State of Brand Management Tools 2024

The 3 biggest obstacles for a DAM project: Extensive asset maintenance

The Digital Asset Management (DAM) is successfully implemented, the assets are transferred, and the users are trained. You are now expecting the DAM to run and harvest the promised key business benefits from such a tool.

This is where the last part of the comspace guest post series begins. In reality, one or several employees are almost exclusively occupied for a long time with manual tagging and categorization of existing assets. In case of an international company, the metadata needs to be translated as well.

Why is this necessary? If the assets are not consistently and precisely tagged and categorized, then a high-performance search in the DAM is not possible. The anticipated simplification of daily work does not take place.

Essential Tagging and Categorization of Images

In most organizations, huge volumes of images constitute the bulk of the assets residing in the DAM. These often do not have a descriptive name, but are stored under titles such as 2023948_201365.png or DSC_9001.jpg. How is the DAM supposed to know which images fit, for example, the product category “garden shears” and for which search query should they pop up?

A powerful search is only possible with tags

Without appropriate tags, images get lost in the virtual vastness of the DAM and are very difficult to find. This is especially true, when data volumes increase dramatically and many people use the system.

However, manually tagging images cannot be the solution. Tagging a single image can already take several minutes. Who is going to do this tedious work, especially since most companies upload large volumes of new images regularly?

Some DAM systems make the work easier by providing a clear capture dialog or bulk tagging of several assets at the same time, but only automatic solutions provide a real remedy.

Automatic tagging solution

One of these solutions is our automatic image recognition tool, Autotagger. It is based on the AI solution Imagga and handles the automatic tagging and categorization of huge amounts of images in a very short time.

AI dramatically speeds up tagging and categorization

How does the solution work? Via API, the image files are transferred to the artificial intelligence, where they are checked by the machine learning algorithm. The AI then reports the tags and categories back to the DAM and the data is stored at the respective image. In this process, the tags are used to quickly find the images, while the categories allow the user to search for abstract collective terms such as animal instead of parrot, or fruit instead of apple. By searching for appropriate categories, all users in the company have faster access to suitable images – whether in the DAM, CMS or in the connected third-party systems. Suddenly, the DAM answers search queries such as “car blue”, “sunset beach” or “woman laughing” with a selection of suitable assets. The result is a positive experience for all users. They get what they need for their use case quickly and easily.

Custom training thanks to learning AI

Autotagger can assign many images to suitable keywords and categories by default. When it comes to industry- or company-specific product images, the AI can be trained on new tags and categories in a custom training session. The Autotagger then is also able to recognize the W 6413 drill or the CV312 connector.

Metadata maintenance in multiple languages

Your assets need to be used on international websites or in stores, but the metadata and tags are only available in German? Then the employees in the foreign country or foreign sales companies cannot use the DAM adequately, because the internal search for tags will not be successful.

In most companies, translating metadata into different languages is a time-consuming and error-prone process that involves exporting metadata from the DAM, passing it on to a translation agency, and then importing the translated version back into the DAM.

With an automated translation solution such workflows are drastically shortened. Our integration solution for the DeepL translation software is suitable for this purpose. DeepL is one of the best machine translation technologies currently available. Its high quality is based on artificial neural networks that understand words in their context instead of translating them individually.

In addition, the DeepL service is 100 percent DSGVO-compliant. Texts are deleted after translation. With the DeepL integration into CELUM Digital Asset Management, editors can have automatic translations performed in 49 language combinations. This can be done after manually selecting specific assets or asset folders, or automatically when importing new assets.

This means significant time savings for editors, while maintaining very high translation quality and minimizing errors. In addition, the costs for a translation agency are saved.

A DAM implementation is work – but it’s worth it!

A DAM system brings structure to data chaos, positively influences internal organizational processes, and pays off in terms of a consistent corporate and brand presentation. Like any new software introduction, a DAM project must be well planned and rolled out in a coordinated manner with all parties involved. Preparation, implementation, and communication take work, require resources, and are not a foregone conclusion. DAM projects are more difficult, when

  • integration with key systems is missing or does not work,
  • no satisfactory solution has been found for tagging,
  • change communication has been neglected.

We have had good experience with repeatedly pointing out these issues before and during projects and providing support when needed. Some things can be solved excellently technically, for others only experience, empathy and clear communication can help. But once the hurdles presented here are overcome, a good DAM is a powerful friend for any organization.

You can learn more about how to overcome the hurdle of integrating a DAM system into the existing IT infrastructure in the first part of this blog post series. The second part is about how you, as a project manager, can ensure acceptance within the company.

Are you facing any of these hurdles? Or have you already mastered them? In any case, we look forward to learning about your experience and answering any questions. Contact us, we look forward to hearing from you!

About comspace

comspace develops and integrates complex web solutions for content management in medium-sized and large companies. As a digital agency with a tech focus, comspace focuses on the highest quality in code and project management. Consulting, development, services: more than 100 colleagues cover the entire spectrum – always with an eye on tech, feasibility and transparency. As a trusted CELUM partner, comspace supports companies in implementing and customizing the CELUM DAM. With its connectors and extensions, the agency ensures optimal integration into existing CMS environments and creates an outstanding experience for CELUM users.

Interested in more insights? Subscribe to our monthly newsletter.

Related Stories


Live Webinar

How DANFOSS engineers the future of product asset data management.

5 December 2023 @ 16:00 CEST

Join us and learn how Danfoss excelled in the realm of asset management.


Guest Speaker:

Mia Rasmussen

Senior Product Owner at Danfoss


Streamlining content creation and management of assets, simplifying online proofing and collaboration, to facilitate a content exploration experience that wows your audiences.

Asset 20

A platform for global brands to manage their digital content, deliver omnichannel, personalised marketing strategies, and overcome complexity in product assets, audiences, integrations and more.

High-End Digital Asset Management

Specialised in mastering the complexity of assets, audiences, integrations and more.

Product Content Management

Centralise, connect, manage in bulk product assets.

Creative Content Collaboration

Give concise feedback. Create workflows. Collaborate efficiently.


100+ applications and extensions accessible via integrations.