Plan for DAM: Move assets and metadata into your DAM system
Finding a digital asset management (DAM) system is like finding a new house or apartment. First, you look at a lot of places, compare features and prices, and eventually pick a new home. Next step: moving in. In the DAM world, moving your assets into a new system is known as implementation.
This article will outline implementation steps, like revisiting your needs assessment, gathering assets, developing metadata, setting permissions, testing, uploading and cataloging, training users, and launching your DAM.
Yes, it’s a long list. No, you shouldn’t be intimidated. It’s likely you’ve completed some of those tasks already, like assessing needs, developing metadata, and thinking about permissions. Now it’s time to set your plans in motion and get assets into your DAM system.
Moving all those assets from hard drives, discs, and other secret hiding places has great value for your company. The reduction in time spent searching for assets is the number one return on investment (ROI) for DAM users. You might also spend less on stock images and color correction services, and you can also avoid fees and fines for misuse of licensed assets.
You may have done an internal assessment while you were searching for DAM solutions. They are common (and encouraged) during the research stage because they help an organization define its needs and digital asset use cases before meeting with demo vendors.
Like during a physical move, figuring out what you have and where it’s kept now is an important part of determining where to put things once you get into the new place.
Here’s a quick outline of important questions:
- How many assets do you have and where do they reside? Make a detailed list of the sources. You’re going to need it.
- What kind of information (metadata) is associated with those assets? You might get a range of answers, such as: “What do you mean by metadata?” (i.e., The assets are all in a jumble and we just browse to find them), a detailed folder structure, and “we have a DAM, and I’m not letting you touch it.”
- How do you search for assets now? Closely related to the previous question since metadata is often used to search, but addresses the process from a different angle. Some people store assets, some people only retrieve them. It’s worth asking both questions to get the best answer.
- How do you use the assets? This will give you ideas about different levels of control your company will need. For example, if creative services is producing print ads, they probably won’t want human resources to use the assets. In fact, it might run counter to a licensing agreement for someone else to use the image.
- How many people will need access? And what kind of access? These questions have complementary goals. The first is obviously just to get a handle on how many people will need access, but the follow-up question will again inform your permissions. Will those people need to be able to upload assets? Update the metadata?
For more information on needs assessments, see our blog posts “10 short questions that assess your DAM system needs” and “One simple step to finding a digital asset management solution.”
Interacting with DAM users has benefits beyond determining how to organize the system. By reaching out to potential users, you are developing the user base for your DAM solution. Getting them involved — and listening to their needs — increases the likelihood that they will use the system once you have it implemented.
Regardless of the monetary ROI, showing management that people are using the system is the biggest indicator of a successful deployment. Get department heads involved early, make the system address their needs as best you can, and you will have made great strides in ensuring the system is adopted.
Gathering digital assets
There are two rules of packing (your belongings or digital assets):
- You have more than you think you do.
- It will take longer than you think it should.
Get the biggest hard drive you can get your hands on tomorrow and start copying whenever the opportunity presents itself. Carry that drive with you around the office and gather everything you find. Create a folder for each department and copy away. Inside the departmental folders, keep their folder structure for the time being. Don’t spend too much time analyzing what you have. Wait until you have most/all of the assets so you only do the analysis once.
You should also gather metadata that exists for the assets. Maintaining the folder structure is one way. You’ll also want to gather spreadsheets and keep them with the assets they relate to. For example, if PR keeps a list of all the images and who is in them on a Google Drive document, get a copy of that document and save it into the folder with the PR assets. If you have an existing DAM system, start talking to IT and/or your vendor about how to extract the metadata.
At the most basic, metadata is data about data. Metadata often includes the following fields of information:
- Metadata type (marketing, sales, etc.)
- Story (why the digital asset exists)
- Description (what the asset is about)
- Type (photo, video, document, etc.)
