Data. GREEN
Savings Insights – Clean Data

This is the first in a series of articles exploring our Savings Insight Service. Savings Insight is designed to bring your data to life and turn it in to actionable insights. If you want to know the maturity of each of your categories or what your next savings projects should be, we have an innovative solution that helps you to draw this insight from spend data.

The first article looks at how we get to clean data as the foundation of our Savings Insight Service. 

How clean is your data?

Clean data is data that is consistent, reliable and free from duplication. It is data that is stored centrally and is easily accessible across the organisation. In the world of procurement, complete and accurate records of data on products and suppliers is key to managing spend, creating new value and gaining deeper insights.  Despite its importance, the number of organisations whose data is full of inaccuracies, duplicates and inaccurate categorisation is worrying. Many organisations we work with believe that their spend data is >80% accurate before starting to work with us. We are typically able to correct between 20% to 35% of data and estimate that at the end of our work data is at least 90% accurate. That means that most organisations are, in reality, only at a 55% to 70% level of accuracy.

Ask yourself, what value is there to building a category strategy where the data that underpins it is only 55% accurate?

How we clean your data?

We begin by conducting an overview of transactional data and supplier records. We are able to load all your data – purchase orders, invoices, contract, preferred supplier lists and supplier master records into our Savings Insight engine.

We use our own proprietary data as well as 3rd party data sources to evaluate your suppliers and allocate them to appropriate categorisation based on their business activity. Our Savings Insight Engine can then quickly identify any likely mis-categorisations. We then apply assisted machine learning to recommend the right categorisation. The engine reviews the details of invoice and purchase order lines to identify the best match for the supplier and transaction. The benefit of using assisted machine learning is that the engine keeps learning based on the corrections that are accepted or rejected, so can speed up the correction process.  

Our solution also identifies potential corrections in your supplier master data. We can identify duplicate records and linked companies. This is vital for the management of strategic suppliers, as well as reducing the risk of duplicate payments.

What can you do with good data?

Firstly, we can help you put that data to use in your organisation. We have a suite of standard data visualisations that help you to review and use the spend data. For organisations that have standard formats for your Category Strategies, we can produce bespoke data visualisations that can enable Category Managers to focus their time on their jobs, not being data analysts.

More importantly, we use this data to help you gain meaningful insight. In the next article in the series we will show you how we assess the maturity of your procurement organisation and categories