How do you adapt and customize your DAM metadata standards to different projects and audiences?

Before you dive into metadata creation, put yourself in your audience's shoes. Consider the types and quantity of assets, your project's goals, and who will use the assets. Understand how users will access and search for them. Tailor your metadata strategy to meet their needs, ensuring a seamless experience.

You'll need to not only interview the people who will use the assets and find out what they need, but also go through every automated process that will touch the assets and find out what's the metadata they require or can potentially benefit from. Many times metadata is more valuable for automation than it is for users.

You need to understand a few things, firstly your audience and how they search, use words etc Secondly, needs to be your assets, what kind of assets are they? How do they work? Are they all the same? Are they similar? By understanding two those things you can build off that information and start to customise the metadata and content as it comes into the platform.

Sometimes it's good to allow some duplication in favor of simpler processing down the line. A classic example of this is video framerate, which some automation tools may want as a fraction and others as a sufficiently accurate decimal representation. Instead of tooling that conversion for each recipient, you may find it easier to supply both values and only automate its calculation when ingesting media. A rule of thumb is that duplication should only happen as a result of an automatic process.

Mapping and harmonizing metadata is like organizing a library with books from various systems. Mapping finds equivalent info (like matching "author" to "creator"), while harmonizing ensures consistency (like shelving all thrillers together). For instance, converting dates to a standard format. Just as a librarian organizes a library for accessibility, mapping and harmonizing metadata ensures data coherence.

You want to have a set of standards that everyone will work to - this should be accessible to all. And taught to all. On duplicate detection, most are 99% accurate so enabling this feature will help in the long run.

Just like tending to a garden, metadata requires regular care and attention to thrive. Imagine planting seeds and watching them grow – as your project evolves, so should your metadata. Reviewing ensures you're on the right track, like checking if your plants are healthy and blooming. Updating is like pruning and watering, making necessary adjustments to keep everything in top shape. It's all about nurturing your metadata to ensure it remains relevant and useful to your audience.

It's important to separate metadata that is strictly the result of reviewing the factual information of the assets, such as pixel size of an image, or bitrate and used codecs of a video file, from metadata that is subject to change, such as keywords and notes. In other words, automatically collected metadata should not be editable or updatable by any user.

Crafting and disseminating metadata standards isn't just about creating a blueprint, it's about building a bridge for seamless communication. Documenting meticulously ensures clarity and consistency, while sharing facilitates collaboration and amplifies the value of your data.

It's very important to find out what the number of total assets you foresee ever storing in a DAM system and whether it is capable of handling that number. Handing over 1.5 million video assets to a system that's designed to handle a few thousand at most is not for the faint of heart.