Migrating Documents into Your SharePoint Solution?

When implementing a SharePoint solution, a discussion about document migration is inevitable. Migration is never an easy task. Document sources and owners need to be identified, not to mention WHERE in a document repository the documents reside. Within SharePoint, documents are associated with a Content Type and described by a series of Content Columns or metadata. As documents are migrated to SharePoint Document Libraries and Folders, the difficulty comes from assigning Content Types and metadata to the source documents.

If you are fortunate enough to be migrating documents from a controlled respository that already has a predefined taxonomy, then you can use an automated process to move the documents from the source to the target. If not, then you need to define a process (manual or automated) for mapping the documents to the Content Types and assigning the appropriate metadata. Many organizations struggle with content migration. Some have trouble identifying the source and accuracy of their content; others struggle with mapping the content to its target destination.

I’ve witnessed a variety of migrations over the years. Here are two recent migrations in SharePoint implementations and how I tackled them.

The first is a company that implemented Windows SharePoint Services as a platform to support ad-hoc requests for collaboration and teaming functionality. Their SharePoint environment grew without planning or design. When we were asked to develop a migration strategy, we found a SharePoint environment that grew to more than 800 gigabytes and 1,200 site collections.

No quotas or controls?
There were no quotas or controls on the site collections. Each site or site collection was unique without any consideration for consistency in look and feel of site functionality. Users were relegated to adding the sites to their favorites list as a means of site navigation document location. Another problem this company faced was cross-site searching. Without cross-site searching, information was often lost and new sites were launched that duplicated information that was already stored in SharePoint.

In the second case, the company was implementing SharePoint into an environment that did not have any controlled repositories. They wanted to migrate documents from file systems, both personal and shared drives, into their new SharePoint solution. The problem: they didn’t know where all documents were stored. Not only did they not know where the documents were stored, multiple versions of the same documents were spread across multiple file stores, leaving the organization unsure as to which version was the correct one for migration.

In this (or a similar) scenario, it is critical to identify a single version of documents that are to be migrated in order to establish an accurate inventory. Only then can the migration happen. Tools are available that compare documents or produce a likeness ratio. However, extensive user involvement was required to produce the document migration list.

Successful migration
Successful migration requires a methodical approach to create a map from source to target that clearly defines the source document, its target location, and the Content Type and metadata for that document. To address migrations, I believe in a four step methodology: Document Identification, Document Mapping, Document Preparation, and Migration Processing.

Applying this methodology to both examples, we were able to create a spreadsheet that inventoried the documents that needed to be migrated and a mapping of those documents from their source to the target. In the case where we were migrating documents from one SharePoint environment to another, we used a migration tool that took our mapping spreadsheet and used it to automate the movement from the source to the target.

In the case of the migration from file stores, we used the spreadsheet to populate our SharePoint repository with the correct Content Types and metadata and then pointed to the associated files for bulk upload. Using a defined, planned approach enabled us to migrate what could have been viewed as a “mess” into an ordered, controlled repository.

Alan Weintraub (alan.weintraub@perficient.com) is a Principal, ECM Solutions for Perficient (www.perficient.com). Alan has extensive experience in all phases of Enterprise Content Management solution implementations. He has worked as a Research Director at Gartner, focused on content and document management markets and as a consultant, where he designed and implemented document management systems. Prior to his consulting experience, Alan engaged in technology management for major pharmaceutical companies. He has over 25 years of experience in the information systems profession.