Content Management and Capture

 View Only

Six reasons why you shouldn't DIY Content Management

By Sydney Adams posted Mon March 20, 2023 12:03 PM

  

“Why do I need a content management system when I can just use S3 / Hadoop / NoSQL databases to manage my critical business documents?”

More often than not, these questions originate from a perception that a content management system (also referred to as content services, enterprise content management or ECM) is doing a relatively small number of things – storing content, managing some metadata, and applying some security. Content services solutions do these things….and much more. Many content services capabilities are often overlooked due to a lack of knowledge about the capabilities needed to effectively manage content throughout its entire lifecycle, a lack of understanding about what long-term functional and operational content management needs will be (in terms of months, years, or even decades), or a combination of both.

Content services helps you avoid extensive custom development in these six areas:

  1. Without content management ….you would need to define a document model and metadata fields in a database for each application that might require a different set of properties

The assumption that having “metadata” is simply the ability to define properties associated with a document is incomplete. Most content services systems will have a much more sophisticated set of capabilities including (but not limited to) custom property definitions, encryption options, or the ability to affect security based on property values.

  1. Without content management….you would need to write custom code to implement robust document security at several levels

Security is another area of functionality that is often oversimplified and understated. Security requirements can vary significantly based on industry, compliance requirements, municipal, state, or federal agency requirements, and many other contributing factors. A robust set of security capabilities is required to control access to documents by end-users, based on a wide variety of factors and conditions, including (but not limited to) a document’s metadata attributes, an individual user’s role within the enterprise, or the ability to redact information from documents to further restrict what an individual and / or role is able to see.

  1. Without content management….to support retention, legal hold, and governance requirements in regulated industries you would not only need to write custom code, but also ensure it was fully audited and certified to meet those regulatory requirements

Governance policies span a range of functional services, including proper identification and classification of documents, identification and declaration of certain documents as business records, and an appropriate disposition and deletion framework based on lifecycle policies. On the surface, it may seem relatively simple to set a retention date and then delete the document when that date is reached. In reality, retention policies can be quite complex and require the ability to manage many different types of policies - and the policies themselves are driven by different document classifications or even metadata attributes that provide more information about a document’s characteristics. Sometimes, a retention date isn’t even known immediately because the retention cannot be determined prior to a specific event.

  1. Without content management….you would need to define a set of APIs to enable applications to store, search, retrieve, update, version, and check documents in / out across your organization

A wide variety of document types get used today, and an ECM solution will provide a much more holistic set of capabilities to enable ubiquitous access to content. This idea also applies to the idea of transforming documents from one type into another (with the most common scenario by far being transforming a document into a PDF) or document manipulation (annotating, redacting information, editing, splitting, or merging documents as part of the viewing and editing process).

  1. Without content management….you would need to implement a custom system that performs well at scale.

All of the functionality listed above needs to be provided in a highly scalable manner. Providing good performance for end-users is often measured in response times being less than one second for most operations. While it’s certainly possible to implement a system that scales and performs, doing so across such a broad range of functional capabilities should not be underestimated. It represents a significant amount of development time and energy to achieve. The only effective way to achieve this goal is to build a relatively comprehensive and sophisticated set of both functional and scalability test scenarios. Each of those can equal or exceed the initial functional development cost of the features themselves.

  1. Without content management….you would need to write a user interface that business users could use to interact with content.

This user interface would have to be smart enough to only display the metadata fields that were relevant for a class of documents, support custom searching, and implement your security requirements. You would probably also need to write an administrative interface to define classes of documents, metadata properties, retention, and the other document attributes mentioned above. The user interface would also need to include a viewer and collaboration tools to interact with content in all the ways described above - annotating, transforming, templates for adding new documents, easy ways to edit and version documents in your favorite applications….you get the idea.

A true content management system will provide the capabilities listed above - and more - out-of-the-box and require little to no custom development. In addition to the heavy lift of custom development in the categories outlined above, it is important to remember that maintenance is likely to be the largest challenge over time. Content management systems are tricky to maintain and upgrade because of the need to provide compatibility and an upgrade path for large volumes of information and metadata. The cost of maintaining a system like this over numerous years will typically be much higher than the initial development cost of the system.

By having a better understanding of the complexities involved with the long-term management of a highly performant, scalable, secure, and properly governed content management system, you’ll have a much higher likelihood of success with implementing one.

This blog is an adaptation of the paper “Considering building your own ECM?” written by @Mike Winter. It has been edited and shared on the IBM Community by @Sydney Adams and Mike Winter. 

1 comment
73 views

Permalink

Comments

Sat April 01, 2023 02:14 PM

When it comes to home grown content management solutions, there has been a trail of disasters during the past couple of decades, resulting from ambitious projects that end in technical cul-de-sacs.