Thanks for the info and experience. Very interesting.
From what I know, the content management sector has a problem of being too generic, trying to handle any form of content, including digital media content, which has very different requirement than documents. Version management is low in the requirement list. Also indexing and searching are a big part of the requirements, which lead to the emphasis on OCR and different format conversion. Archiving and document retention are pretty big with emphasis on the compliance with the various laws. This leads to bloated products that the customers find hard to use and hard to fit into their process and workflow.
I heard of effort by law firms trying to implement content management systems but couldn't because of diverging requirement. At the end just threw the printed PDF into the system and called it for the day.
I heard of contract version management system, just to track the changes to the text of a contract, which was pretty successful because of focused effort. Contract can have numerous versions during negotiation and small changes can have huge impact.
What we need is a system with a small well defined feature set focusing only on the version control of text documents. Outsource the format conversion, OCR, index and search work to other systems.
You are absolutely correct. What typically happens is a company will bring in a focused solution like a contract management system. Because of its focus, it is able to do its job pretty well.
As that tool is successful, people will start thinking, "I sure wish I could do that with my [Word docs, Photoshop files, video files, etc]," and they start looking. Then they find Content Management and think they've found the perfect solution (and it really does sound good on paper).
But there are three fundamental problems: Unlike the contract management system, you generally don't use the CMS to create or edit documents[1], so you always have a point of impedance between the authoring and the management. Now you are relying on people to "do the right thing", and the right thing isn't the easiest thing.
The next problem is managing versions of the documents gets ugly and confusing pretty quickly. Most file formats we use today don't diff well, so the benefits of e.g. seeing just the changes from v2 to v3 are hard-to-impossible to implement.
The final problem is the content lifecycle of most documents is very fluid. Whereas a contract has a well-known lifecycle with a point in time that it is DONE (e.g. the contract is signed) and a well-known set of approvers (companies generally understand their money-spending approval process well), most documents people interact with follow very ad hoc processes. Invariable, people try to build formal processes out of those ad hoc processes, and people get frustrated. Given the first problem (authoring outside of management), people can easily make an end-run around the tool, perhaps putting a document into it only when they absolutely have to.
Anyway, I could go on for a long time, but my take-away from my experience in content management is to solve small, well-defined problems and forget about the siren's call of "all our documents managed in bliss" because it won't happen.
1. Like contract management, web content management tends to be more successful, because the authoring environment is usually the management environment.
Comments
Thanks for the info and experience. Very interesting.
From what I know, the content management sector has a problem of being too generic, trying to handle any form of content, including digital media content, which has very different requirement than documents. Version management is low in the requirement list. Also indexing and searching are a big part of the requirements, which lead to the emphasis on OCR and different format conversion. Archiving and document retention are pretty big with emphasis on the compliance with the various laws. This leads to bloated products that the customers find hard to use and hard to fit into their process and workflow.
I heard of effort by law firms trying to implement content management systems but couldn't because of diverging requirement. At the end just threw the printed PDF into the system and called it for the day.
I heard of contract version management system, just to track the changes to the text of a contract, which was pretty successful because of focused effort. Contract can have numerous versions during negotiation and small changes can have huge impact.
What we need is a system with a small well defined feature set focusing only on the version control of text documents. Outsource the format conversion, OCR, index and search work to other systems.
You are absolutely correct. What typically happens is a company will bring in a focused solution like a contract management system. Because of its focus, it is able to do its job pretty well.
As that tool is successful, people will start thinking, "I sure wish I could do that with my [Word docs, Photoshop files, video files, etc]," and they start looking. Then they find Content Management and think they've found the perfect solution (and it really does sound good on paper).
But there are three fundamental problems: Unlike the contract management system, you generally don't use the CMS to create or edit documents[1], so you always have a point of impedance between the authoring and the management. Now you are relying on people to "do the right thing", and the right thing isn't the easiest thing.
The next problem is managing versions of the documents gets ugly and confusing pretty quickly. Most file formats we use today don't diff well, so the benefits of e.g. seeing just the changes from v2 to v3 are hard-to-impossible to implement.
The final problem is the content lifecycle of most documents is very fluid. Whereas a contract has a well-known lifecycle with a point in time that it is DONE (e.g. the contract is signed) and a well-known set of approvers (companies generally understand their money-spending approval process well), most documents people interact with follow very ad hoc processes. Invariable, people try to build formal processes out of those ad hoc processes, and people get frustrated. Given the first problem (authoring outside of management), people can easily make an end-run around the tool, perhaps putting a document into it only when they absolutely have to.
Anyway, I could go on for a long time, but my take-away from my experience in content management is to solve small, well-defined problems and forget about the siren's call of "all our documents managed in bliss" because it won't happen.
1. Like contract management, web content management tends to be more successful, because the authoring environment is usually the management environment.