I have been exposing the many layers of the “chaos” that resides within training content workflows over the past few weeks. Today, I want to share with you the first of many examples of “chaos” that I personally have experienced in my own work. It is the story of how one of my colleagues spent 3 hours of his life searching for what he thought to be the “final” compliance module for a specific program that he was working on. After digging through multiple folders on our shared drive, he finally found what he was looking for. Or so he thought. It turned out that not only was he one of the two designers working on the program, but he was also updating a copy of the content that would be delivered to learners, while the other designer was updating a copy of the same content as well. As you can probably guess, there were updates to the content that the first designer did not know about, and when the content was published for the learners to take the program, there were two lessons that contained conflicting information. It was not until a learner found the discrepancy in a live session of the training that anyone found out about the problem.

I like to tell the following story from a few years ago: A colleague of mine spent 3 hours of his time searching for the “final” version of a compliance module. It turned out that 2 instructional designers had been working on different versions of the instructional content for the past week or so without the other knowing that the other was working on the same module. In the end the training was deployed

The chaos hiding inside content workflows

Training content can quickly spiral out of control. A single onboarding program can grow into a dozen or more modules, plus all the supporting job aids, quizzes, tests and more. And the content is spread across multiple locations (like a shared drive or email) and project management tools. This is the kind of “content chaos” that an orchestra of musicians writing their own sheet music in pencil would look like. And, with no eraser in sight, it’s a mess to manage.

While many believe that the greatest challenge to organizations managing learning and performance is that learning content multiplies out of control, the reality is that by far the greatest problem is that of version drift. Such as in the story above where a writer in a team updates a product screenshot in Lesson 4 of an organization’s onboarding program. The person who maintains a related PDF-based job aid continues to update the job aid mirroring the old content in Lesson 4. Six months pass and two apparently identical pieces of information turn out to be in conflict. A learner from the organization using the information from one of the sources of apparently identical information takes action in his work with potentially serious consequences for the organization.

Training content multiplies or grows in the form of modules, job aids, assessments or even supplemental training materials for a particular onboarding program. As more content is created, it spreads across a shared drive and the various teams can end up working with a number of different files. To effectively and scalably manage training content within a training team, a system for tracking, storing and managing content must exist. This is what an LCMS or learning content management system does.

What “reusable components” actually means in practice

An example for a learning object is a safety procedure. This safety procedure is part of the New Hire Orientation program as well as part of the Annual Recertification program for existing employees. For Manager’s track a slightly modified version of the safety procedure is part of the learning program as well. The mobile just-in-time learning module also contains a version of the safety procedure. Today all four learning objects contain a version of the safety procedure that has been copied and pasted as a text block in the respective learning object. All four learning objects were developed by different people at different times. Chances are that not all four learning objects will be updated when the safety procedure changes. This means learners of an organization will have the worst information possible while they believe they have the best information possible.

Now think about a safety procedure that is used in a number of training courses. For example, it might be included in a new hire orientation course, annual recertification course, manager’s track, and in a number of just-in-time (JIT) mobile modules. Right now that safety procedure is likely to be a number of separate text blocks – probably created by copy and pasting from a word processor by the person who wrote the procedure. After that, the writer probably won’t give a second thought to that procedure again. But if the safety procedure changes at some point, all of the copies will need to be updated. This can be difficult for a writer, but it is an architectural problem for the LCMS to solve.

An LCMS would treat that safety procedure as one piece of content, that is used in four different pieces of training. So, if the safety procedure was changed in some way, then that change would automatically be made to all four pieces of training – including new hire orientation, annual recertification, manager’s and a just-in-time piece of training that’s delivered on a learner’s mobile device. For training created by training teams that are dealing with compliance issues, that is regulated in some way, that create lots of pieces of technical information, it’s not a feature that’s in a particular price point. This is a fundamental part of the work that an LCMS does. It’s the difference between a workflow that is manageable and a real liability as time goes on.

Where training teams typically feel the difference

• Translation and localization: Revising a single source object instead of excavating through twelve language files one by one
• Compliance updates: Rolling out regulatory changes across every affected course in hours, not the weeks it currently drags on
• Onboarding content: Keeping product information accurate when your offering shifts every quarter (and it always does)
• Multi-format publishing: Pushing the same content to an LMS, a PDF, and a help center without manually reformatting everything from scratch each time

The collaboration question nobody asks early enough

When Training Teams Wait Until Something Breaks. Most training teams wait until something breaks. A painful audit finds non compliance with regulations. A training program is found to have a compliance gap. A project is finished to find that for three hours in the middle of the project two people were working on the “final” version of a compliance module, each updating a different part of the content that the other would later include in their version. And nobody found the conflict until it was too late. The training was conducted with learners receiving conflicting information in two different lessons. The fact that the conflict was found in a live session was a lucky break. That had to have been painful to fix. The pain would have been worse if it had been found in a recorded version that was then sent to many people.

An LCMS supports collaboration by supporting the structured roles of writers, subject matter experts and approvers for the review of content. In addition, every step of the review process is tracked. A log of the review process is maintained including a timestamped record of each action by each person involved in the review. The review process of a team that is using an LCMS to manage their learning content is one of the key reasons why such teams are able to work as a team effectively even as the team grows. For small teams of people who already know each other the structured process of a review in an LCMS may seem to be excessive but as the team grows and people leave the organization it can be a real lifesaver to prevent slow down of the team.

(One final note here. I’ve seen teams try to fight the kind of structure an LCMS demands when teams are small and everyone on them is familiar with each other. But as teams grow and people leave the kind of structure an LCMS demands becomes more and more accepted. The chaos an LCMS prevents is much slower than the chaos you’d be in trying to work around an LCMS.)

Workflow stage	Without an LCMS	With an LCMS
Content updates	Manual edits scattered across multiple files	Single source object updates everywhere at once
Version control	File naming conventions (v2_FINAL_revised_USE_THIS)	Built-in version history with rollback capability
Review and approval	Email chains and comment-laden docs	Structured workflows with tracked, attributable sign-offs
Multi-format publishing	Manual reformatting for each output	Automated publishing to multiple targets simultaneously

It’s not magic. It’s architecture.

This brings us back to the story of the compliance module. Yes, the conflicting information was found in the end and fixed. But the real damage was the three hours of someone’s life spent searching for a file. The discovery that two people had been working on separate versions of a file for six months. The damage to trust that occurs when people can’t be sure if they’re working with the latest version of a document. This kind of damage accumulates slowly over time. Until, one day, you realize your organization is operating on a background hum of uncertainty about whether or not your training materials are current.

And last, an LCMS is not a fix for problems with review processes or content ownership that are amorphous at best. These are organizational problems that can’t be ‘solved’ with software, but a good LCMS removes a lot of the technical friction that makes it harder to establish and sustain good processes than it does to put in place and sustain poor ones. (If it takes your team more hours to manage your learning content than it does to create learning moments in the first place, then you need to ask yourself why that is.) The content is not the problem. The system that the content was developed for is.

So, if your training team is spending more time than necessary managing files, that’s a problem you should recognize and do something about. The content isn’t the problem; the system in which the content resides is the problem.