When a large organization decides to undergo a content migration, it's usually to either implement a new DAM, switch to a new CMS, or to make UX improvements for site users or EX (editor experience) improvements for employees.
Simply put, content migration is moving existing content from the CMS, DAM, or shared drives you use today to a new CMS or DAM. Sounds easy enough, but for mid-to-large-sized organizations, it can take hundreds of resource hours to achieve - even if you automate extraction and mapping.
Not only do you need to locate all your assets and copy (both current and legacy), but you need to audit this content and analyze and map its metadata. You may also need to rename and recut hundreds of image files, and if your re-platforming project includes new page designs, you likely need to create entirely new assets and write new copy.
How do you get started on such a Herculean task? Well, I can't emphasize this enough - content migration is its own end-to-end project. Therefore, start where you would with any project. Find a strong Project Manager, create a project schedule, and get a work team in place. But this PM should NOT be the same person running your platforming project. Content migration is a beast. It needs its own dedicated PM to focus exclusively on asset inventories, audits and be responsible for selecting the best migration strategy for your organization.
At Tahzoo, we use the same migration strategy and methods, whether we are doing the migration work ourselves or only coaching the client's migration PM. Our framework helps clients manage the overwhelming task of migrating tens of thousands of assets and millions of words to new platforms by breaking the work into 3 phases and 7 stages. And we govern our progress through the phases with two 5-point gateway checks. These gateway assessments are not tactical. Instead, they are based on learning objectives to help ensure that no content gets left behind and that all decisions are being made based on evidence and a shared understanding.
PHASE 1: PLAN, IDENTIFY, AND PRIORITIZE
During the first stage of Phase 1, you are going to establish your migration lead, stand-up a migration sub-project, and have a kickoff meeting. Once your project team is in place, you will move on to Stage 2, which is one of the longest phases in the framework. Lots of work happens in these three to six weeks. You will identify all content repositories, work with your IT teams to evaluate extraction options for these repositories, conduct a comprehensive inventory of all your content, componentize those inventories, prioritize your object list, and create a batch plant.
Before you can close out Phase 1 and move onto Phase 2, you'll conduct a gateway check. If you can answer Yes to all five of these outcome-focused questions, you are ready to move to Phase 2.
- Have we established a migration lead, sub-project framework, and held a kickoff meeting with key stakeholders?
- Do we understand how content will be structured for reuse and inheritance within the new platform?
- Have we componentized the UX designs and conducted an object-based needs assessment on all layouts?
- Have we conducted a comprehensive audit of global content based on one or more inventory models? (See Leave No Content Behind: Part 2 for more information on these three different approaches to conducting content inventories.)
- Have we audited and prioritized content to establish P1, P2, legacy, and "leave behind" lists?
PHASE 2: EXTRACT, CONVERT AND TRANSFORM
Phase 2 deals with the extraction of object information from current repositories, conversion of file formats, the creation of attribute maps, and enrichment and transformation of all content objects.
During this phase, you will extract a representative sample of data and analyze the output results. What needs to happen to normalize the extract results for import into the new systems? Do the file types, schemas, attributes, and tags align with your new models? Where do you have data gaps, and should you address those manually or apply custom scripting? You will create your attribute maps and identify net new assets that need to be made. This is also where you want to start planning for content projects that are currently in progress but not yet live. Schedule juggling and resource planning becomes critical in this phase.
The last stage in Phase 2 is Transformation, which can take several weeks. During Transformation, you will create your attribute maps, validate metadata models against your sample sets, confirm that DRM, IPTC, and existing metadata is porting correctly and that everything is correctly set for system loading. Expect to find yourself neck-deep in Excel sheets, code strings, and checklists while you work closely with content SMEs, marketing teams, and the developers implementing the new systems.
Gateway criteria for moving out of Phase 2 include:
- Have we identified all assets by MIME type and know which repositories hold which files?
- Do we have a documented extraction plan for each repository?
- Have we mapped source attributes to target fields and identified all gaps to be filled?
- Have we cleansed, enriched, and deduped all migration files?
- Do we have a plan and schedule for capturing new projects currently in creative development?
PHASE 3: LOAD, TEST AND DISTRIBUTE
I won't spend a lot of time on this final phase because it's relatively standard. In Phase 3, you will load the fully transformed objects into new systems and thoroughly test your results with all SMEs and end-users. This is also the time to add content from any in-progress projects depending on schedule alignment. This final phase's activities and schedule are typically owned by the IT department and a Scrum master. The content migration project lead will often pivot to focus on change management, stakeholder communications, and user training during Phase 3. Solving the Content Inventory Puzzle
Before delving into the tactics of conducting content inventories, it helps to understand what you are looking for. We call this your content inventory puzzle. A good content inventory starts with a basic additive formula, which can be expressed as four pieces of a puzzle.
The first puzzle piece is content you KNOW YOU HAVE. These are images, artwork, videos, copy blocks, and reference documents that you show to site visitors or have in print today. There is no doubt these content objects exist because they are actively used.
The second puzzle piece is content you THINK YOU HAVE. These are assets that are not currently shown to consumers due to seasonality, product availability, or asset lifecycle. For example, you know you have a series of holiday-themed lifestyle photos, even if you don't have them on the website today because it's Spring.
The third puzzle piece is content you know you SHOULD HAVE. This may be content managed by other departments like warranty, support, or regulatory documents, assets created by outside agencies, PIM content being syndicated out to 3rd parties, or regionally produced content exclusive to local markets. Basically, you know these assets exist somewhere, but it will take some legwork to track them down.
When you put these three puzzle pieces together and compare them to your new site designs, you reveal anything missing. This delta is your fourth puzzle piece: WHAT YOU NEED.
This is where the componentization work you did earlier comes into play. From that work, you created a list of all assets and words you need to create pages and documents in your new system. When you take your KNOW YOU HAVE, THINK YOU HAVE and SHOULD HAVE lists and compare them to your componentization master list, you'll find where you need to create net-new assets or write new copy.
Adhering to this three-phased migration methodology and understanding the pieces of your content inventory puzzle are the foundations of a successful content migration project. You now know how to manage your migration project, and you understand what content to look for. The next step is to conduct the content inventories to establish those KNOW, THINK, and SHOULD lists. In Part 2 of our Leave No Content Behind series, we'll explore three different approaches to completing a quality content inventory.