The Spikes Cavell process to extract and enrich the raw financial data

The raw data is extracted from the organisation's financial systems and put in to a standard format. Classification is added to help people see what specifically the purchase relates to. The enrichment process adds extra information about the suppliers, such as the number of employees. This information comes from Dun & Bradstreet. All the purchases from a specific supplier are then aggregated across the records held in the database from all the participating organisations. The data is then ready for display in the spotlightonspend application which is installed on each organisation's web site.

Extract

Raw financial data is extracted from the financial system. To make it easy, we provide a documented data extraction specification. An experienced project manager is on hand to provide practical advice, guidance and support.

Standardise

The validated data extract is processed using a specially developed engine to standardise the data, remove duplicates and prepare the file for classification.

Classify

A specially developed, sophisticated matching and inference engine is used to match the supplier record to the Dun & Bradstreet and Observatory reference datasets and append a primary classification to the supplier record. If purchase order descriptions are also being processed, the item record is matched to our UNSPSC reference dataset and advanced statistical techniques are used to append a UNSPSC classification to each item. The most granular a level of the UNSPSC taxonomy that can be supported by the source data is used. All classifications that represent 97% by value of spend are manually validated by classification experts.

Enrich

The data is enriched with additional information about the suppliers using the Dun & Bradstreet and Observatory reference datasets. The type of information added includes: the Number of Employees, Annual Revenue (Actual or Modelled), Date of Incorporation (Birth Year), Geographic Location, Risk Classification (Modelled).

Aggregate

The standardised, classified and enriched records are aggregated together and each suppliers' record is linked to a master record in the Observatory reference dataset. This supports the identification and analysis of common suppliers irrespective of how that supplier was named in the original financial management system.

Universal Taxonomy Summary

The Spikes Cavell Universal Taxonomy is a unique public sector specific master classification system used to classify suppliers, items or extended items in a standardised way to support analysis and comparison of spend irrespective of the type of public sector entity. The Universal Taxonomy underpins the category based analysis in spotlightonspend.

Get involved

Please tell us how you would like to get involved.


Publish my organisation’s data
I would like to join the Advisory panel
I would like to join the People's panel
I would like to be a sponsor
First Name:
Last Name:
Email:
Phone:
Organisation:
Post Code:
Send me the transparency bulletins

Where does the data come from?

spotlightonspend uses data from public sector bodies who have had their invoices processed and classified by Spikes Cavell.

Find out more about the process...

Who is using spotlightonspend now?

A number of local governments have already published their spend data on their own websites using spotlightonspend.

See who has already published here...

Eric Pickles, Secretary of State, CLG

"I'm really excited about the opportunities of transparency and it's something this government is utterly committed to ..."