Home
About us
Solutions
Case studies
News
Procurement
TSV
Indirect spend is everywhere. Making it readable is the hard part.
A platform for analysing indirect spend · AI classification · data normalisation · standard or custom views
Schematic example of a spend model. 95% is the product’s reference accuracy; performance on your own data is verified during the assessment.
The context
Spend is scattered. Putting the picture back together is the difficult part.
In a group, purchasing data comes from different companies, different ERPs and different countries. Descriptions follow no common rule: the same item can carry several names, one supplier can appear under several legal entities, and the same category can be classified in different ways. Without shared criteria, comparing and analysing spend becomes hard.
Information spread across systems
Orders, invoices and contracts sit in different applications. Bringing them together and rebuilding an overall view takes collection, reconciliation and checking.
The same item, different names
Abbreviations, typing errors, languages and internal codes make it hard to recognise a product or service when it appears under different descriptions.
Small amounts stay invisible
Small purchases spread across many suppliers are hard to spot in aggregate analysis. Looking at them completes the picture and shows where consolidation is possible.
Data that arrives too late
Manual classification takes time and usually delivers the answer after the fact. Procurement loses the chance to use current data to steer decisions and negotiations.
How it works
From purchasing records to a structured view of spend.
TSV collects the data from your own sources, normalises the descriptions and assigns every line to the right spend category. The result is a consistent base you can read at several levels of detail, from the overall picture down to a single order.
Semantic Engine
Recognising the same item , however it is described.
The Semantic Engine recognises the same item even when the description changes by language, abbreviation, unit of measure or internal code. Everything is traced back to one common reference, which makes purchases comparable across companies and countries.
Every classification can be checked and, where needed, corrected through rules you can read. Corrections apply to historical data and to later loads, which keeps the criteria consistent over time.
“CRT A4 80GR RSM 500FF”
“white A4 paper 80 g — ream”
“A4 paper 80 gsm, ream of 500”
ITEM IDENTIFIED
Paper for printers and copiers
UNSPSC category · Office supplies · 3 suppliers · 2 countries
Illustrative example of normalisation and item recognition.
Before and after
From messy records
to spend that is
classified.
Drag the handle: on the left the purchasing data as it arrives from the ERPs, on the right the same information after normalisation and classification.
- Level of detail
- from group total to a single order line
- Criteria
- UNSPSC or your own model
- Languages handled
- Descriptions recognised across languages
- Changes to the ERP
- 0
Schematic representation of the transformation. Block sizes do not represent real amounts.
Inside the product
Three views
of the same spend.
One base of data answering three different needs: supporting negotiations, keeping the budget under control, and managing supplier relationships.
Finding the buying opportunities
For each category you can look at amounts, number of suppliers, prices and the differences between group companies. Negotiation then rests on comparable data instead of scattered information.
- Spend distribution by category
- The same item priced across companies
- Tail spend identified, with the room to consolidate
Checking spend against budget
Actual spend can be analysed by company, country, cost centre and contract. A shared classification lets the group’s entities be compared on the same basis.
- Analysis by company, country and cost centre
- Contracted spend compared with off-contract spend
- Export for management accounting
A complete view of the suppliers
Reconciling the master data rebuilds the relationships between the group’s companies and their suppliers: who buys from whom, on what terms, and under which contracts.
- Suppliers recognised across different legal entities
- Supplier concentration and dependency analysed
- Classification traceable down to the single line
What it takes to start
We start from the data you already have.
| Stage | Data needed | Result | Your IT effort |
|---|---|---|---|
| Data collection | Order lines, supplier invoices and supplier master data | Scope and data quality checked | One extract |
| Classification test | A sample of historical data | A first classification of spend against your own model | None |
| Service start | A periodic data feed | Platform kept up to date, with views for each role | Setting up the feed |
Timing and frequency of the updates are agreed according to scope and business needs.
Keep exploring
The other Add Value products
Intelligence · Insurance
Calamìta
Cross-references seismic, weather and hydrogeological data with geolocated policies to analyse the NatCat exposure of an insurance portfolio.
Intelligence · Banking
Tequery
Lets you query databases in plain language and check the SQL generated behind every answer.
Intelligence · Raw materials
PriceSense
Reads news and market data to estimate price movements and raw-material availability, with an analyst validating the result.
Data · Cloud and AI
frAIme
Integrates different data sources into a cloud-native data fabric, leaving the information in its systems of record and making it available to AI applications.
Intelligence · Mainframe
OptiMIPS
Analyses mainframe consumption and attributes it to the applications, to find where resources can be optimised.
Software · Modernisation
L2M
Core applications, modernised and verified against real data.
Intelligence · AI adoption
Officina AI
Finding where artificial intelligence can create value in business processes.