Product data glossary

Data governance is the set of roles, rules and processes that decide who owns product data, what good data looks like and how changes are made, checked and approved.

Also called: Product data governance

For product data, governance answers practical questions: who can create a product, which attributes are required for each category and channel, how GTINs are assigned, who approves new content and how quality is measured. It assigns owners and data stewards, sets standards and validation rules, and defines workflows for changes. A PIM or MDM platform can enforce many of these rules, but it cannot decide them.

Governance works best when it starts small, with a few measured rules tied to real problems such as retailer rejections, and grows with the program.

Related: data quality, golden record.

Learn more: data literacy and governance, Data quality in PIM.

Digital asset management (DAM) is software that stores, organizes and governs images, video, documents and other files, with the metadata and usage rights that control where each asset may be used.

A DAM gives creative, marketing and product teams one library of approved files, with versions, search, renditions for different channels and rights information such as license expiry. For product content, the DAM holds the product photography, lifestyle images, videos, spec sheets and manuals, while the PIM holds the structured data and links each product to its assets.

The key design decision is which system is the authority for what: the DAM for the file and its rights, the PIM for which assets belong to which product and channel. Getting that integration right avoids duplicated or outdated images on retailer sites.

Related: enhanced content, product data syndication.

Learn more: DAM implementation, PIM vs DAM vs CMS.

The GDSN is GS1's network of certified data pools that lets suppliers publish standardized product data once and share it automatically with the retailers and distributors that subscribe to it.

A brand owner or manufacturer loads product data into a GS1-certified data pool. Basic information about each item is registered in the GS1 Global Registry, and trading partners subscribe through their own data pools to receive the data and any later updates. Because every party uses the same GS1 attribute definitions, the data can move between companies without custom mapping for each one.

GDSN is common in grocery, food service, healthcare and consumer packaged goods, where retailers often require it for item setup. A PIM is not required for GDSN, but many companies use one to prepare and validate data before it is published to the data pool.

Related: GS1, GTIN, GPC.

Learn more: GDSN expertise, GDSN implementation, What is GS1? What is GDSN?, GS1 GDSN (gs1.org)

A GTIN is the GS1 number that uniquely identifies a trade item, such as a product or case, and is the number encoded in barcodes such as UPC and EAN.

Also called: GTIN-8, GTIN-12, GTIN-13, GTIN-14

GTINs come in four lengths: GTIN-8, GTIN-12 (the UPC used mainly in the United States and Canada), GTIN-13 (commonly called EAN) and GTIN-14, often used for cases and other packaging levels. Companies usually get the numbers by licensing a GS1 company prefix from their local GS1 member organization, then assign a GTIN to each distinct product and packaging level.

The GTIN is the key that retailers, marketplaces and the GDSN use to match a product across systems. Duplicate, reused or missing GTINs are a frequent cause of listing errors, so GTIN assignment rules belong in data governance.

Related: GS1, data pool.

Learn more: GDSN expertise, What is GS1? What is GDSN?, GTIN (gs1.org)

A golden record is the single, trusted version of a product (or other entity) created by matching and merging data from several systems under agreed rules for which source wins.

Also called: Master record

When the ERP, PIM, supplier files and commerce platform each hold part of a product's data, and some of it conflicts, the golden record is the agreed result. Rules decide which source is trusted for which field (for example, pricing from the ERP and marketing copy from the PIM), how duplicates are matched and how conflicts are resolved. These are often called survivorship rules.

Golden records are a core idea in MDM, but the same thinking applies to any stack where product data is spread across systems. The hard part is usually agreeing ownership, not the technology.

Related: data governance, data steward.

Learn more: Defining a golden product record in a decentralized tech stack, MDM expertise.

Master data management (MDM) is the discipline and software for keeping one governed, consistent record of core business entities, such as products, customers, suppliers and locations, across the systems that use them.

Also called: Multidomain MDM

MDM matters when the same entity is created and changed in several systems, such as ERP, CRM, supply chain and commerce, and those systems disagree. An MDM program sets ownership and rules, matches and merges duplicate records into a golden record, and shares that record back to the systems that need it. "Multidomain" MDM covers more than one entity type in one platform.

Compared with PIM, MDM is broader in scope and usually more focused on operational accuracy and governance than on marketing content. Some platforms offer both, and many organizations run a PIM for customer-facing product content alongside MDM for enterprise master data.

Related: data governance, golden record.

Learn more: MDM expertise, MDM implementation.

Product data syndication is the process of mapping, validating and sending product data and assets to retailers, marketplaces and distributors in the format each one requires, and keeping it current as products change.

Also called: Content syndication

Each channel has its own attribute names, required fields, image specifications and category structures, and those requirements change often. Syndication maps the internal product record to each channel's format, checks it against that channel's rules, sends it by feed, API, portal or the GDSN, and handles rejections. It can be run from a PIM with built-in syndication, a separate syndication tool or a mix of both.

Good syndication depends on good source data: a clear taxonomy, consistent attributes and a measure of channel readiness before anything is sent.

Related: product feed, enhanced content.

Learn more: syndication in managed services, marketplace implementation.

Product experience management (PXM) extends PIM by tailoring product content to each channel, audience and market, so the product is presented in the way that helps each shopper decide.

Where PIM focuses on getting product data complete and correct, PXM focuses on how that data performs: channel-specific copy and images, localized content, enhanced content and feedback from the digital shelf such as search visibility and ratings. Several PIM vendors use PXM to describe their platforms, so the term describes a set of capabilities and practices more than a separate category of software.

In practice, a team is doing PXM when it adapts content for each channel and measures the results, not just when it publishes one version everywhere.

Related: channel readiness, AI enrichment.

Learn more: PIM & PXM expertise, managed services.

Product information management (PIM) is software and practice for collecting, enriching, governing and distributing product information, so every channel receives accurate, complete and consistent product data from one source.

A PIM holds the attributes, descriptions, categories, relationships and translations that describe what a product is and how it should be presented. Data usually arrives from an ERP, suppliers or engineering systems, is enriched and approved by product content teams, then sent to an eCommerce site, retailers, marketplaces and print. Without a PIM, that work tends to happen in spreadsheets, and each channel ends up with a slightly different version of the product.

A PIM is different from a DAM, which manages files such as images and video, and from MDM, which governs several kinds of master data across the enterprise. Many companies use a PIM and a DAM together.

Related: attribute, product taxonomy, product data syndication.

Learn more: PIM & PXM expertise, PIM implementation, PIM vs DAM vs CMS.

A product taxonomy is the hierarchy of categories used to organize a catalog, with each category defining which attributes its products should have.

Also called: Category hierarchy

The taxonomy does two jobs. For shoppers, it drives navigation and filters on a site. For the product team, it defines what "complete" means: a product in the drills category needs voltage and chuck size, while a product in the shirts category needs size and fabric. A well-designed taxonomy makes both jobs easier, and a poorly designed one creates duplicate categories and inconsistent attributes.

Companies often maintain more than one taxonomy: an internal master taxonomy in the PIM and mappings to each retailer's or marketplace's categories. Industry standards such as ETIM and GS1 GPC provide shared classifications for specific sectors.

Related: attribute, product data model.

Learn more: taxonomy, attribution and enrichment, Plezio Root.