You know that it’s important to keep your eLearning files small. The bigger your files, in most cases the slower they will be to open, save, and edit. In addition, unnecessarily large files stand a bigger chance of getting corrupted, especially when you save the file to a network drive.

On top of that, when you publish your lesson you may find that the resulting set of files is larger than needed, making your lesson difficult to deliver across anything but the fastest internet connections. So how do you keep your eLearning files light? Let’s explore the ways.

Your content

The buzzword for the last few years has been microlearning. It has its supporters and detractors. Like any other idea, it makes sense in some areas and not in others. I’m convinced that the main reason that this movement started is that learners complained that lessons were too long, boring, and pointless.

Focus your content on the main point. Don’t crowd it with a lot of extra material. If you feel strongly that those sections should be included, make them optional or accessible through links.

Just get to the point!

Your design approach

You certainly don’t want to create learning that is a lot of text with little else. However, the solution for many designers is to retain all that text and spice it up with images that may be tangentially supportive of the lesson topics. So you end up with lots of pictures of people sitting around a table, executives shaking hands, or a construction worker leaning on a shovel. Do these really help learning happen? I argue they don’t.

What if learning was more like the Choose Your Own Adventure books? Those books contain few, if any, images, but they’re engaging and interesting because after reading a few paragraphs, you have to decide the next step to take, hence jump to page x or y. The resulting story could differ tremendously depending on the path you took.

Isn’t that the point of eLearning? Letting learners make decisions in a safe environment and see the consequences of their actions? I’m not suggesting your eLearning should be all text, but take great care choosing images, audio, video, and other elements that weigh down your learning. They should be helping learners understand better, not distracting them.

Media, of course

There’s no question that the biggest culprit in increasing lesson file sizes is the use of images, audio, and video. However, that’s not an excuse to create just plain vanilla eLearning with text bullets! There are plenty of ways to ensure that your media files won’t weigh down your learning files.


When you import images into your lessons, it’s OK if you need to adjust their sizes a little. However, if you import an image that has a width of 2,000 pixels and a height of 1,500 pixels, and the maximum size you need the image to be in your lesson is 500 x 375 pixels, it’s an excellent idea to first resize the image to the smaller resolution in an image editor before you import it in your authoring tool (Figure 1). Do this with all the images that you import and you’re guaranteed to have a smaller file size.

Photo of sunset over a lake

Figure 1: Resize images before importing. This photo does not need high resolution.

As you’re aware, there are many formats for images, including BMP, JPG, PNG, GIF, and others. The same image stored in one format may have a much larger file size than in other. This is because a file format may be:

  • RAW: Not compressed at all, so the largest file size. Most digital cameras let you save photos to RAW format for the best possible quality.
  • LOSSLESS: Compressed in such a way that you could convert it back to RAW.
  • LOSSY: Compressed in various ways to reduce file and transmission size. Depending on how compressed it is and the method of compression, the image should still look very good to the human eye.

The above all refer to the most common types of images we use and see and are typically called raster, pixel-based, or bitmap. Almost all photographs and many drawings fall into this category.

Important: Check if your authoring tool converts all images you import to specific formats, such as JPG or PNG, regardless of whether the imported images were BMP or another format. If so, you may find that it’s better to convert your images first to those formats before importing them into your lesson for maximum file size savings.

Scalable vector graphics

If your authoring tool supports importing vector graphics (SVG), strongly consider using them instead of another format. SVG are typically images drawn in a tool like Adobe Illustrator, Inkscape, Method Draw, or Adobe XD (Figure 2). The file contains XML-tagged information that allows the browser to reconstruct the image instantly and faithfully at whatever size you like. There is no blurring or pixilation, and the file sizes are very small. Again, these are not for photographs, only for drawn images. In a project I’m working on, almost every image the client provided was in SVG format. The resulting lesson file size is much smaller than it would have been if I had used raster images.

Logo for Loudoun County, Virginia

Figure 2: A logo saved as a scalable vector graphic (SVG).

By the way, scalable vector graphics also come in different formats for architectural drawings, CAD-CAM, and many other fields. The standard for HTML5 is SVG, so you should convert to that format first.


You know that friend of yours that swears that music on record albums is superior to the music on your MP3 player? Well, it’s true, because that audio on vinyl is in analog format, whereas all the music stored on computers and devices is digital, which means it must be converted and, in most cases, compressed. We talk to each other in analog, and our microphones pick up our voices in analog as well, but that microphone is attached to a computer, which converts that analog signal into digital. Conversion shrinks the dynamic range between the quietest and the loudest parts of the audio. It makes the quieter signals louder and lowers the louder parts as well. You may find that this lowers the music’s quality.

We rarely use music in eLearning, though. Most audio in eLearning is narration. The dynamic range for human voices, especially for male voices, is much tighter than for music.

As with images, audio compression can be lossless or lossy. Most music formats are lossy, with MP3 being the standard for web delivery.

You can likely convert those huge audio narration files into much smaller formats without any noticeable degradation in quality. If you look at the properties of a sound file on your computer, you will see that it is stored using a certain bit rate, for instance, 256kbps. Three main factors determine the bit rate:

  • Sample rate: Can be as high as 192,000 hertz or as little as 8,000 hertz. CDs use 44,100 hertz, but I have found that for typical narration, 22,050 hertz gives the same results.
  • Bit depth: Usually set at 8, 16, 24, or 32. The higher the number, the more detail in your audio. For narration, you’ll find that a bit depth of 8 is fine, and you won’t get any extra detail with higher numbers.
  • Channels: Usually this is set to 1 (mono) or 2 (stereo). For most narration, stereo is unnecessary, and you can cut an audio file size in half by saving it as mono.

While there are many audio file formats, the standard for web delivery is MP3 and you may find that your authoring tool will convert all audio you’ve imported to that format.


Just as for audio, there are several factors that determine the file size of your videos. The most important are:

  • Frame size: The width and height of your video. Definitely do not import videos that are larger in width and height than you need. Resize your video to the largest size needed before you import it into your lesson. The difference in file size can be huge.
  • Frame rate: The number of frames per second. Video is nothing more than a series of photos. The number of photos you display each second is the frame rate. While 29.97 fps is standard for television and other professional media, you may find that a frame rate of 24, 12, or even 8 fps will work for your videos, depending on how much movement and action are in the videos. For instance, a talking head will require a much slower frame rate than showing a car racing along a highway.

The audio your video files contain also should be addressed using the suggestions in the audio section above.

Online video

If your video is already resident or if you can place it on a fast video server, either your own or a site like YouTube, you can avoid weighing down your eLearning files by simply linking to the video through a web object in your authoring tool. Almost all authoring tools now support web objects; even PowerPoint allows you to include them. If it’s possible to do so, you may find it very advantageous.

Your thoughts

I have found that using the methods I’ve described help learners engage and enjoy eLearning a lot more than wading through a lot of unnecessary material or seeing and hearing media that doesn’t really add value. When media does add value, having it delivered fast, without delay, also helps to keep learners engaged.

What do you do to ensure your eLearning is light and tight? Write your comments below.