One of the major challenges with Yard data is the bad quality of data. For example the Yards may send “ALBERTA TUBULASR” when referring to a sage entity named “Alberta Tubular Products Limited”. A custom-developed learning module was developed to enable the system to do a fuzzy match between the two and determine that they mean the same thing. If suitable corrections could not be made, the system raised the data for correction to the users. The system was designed to learn from any corrections so that a correction once made by users was remembered by the system for future references for that customer.  Middleware services also ensured that stock records were processed in order and that a stock out movement related to a stock in the movement were treated in the correct order.

An SOA oriented middleware is envisioned with BizTalk acting as the common services hub. BizTalk services exposed standardized middleware web services that current yards and future systems could use to load data into Sage. Business and transformation rules along with business processes implemented in BizTalk ensured that a central normalized view was available throughout the enterprise and all its systems.

An add-on Sage Adonix X3 enabled services to be exposed for all objects and subprograms available within Adonix. This enabled almost all the functionality available within Sage to be available to external integration via the service bus.