- Content source
- Rights management
Metadata topics can be grouped these types:
- Descriptive is what you’ll spend the most time thinking about. That’s information that describes your asset: locations, people, colors.
- Administrative helps you manage your collection of assets. Think expiration dates, licensing information, and geographic restrictions.
- You’ll also hear the term structural metadata. That is related to writing code and being able to navigate pages.
A more detailed description of metadata (and related terms) is available in the post “Taxonomy, metadata, & advanced search capabilities.”
Another term you should be aware of is schema. Schema refers to all the metadata fields and the values attached to them. Your schema could have six fields or 60 fields (though that many fields is not recommended). For example, if your organization applies metadata to indicate which department developed an asset, the metadata field would be department and the values would be the names of the departments, like marketing, HR, etc.
If the list is predetermined, as in you have an unchangeable list of values, that is a controlled vocabulary field.
Back to the assessment we discussed in the first step. The questions you asked about how your users will search and what kind of assets they look for will be a great source of ideas. Keep a list so you can see what terms show up most often.
Other places to get ideas for metadata fields:
- Your company website
- A legacy DAM system
- Social media
- A Content Management System (CMS)
- Other companies
- Your DAM provider. Ask for help from them!
Building your metadata list
Combine your notes, lists, and ideas to identify the most important metadata fields. You’ll find you need to balance thoroughness with ease of cataloging. That is, you could create a very detailed schema that has 25 fields that have to be populated for each image, but do you have the staff to handle the work that goes with that much cataloging for an image? And does the staff have the skill and knowledge to manage the job?
Alternatively, having a schema that is designed to make image cataloging easy for the staff may not be thorough enough. If there isn’t enough metadata, users won’t be able to find what they want.
Test and tweak
Get feedback on the metadata structure you use before you can call it complete. Your boss, stakeholders, and future users are great people to have review a metadata structure. Once you’ve deployed the initial structure you can do some testing and give the same people a visual example of how the metadata will look and how it will support searching.
To test, start by uploading a group of assets. Make sure to add TEST_ to the beginning of the filename so you can find the assets quickly and remove them from your system before launch. Next, come up with some example searches. Find terms that cross your metadata fields. For example, you might have a product called “Tulip,” you have multiple pictures of it. You also have pictures of fields of tulips. A search for “tulip” will return both types of images. This is important because you’ll be able to show how the filters will help testers find assets of the product quickly. Now you’re ready for some demos.
During a demo, you can show how the metadata works to help users find assets. You can do demos in a group or work with individuals. Both methods have benefits. Whichever you choose, take lots of paper with you because demos are a great time to get feedback and ideas from the people who will be using the system.
After the session, institute changes that make the system more usable. You might find that some suggestions sound good, but also be prepared to explain why something won’t work. Do demos for new people so you get fresh feedback — and also so you keep everyone involved and engaged.
At some point, you’ll have to stop tweaking your structure and start uploading assets, but you should review your metadata on an annual basis.
Upload and catalog your assets
This is the part of the move where you load all your personal belongings into a truck and drive across town (or the country depending on your move). Once you are there, it’s time to unload. You might have a system: all the kitchen stuff gets stacked in a corner next to the fridge, bedroom stuff along the back wall of the bedroom, etc. You’re cataloging things as you unpack them.
In a DAM move, while you’ve been working on the metadata, behind the scenes you’ve also been carrying around a big hard drive, getting digital assets from personal computers, shared drives, and more. Hopefully, your drive is full about now. Once you get your assets all in one place, here are some things to consider.
How will you upload? Your assets and data probably fall into one of these five categories:
- Assets with embedded metadata, but no other information. Photographs usually always come with capture date as the minimum in embedded metadata.
- Assets with a filenaming convention that allows you to extrapolate metadata, like “Fall_2015_newsletter_header.jpg.”
- Assets stored in a folder structure that provides meaning. For example, the product department might store product images in a folder for each product and a sub-folder for the year. This structure can help you apply metadata.
