Is Your Master Data ready for the future?

If you answered "not really" to the question above, don't feel alone; almost every large enterprise running SAP or Oracle (or Microsoft or IBM or take your pick of software) we've seen has at least tens of thousands of potentially dirty data lines today messing with their systems. And all of them want to rush headlong towards the AI-enabled horizon.

Gartner has been blunt about this in 2026 for enterprise engineering and procurement benefits: "generative AI has entered the trough of disillusionment, with many organizations seeing uneven results because AI was layered onto fragmented, messy data" rather than built around "mastered data" from day one.

Exactly 20 years ago we submitted a patent application, the first sentence of which read: "This invention relates to the tool which is embodied in a complex but easy-to-use software program aimed at business end-users which accurately and automatically converts material description data from a "disorganized" or "dirty" state to conform to and comply with a defined data standard".

Key to this invention was a set of data standards for goods and services used in your specific industry. Computer scientists refer to these data standards as dictionaries, taxonomies or ontologies and today they are back again at the absolute forefront of modern Agentic AI software.

You see, modern AI often suffers from failures because LLMs lack a formal logical layer. But in our software, Refresh, a neurosymbolic taxonomy-based approach solves this, limiting the questions asked of LLMs to those that can utilize its probabilistic power, and keeping the logical rigor of ontologies to drive what classes and characteristics are practically applicable to your materials. 20 years of collecting battle-tested industry specific dictionaries and ontologies means that we have done the hard work for the models.

What does this all mean for you? You get to use deep AI automation that just works on day 1, delivering for you a cutting-edge, novel approach to data standardization projects that didn't exist, even a few years ago. This gets you set up properly for the AI-enabled future – with your cleansed and de-duplicated data foundational for your own moat, from day 1.

This breakthrough approach opens up a whole new seam of procurement and maintenance savings, sales opportunities and risk reductions, especially for large enterprises, given their significant heterogeneity of materials. Quick new wins at our well-known customers today include:

Consistent descriptions for easy-to-find materials.
Reduced working capital & inventory levels.
Increased sales order, production order, customer order fulfilment.
Increased effectiveness of your ERP, PIM, CRM, EAM systems.
De-duplication of materials within and across sites.
Increased spend visibility and reduced spend per commodity.
Increased Tariff, Tax and Accounting compliance levels.
Trustworthy specifications, manufacturer & part number data.
Automatic, reliable, technical enrichment & translation of each item.
Reduced maverick purchasing, bought right first time.