OVERVIEW

Global Content Works builds incredibly great data. Our projects are defined by a full understanding of goals, objectives, scope and timelines. We work with any input form and format. We construct supplier databases, parts databases, customer databases. We can meet any need. We can take work on in phases, tackle the entire project, and we can assign a dedicated team for hire. We are here to help you.

Taxonomy, Schema, Data Model, Data Architecture, Consulting
The design of the foundation for your business is not a trivial task and it should be architected to meet present needs and future requirements. We can recommend best practices depending on your requirements and we can work with your existing business system vendors and / or your internal resources driving toward the optimal solution.

Duplicates Analysis and Reduction: Based on Given Data
Duplicate part numbers, manufacturer & supplier names, descriptions are flagged by type and identified as candidates for elimination based on rules. Reducing duplicates in your data is like reducing inventory held unnecessarily. When you have multiple systems across multiple locations, many suppliers and lots of commodities, and when your data stewardship historically has not been adhered to, there are bound to be a lot of duplicate items and duplicate suppliers.

Classification / Categorization
Classifying the data in to an industry-accepted or customized taxonomy is the first step in organizing data and driving value out of it. Data that is classified can be used for detailed item-level or bottom-up spend visibility and analysis. Transactions executed with items that are classified can be analyzed. Commodity managers can get insight into the spend habits across their enterprises when items have been rationalized by deduping and classified.

Important Note: We employ a three-stage process for the highest yield possible. Yield is the proportion of parts correctly classified to the deepest level possible within your category tree, plus the proportion of the remaining parts that are correctly unclassified. Yield must take into consideration items that are classified incorrectly and items that are unclassified falsely. We do use technology and automation when and where appropriate, but we do not let that overcome the human element. Nothing works better than human experience.

Normalization & Standardization
Ensuring that all data fields and parameters conform to a uniform, clear and consistent style (nomenclature, acronyms, units of measure, abbreviations, spelling, etc.) This helps eliminate new parts introduction that otherwise shouldn’t happen. It helps people placing and tracking orders, or receiving orders do their jobs more productively. It helps give visibility into what locations have which parts from what suppliers – so that true supplier rationalization, parts rationalization, contracts optimization and spend minimization can be accomplished.

Sourcing Data
When your existing data just doesn’t have enough “bits” and “bytes” to be unambiguous, when your data is too ‘toxic’, when classification isn’t possible, and when normalization and standardization cannot be done, we can enhance your data so it becomes workable. To comply with your data model and business requirements we perform researching and gathering data from sources you approve, migrating that data, conditioning it and cross-referencing it to your data, item by item.

Enrichment
This is the process of extracting and transforming raw data from your internal sources or the external (as described earlier) and populating the appropriate information correctly into the various attributes and fields within your data model. We apply extremely rigorous testing to assure that the data populated is in fact not only what belongs in each field, but that the value is correct and that it meets the data type, range, and enumeration or other rules, for each category, for each attribute.

Form, Fit, Function Duplicates Analysis and Reduction
With a fully populated data model, we evaluate attributes to identify identical parts for elimination, or for second sourcing for Product Lifecycle Management (PLM). Having all the necessary attributes in place, we can evaluate form, fit and function. For example if two parts have the same size, the same color and the same style, if they have the same purpose and can be substituted with no loss of performance, then in this case we might consider the two items identical. Different degrees of identical will occur depending on the category and the strategic importance in your organization.

Rules-based Long and Short Description Generation
Once your data model has been fully populated, this step provides the highest quality, human readable, full and complete descriptions possible. Typically the noun-modifier pair is appended with each attribute name and the corresponding value, in essence constructing a description for every item in a particular category in exactly the same manner. Different rules can be created for each category. Short descriptions, that meet certain business system criteria, are created by applying rules-driven truncation, abbreviation or other methods, and by prioritizing that information most important for your particular application.

Guided Selling, Persuasive, Marketing Driven Copy
For your electronic catalog, web site, CD ROM, paper catalogs, we can aggregate your finished data with images, crop, resize, format and verify, to help you speed to market your outward looking content. We work closely with your experts to help you become more efficient and effective.

Project Management, QA / QC, Customer Satisfaction
We will hand-hold our clients through the process, work virtually side-by-side with your people, perform extensive quality checks, and make sure we perform according to a mutually agreed-to statement of work and project scoping specification.