Enterprises of all sizes and shapes have their own training departments. If a formal training department doesn’t exist, it is often easy to find an informal one. Training and creating the data that comprise training consumes resources and if the data is not properly managed, the results are a waste of time and poor training materials. There are basic data management tools and techniques that can help.
Enterprises need to train their employees; this is true in most enterprises no matter their size. What use is a new app or service without training? If nothing else, the training material can be used in marketing. Many enterprises don’t have formal training departments though, but rely on teams to train their own people. A quick search, however, will reveal a person or group of persons who create training materials. These persons quickly drown in data.
Consider the process of making a training video: A camera records the video of a demonstration on how to use an app or service; sometimes a person talks during the video to enhance the content; hence, there is a raw video and audio. Next, someone edits the raw video, often cutting small sections to reduce time. The editing may also remove the audio or add more audio. Still images are often pulled from the video to place into other training materials. The result is (1) raw video, (2) raw audio, (3) removed video sections, (4) removed audio sections, (5) added audio sections, and (6) removed but saved still images.
Editing the initial raw video, i.e., one data element, produces six data types and usually a dozen or two individual data elements. Note the trend: the number of data elements to manage is quietly exploding. Note the process: the video editor is creating all these yet-to-be-managed data elements while concentrating on the art of communicating. The science of Data Management is absent.
The above scenario is a simple case. Add the case where an expert records voice only to provide needed information for the training video. That voice recording is also sliced and diced and produces half-a-dozen more data elements.
Now consider the old-fashioned paper documents and their online screen-display cousins. A proven technique for saving resources while creating paper documents is to create small modules of content. These can be single pages of text and illustrations that each explains how to do something of use, e.g., how to add data to a database and how to attach a file to an email. Training organizations then mix and match the single pages into different documents intended for different clients.
It is more efficient is to reduce the size of these training modules to the paragraph level. For example, how to start your email client is a paragraph-size training module that can be used as part of several different email training lessons.
Data Management professionals no doubt have noticed the other side of this small-module technique. Smaller modules means more data elements to manage; if these data elements are not managed, any efficiency gains through reuse of training modules is more than consumed by time wasted trying to find the right module.
The result of these training technologies and practices is a mountain of individual data elements. Each data element resides in a separate file on the training organization’s network. Long, descriptive file names are helpful. Nevertheless, they are rarely sufficient to allow another person months or days later to find the right file for the right situation.
The problem – mountains of individual data elements – is compounded by the people who create the data elements. Note the use of the term “create” instead of “build” or “develop.” The video, audio, and imagery recording and processing people are usually creative types of people, not engineers and scientists and certainly not data managers. They are focused on the art and nuance of persuasion. File names, folder structure, and database entries, while important for them to understand, are isolated and often irrelevant in terms of a greater Data Management system within the enterprise as a whole. The creators are not passing these tasks along to someone else for some other time. They simply don’t acknowledge the existence of the tasks.
Waste, Mediocrity, and Anxiety
The results for the typical training organization and its practices are predictable. The first result is waste. Much of the efforts to create small, usable data elements are lost. When it comes time to alter a training video or document, people search for the pieces they created days or months earlier. The search for the right data element takes time – more time than is estimated. Modifying existing training materials and creating new ones devours far more resources than is affordable.
Another result is a lack of quality in the training product. Creators simply surrender to the annoyance of searching for a desired data element. The statement, “this is not what I had in mind, but it is the only thing I can find,” is muttered as yet another less-than-desired training product is sold to a soon-to-be-disappointed client.
Then there are the persons involved. Searching sometimes finds a desired data element; it almost always finds a headache. Poor quality products also bring their headaches. The creative persons who fashion training products may seem overly casual in their attitude towards Data Management. They are not, however, casual about the result of their work: anxiety rules. They suffer emotionally when something that is less than they desired goes out the door. This anxiety carries into tomorrow and the next day and diminishes their effectiveness and efficiency.
Metadata and a Helping Hand
There are tools and techniques that can help the persons in a training organization create high-quality products. Consider, first, the case of the video, audio, and image files that a single raw video file spawns. What is essential is to capture the processing chain. This chain is the sequence of processing tasks applied to a raw video or audio file.
For example, starting at 15 seconds into a video, 35 seconds of “dead air” is removed, but the dead air contains a possible source for still images so it is saved to a file. The file that was shortened by 35 seconds should be saved with the processing chain, “Input: file abc.avi, cut 0:15 to 0:50, Output: file def.avi.” The 35-second piece removed from the original file should be saved with: “Input: file abc.avi, vut 0:15 to 0:50, cut saved to Output: file xyz.avi.” Next, a still image is taken from the second file at time 12 seconds. This still image should be saved with the processing chain, “Input: file def.avi, still image at time 0:12.” The processing chains should be stored in a searchable database so that the training persons can find the section of video, audio, or still image they desire.
Several commercial video processing apps enable attaching Metadata such as processing chains to files. If the training organization doesn’t have this capability, a data manager can construct a simple HTML file that contains the Metadata and links to the files.
Now consider the creation of video and audio files. Hollywood has provided a simple, time-proven technique: the clapperboard. This is that simple little sign that a person holds in front of the video camera before recording the content. It states the date, subject, and such of the video. Data managers will feel silly holding a board in front of a camera. They won’t, however feel silly 13 months later when they are trying to find that one video they need. An audio version of the clapperboard should also be used. Simply say the date, speaker, and subject at the start of an audio recording.
Next consider the use of text modules for paper documentation. Each module is created so that it can be used in different situations for different audiences. There is the (1) module name, (2) file name, (3) intended situation, and (4) intended audience. This provides four attributes for each training module. A simple, searchable database can hold this information. The training modules are now usable and reusable.
The above tools and techniques are not complex. They will not, however, be applicable to the typical training organization. Remember that the creative persons who fashion the training materials focus on their art. No amount of chiding will change them. Take advantage of their obstinacy. They don’t want to manage the data, so don’t ask them. Have a data manager join the training organization and let this person look over the shoulders of those who create the materials.
Adding a data manager to the training team carries its own risks. Crucial to this technique is choosing a data manager who can become part of the team. It is not straightforward for a data manager to slow the training creators a little now and then while creating Metadata. It is easy for a data manager to anger the creators and cause them to hide their work. That result would be worse than doing nothing.
Training organizations quickly create mountains of data elements that contain valuable information for an enterprise. These data elements easily become unmanageable resulting in waste, mediocre products, and frustrated people. There are, however, basic Data Management techniques that can bring the situation under control. Add a data manager to the training team, and take care so that the data manager has the right personality for the team.