- Assets without metadata. Zip, zero, nada.
- Assets with metadata associated with them via a spreadsheet. A wonderful thing if you have it.
You may have assets in each of the categories, which means you’ll have to decide how to address each group. Below are my recommendations. I’ve arranged the categories by desirability.
Best-case scenario: Assets with metadata associated with assets via a spreadsheet
If you’re in this situation, you are probably migrating from another DAM system — or you’ve got an information professional on staff, in which case you are very lucky. In the first case, your biggest challenge is getting the data out of the old system, then mapping it to the new system.
Next best case: Assets with embedded metadata, but no other information
Your new DAM system should be able to scrape the embedded metadata and put it in whatever fields you want. There are an array of common embedded metadata fields, such as tags or keywords; photographer or author; caption; and title. Work with your vendor to identify the fields that are embedded and to map those fields to a field in your new system.
Somewhere in the middle: Assets with a filenaming convention that allows you to extrapolate metadata
The uploading options are much the same for assets with filenaming conventions. When you upload files that use a naming convention, you can easily run searches on filenames and make bulk updates. You can do an advanced search of filenames, looking for codes your company uses and update the appropriate field.
OK situation: Assets stored in a folder structure that provides meaning
You will want to upload these on a folder-by-folder basis. Your DAM system should either let you assign metadata as you upload each folder or will allow you to make bulk changes to the assets once they’re in the system.
A long row to hoe: Assets without metadata
Sorry if you fall into this category, because it often means more work for your implementation because you have to look at a lot of assets and tag them appropriately. It’s also unlikely that all your assets have no metadata. I’ll spend a little more time on this section, since it requires more work.
When it comes to addressing assets without metadata, you can start in one of two ways:
Start sorting assets before uploading them to the DAM system.
Upload everything and do all the sorting and tagging within the DAM system.
If you decide to sort before uploading, here are a couple ways to begin the pre-upload work:
Create a spreadsheet with all the files and a column for each metadata field you are going to use.
Rename all the files using a filenaming convention that will add meaning.
Embed metadata into all the assets using a tool like Bridge or Photoshop.
Create a folder structure and drag and drop all the assets into the right folder.
If you wait to sort and tag after uploading your metadata-less assets, you have a lot of work to do in the DAM system but you can do the changes in bulk. And if your new DAM software has the capability, you also have the benefit of weeding duplicates.
Did you notice a trend with the upload options? None of the upload methods (except the spreadsheet) result in complete metadata. Yes, you are going to have to do some work — what I call cataloging — to complete the metadata after you finish the upload. The term comes from the library world, so it carries a certain amount of geekweight, but it’s very accurate. Webster says catalog (transitive verb) means to classify (as books or information) descriptively.
Fully cataloging your assets will have two main phases: bulk updates and updating individual assets.
Bulk updates mean making the same change to multiple assets at one time. It could be a batch of two assets or it could be 500. In this phase, you will start with the big groups of assets and work your way down to smaller groups. You may be doing searches against specific fields and moving the data to a new location where it can be used effectively as a filter. You may be searching for assets with specific attributes in the filename (this is where the naming convention first shows its value) or you may be searching for a city field so you can fill in the country field that wasn’t filled in before.
Get down to the nitty-gritty
It’s both a natural and logical path to start with the biggest batches of bulk updates (like you move in the big furniture first) then work your way down to very fine-grained changes. You will probably find that the bulk changes are related to fields with controlled vocabularies, and as you get to the more fine-grained changes, you will be dealing with free text fields like keywords.
You use the keywords field to capture information that doesn’t fit neatly into one of your other fields. If you find yourself using the same term over and over, you may want to add another field or add it as a value to an existing field. Generally, you’ll be adding keywords to a few assets at once and maybe addressing one image at a time. It’s time-consuming, but worth it. And when you get down to the single image updates, you are putting the finishing touches on the implementation. Your metadata is strong and the users will thank you. Maybe not in so many words, but they will appreciate it.
