What is metadata? A digital object consists of content, the so-called fine data, and contextual information, the metadata. While the former represent the actual content, e.g. an image, the latter make the object really usable.
A lot of metadata is created automatically when an object is created: File type, resolution, creation date or author, to name just a few. You can also add additional information to an object. So photos about your IPTC data offer the possibility to store further information standardized in the file. In other cases the objects are managed by a software that provides additional information for each object.
This can be information that is directly related to the object, for example number of uses, or indirectly, for example source of supply.
Exemplary picture on the subject of milk advertising for keywording
To better illustrate the special importance of keywords as a particularly important form of metadata, I will first describe a scenario without their use.
In a company, a picture with a sunset in the mountains is bought for an advertising campaign on the topic of milk and then stored in the file system in the folder /pictures/advertising/food.
A few months later, another employee looks for a motif for a campaign on the topic of holidays. The picture from the milk advertising would be suitable and could also be used under licensing law. However, the employee searches under /Pictures/Advertising/Landscapes. He cannot find a suitable ad spec there. Therefore, the same picture, even from the same agency, is bought again, used and saved under /Pictures/Advertising/Landscapes.
A good keywording and the possibility to search in it would have solved this problem. The first employee would have tagged the picture after the purchase with keywords that describe it. In our example these would have been words like milk, mountains and sunset.
If the other employee now enters mountains as a keyword during his search, he will now find the picture again and can use it. Furthermore, he can also add the term vacation to the keywords and thus increase the quality of the media collection.
This example illustrates the influence of keywords on the quality of the data: A one-dimensionally structured accumulation of images becomes a collection searchable by many aspects, which enables the user to find the desired assets without having to search the entire collection. This saves time and money.
Pixelboxx analyst Axel Ziegler-Lüdke has shed light on what makes a good keywording in an article which you can find here.
Metadata and especially keywords require a system that allows the user to create, search and manage them. This may seem trivial at first glance, but it requires some thought beforehand.
The main reason for this is that not all metadata of an object is stored in the object itself. The specifications of a format provide for different metadata, partly cross-format, partly format-specific. Additional information must therefore be stored in part in the software that manages it.
Therefore file systems are not suitable for managing metadata. Only with additional programs can the information stored in the objects themselves be made visible and (possibly) searchable. Data that is not stored directly in objects because it is not part of the specified description of the data format cannot be managed at all.
Therefore, the alternatives left are a photo workflow software with an attached database (for example Adobe Lightroom), a database or a digital asset management software. If metadata is to be used automatically, for example by a shop system, even the first solution is not necessary.
You can find a detailed discussion of file storage here.
Metadata is needed to ensure that digital media do not become disposable, but remain versatile and easy to find. This saves the company money, employees can efficiently find the appropriate media instead of time-consuming research and the context of use remains consistent.
Metadata is a must, especially if workflows and media delivery are to be partially or automatically automated.