Monetizing digital asset management: The power of metadata management
Forrester projects the digital asset management (DAM) market to exceed $1 billion by the end of the year. This market—and numerous others—hinges on deriving benefit from metadata management. The capital surrounding DAM is perhaps the most tangible marker of metadata’s transition from supporting passive use cases (such as data lineage) to active ones associated with accruing revenue.
Specifically, achieving ROI from DAM, enterprise content management, content services, and other expressions of modern digital content requires the following three capabilities:
Data cataloging: Without adequate descriptions and tagging of data assets, organizations lack a consistent, reusable means of retrieving those assets for deployment at scale. “AI has the ability to understand the subject of a piece of content on its own and then automatically assign it the appropriate tags,” stated David Schweer, product marketing director, Aprimo.
Uniform accessibility: Universal data repositories (delivering ubiquitous access through multiple and hybrid on-premise clouds) ensure organizations are working with the same content despite distributed settings, systems, and use cases.
Analytics: Real-time, predictive, and prescriptive analytics are the final component for understanding which content has resonated with which audiences and why, delivering a road map for successfully positioning subsequent content.
Each of these requirements for optimizing digital content depends on operationalizing metadata, which is DAM’s foundation and that of data management in general. Metadata management is the latest data landscape domain to emerge from back offices as a means of monetization. According to Gartner, metadata is now actuating dynamic systems and “becoming the primary driver for all AI/ML.”
When those systems underpin DAM, digital content, and data assets, “this metadata can now flow from system to system to power personalization for a better customer experience,” observed Elliot Sedegah, group product marketing manager, Adobe Experience Manager.
When paired with a rich assortment of detailed metadata, modern analytics options present the foremost means of profiting from DAM by increasing conversions and measurable interactions with customers or prospects. There are three requisites for producing these results:
Synthesizing DAM and product information: Combining digital asset information with product or service information yields visibility into how the former is impacting the latter. Universal repositories that encompass data for these domains—or that link them together—assist with this goal. Metadata reveals how assets and products relate.
Descriptive analytics: Real-time descriptive analytics measures is the value of digital assets in relation to products or services. “When you link to something like an analytics suite that can report back on [whether] this asset is performing really well with these markets, you can bring that back into the DAM as metadata,” said Kelsey O’Brien, DAM product manager at Salsify. “That can actually continue to drive more optimal practices within your full platform.”
Advanced analytics: Machine learning delivers a flawless feedback mechanism about which aspects of content resonated with whom, yielding insight into why and how to improve future content. For example, for marketing materials, it would be possible to look at which people were given a document in the past, said Jans Aasman, CEO of Franz. “Then we could do machine learning to learn what was handed out and how effective it was.”
Metadata plays a critical role in machine learning applications for increasing content effectiveness. This technology is largely based on “factors.” Organizations can “take one piece of content and put the official metadata in the factor,” but also put the words used in the document in the factor and other details, such as the customer it was given to and whether there was a sale, Aasman explained. Once these factors are in place, “then the machine learning system will try to figure out what the characteristics are of a factor that leads to a sale,” Aasman said.
The ability to repurpose content across organizations for multiple use cases is essential to boosting the ROI of digital assets and their products or services. Machine learning and AI can be used to automate many of the content creation, sourcing, assembly, delivery, and personalization tasks that need to be done for vast quantities of new assets, said Sedegah. Tagging digital assets with meaningful, accurate metadata to spur the processes Sedegah referenced is indispensable for reusing them—as opposed to redoing this work.
Cataloging this metadata lets users know what content is available, who owns its rights, and any expiration date. “You can have metadata on an image that drives expiration of that image, making sure that you’re alerted when you’re coming up with an expiration date and ‘un-publish’ it from a retailer website and publish the newer version,” O’Brien noted. “[There is also] metadata that ties to the usage rights that can kind of say where an asset can and can’t be used.”