A note about permissions
Nearly every company has groups of people who need access to one set of assets, but don’t need access to assets in another group. There are a lot of cases where this is true. Here are a few.
- Geography. Pictures taken in Barcelona may only be useful to your Europe office. People in North America and Africa offices will never use those images; you don’t want to clutter their DAM experience with images that are not relevant to their work.
- Departmental. Research and development add pictures of prototypes to their corner of the DAM. They would like to restrict access so that the marketing and products teams don’t have access to unapproved prototype images.
- Usage. Images licensed from a stockhouse have certain limitations that must be respected. Design-oriented staff may understand that, while more junior members may not get the nuances.
You should have uncovered tips for this in your discovery phase. Review your notes or go back and talk to people again to get a better understanding of who needs to see what in the DAM.
Different DAM systems have different permission structures. Some assign permissions on a folder level, so you’ll drop assets into a folder and give users access to the folder. Other systems assign permissions on an asset level. The theory is the same, and your vendor should be able to walk you through the process.
Before your official launch, you should spend some time testing. This is a great time to go back to the stakeholders and others whose shoulders you tapped during the assessment.
If you were moving into a new house or apartment, beta testing is the part where you move furniture around to decide where you like it. You might bring in stakeholders like friends and family to weigh in on where to put the couch to best avoid a glare from the windows.
A good strategy is to create a script that takes them through specific tasks, like searching, filtering, downloading, uploading, and other common tasks. Assign different permissions to your testers to make sure they can access the right (not the wrong) parts of the system. Also, find testers who use different browsers and platforms and have a variety of access to images.
On the script, give them room to make comments and respond to the tasks. Ask them to search for a tulip and mark down how many assets in their results. Download an image as a JPG, and mark down how long it took to convert. Use filters for city and country, and make note if it was intuitive and worked as expected.
Beta testing is another important point in your adoption strategy. You’ve talked to your stakeholders two or three times, and they’ve seen the system change and improve along the way. Hopefully, it now does everything that they have asked of it. You’re almost ready to launch!
Training and outreach
Training and ongoing outreach to new users will keep your DAM system in demand. Here are a couple tips for training success.
For your launch, consider group trainings. Plan about 30 minutes to get through the basics — don’t try to pack all the info into one session — but book your room for an hour so there is plenty of time for questions and to go back and cover some of the bells and whistles in your system.
If your vendor offers training, use it! They know the system inside and out and have the training script down pat. They’ll be able to give that 30-minute overview, but you should be there to talk about the specifics of your deployment. You also know your metadata better than the vendor, so you should be the one to communicate that with users.
After your launch, offer regular training sessions to new users. You’ll want to get as many people in the room as possible, though you may also want to target groups with specific needs. For example, sales needs to find the latest product pictures and may need to know how to identify the latest version of assets. You may want to do a special session for them, or you can target specific people in the department who you can teach and let your lessons trickle down to the staff.
While you’ve done all the work to set up and launch your new DAM, you aren’t done. This is a good thing. Just like a new home, your DAM doesn’t have to be perfect from the start, but you need to have a plan in place for reviewing which filters are working and if all the metadata fields are useful.
You can also look ahead to some functionality you didn’t originally implement, like single sign-on or an integration with your CMS. After your system has been running for six months or more, start to leverage features like analytics info (if it’s available to you) to get real data on how people are searching the site and what key terms may be missing. In short, a DAM solution is a living thing that can flex with the needs of your users over time.
This paper covers an implementation from a pretty high level. Every implementation is different, and every deployment has quirks that are specific to the company or the assets. Hopefully, you’ve been able to pick up some basic theory that will help you move forward with confidence to revisit the assessment you made and take all the steps — gathering assets, developing metadata, testing, uploading, cataloging and training. Those steps will result in a DAM system that will benefit your company for years to come.