How to Create a Screencam-Based Video Lesson

So you've been charged with producing a screencam-based training lesson for your organisation, and you're wondering where to start. Well, you're in the right place, because this tutorial will tackle the hardware and software needed, plus the planning, recording, editing, and output of the lesson. 

Holy cow, that's a lot to cover. Let's jump right in. 

Planning and Kitting Up

Any well-executed project begins with planning, so let's start there. 

How long should my lesson be? Video hosting service Wistia analyzed 1.3 billion plays of 567,710 videos and found that either 2 minutes (near 70% retention) or 6–12 minutes (around 50% retention) are the optimal durations (Figure 1, below). Thereafter, viewers begin to drop off, although not precipitously. 

Figure 1. Keep your lessons either under 2 minutes or no longer than 12 minutes or so.

Learning platform Udemy is even more aggressive and recommends lessons that are 2–6 minutes long. Not only are shorter lessons easier to watch, but they're also easier to record and edit. So the shorter you can make your lesson, the better it is for everyone. It's smarter to create and consume four 5-minute lessons than one 20-minute lesson. 

What software do I need? This depends on a couple of factors. If you already have a video editor, you might be able to use a free tool to capture the screen (see a list at go2sm.com/screenrecording, plus QuickTime for the Mac), export an MP4 file, and edit in your existing program. However, if you have to add features such as cursor effects, quizzes, or other interactivity, you may need a tool like TechSmith's Camtasia or Telestream's ScreenFlow. 

I've probably created close to 100 hours of instruction over the years, and my go-to workflow is to record in Camtasia, piece together the different recordings in Camtasia's editor, and export the rough lesson as a high-quality MP4 file. Then, I input the MP4 into Adobe Premiere Pro to finalise the video edit, optimise the audio in Adobe Audition, and output using Adobe Media Encoder. 

I use the Adobe Creative Suite because I don't need any of the aforementioned screencam-related features, I'm faster in Premiere Pro, and Audition's audio filters, particularly compression, are more configurable and more effective. I wouldn't buy the Creative Suite just to produce screencam-based lessons, but if you have it, you can use it. However, Camtasia's editor can certainly get the job done, is way easier to learn than Premiere Pro, and includes lots of content and templates to produce highly engaging lessons. (See my review at go2sm.com/camtasia.)

Should I include webcam video? Although this is my opinion only, you shouldn't include webcam video unless you absolutely have to. When I produce a lesson, my goal is to make it look like the audio and screen were flawlessly captured in a single take, which is relatively simple to do if you're capturing audio and screencam, but nearly impossible to do if you add webcam video. Plus, if you add the webcam, you have to record in an environment with a suitable background and lighting, which might be tough for some producers. For example, I record my lessons in a closet off of my office, which totally wouldn't do for a webcam. 

What about audio workflow? Do you narrate while recording the screen? Or do you create the narration and recorded video separately, then merge them together while editing? Here, I go one of two ways. If I'm producing a lesson for a client that needs script approval, I finalise the script, get it approved, narrate it, and insert it into a Premiere Pro timeline. Then, I record the screencam while playing the audio in Premiere Pro, so the audio and video are roughly aligned. If you're producing a single short lesson that has to look its absolute best, try this workflow: if I'm creating a 10-hour course, scripting takes too long, so I ditch the script, record while I capture the screen, and edit the various takes into the lesson. 

How can I work well with others? What if you're in charge of editing and production, but not recording? Either ask the subject matter expert to record and create the rough cut or simply output individual files with editing instructions via timecode (file 1 through 3:12, file 2 through 2:06, and so on). 

What computer do I need? This one's easy; almost any computer will do. I have recorded hundreds of screencams on an HP ZBook Studio G3 with a 2.8 GHz Intel Xeon E3-1505M v5 CPU. Recently, I bought a refurbished HP Elite 6300 Small Form Factor Business Desktop Computer with an Intel Quad-Core i7-3770 running at 3.9 GHz CPU with 16GB of RAM. It cost a whopping $269 and works just fine. Of course, the computer has to have access to whatever program or content you're demonstrating. Other than that, it has to run silently and be installed in a relatively quiet room. 

What about a microphone? You'll need a high-quality condenser microphone, which should cost around $100 or so. It's simplest to get a USB microphone; otherwise, you'll need a separate pre-amp to supply power and the USB connection to your computer. I have an old XLR-based Shure microphone that I pair with a $50 Behringer audio interface.

You want the microphone to be about 6 inches from your mouth, so mics with short desk stands won't cut it. You'll also need a pop filter to avoid plosive P's and sibilant S's. The bundle shown in Figure 2 (below) checks all of the boxes, pairing the highly regarded Blue Yeti microphone with a Knox boom mic arm stand, pop filter, and shock mount. It all retails for $164.99. Wirecutter has a fabulous review of mini condenser microphones at go2sm.com/usbmics, with the Blue Yeti mic as its top choice. 

Figure 2. A flexible mic stand and pop filters are a must.

Configuring the Capture

All capture tools will let you set some basic capture parameters, such as the resolution that's being captured, the frame rate, and the capture format. You can see Camtasia's capture application and resolution selector in Figure 3 (below). There are a gazillion factors that impact screen resolution, including how and where the video will be deployed and the application you'll be capturing. 

Figure 3. Setting the capture resolution

I recommend that you capture at the smallest resolution that allows you to capture the critical action in the lesson, since a 720p video will be much easier to discern on a phone or tablet than a lesson in 4K resolution. If you're producing to deliver to a learning platform, check its recommendations; for example, Udemy recommends either 720p or 1080p. 

While it's possible to change resolution in your video editor, anytime you stretch or shrink a screencam, you risk blurring or otherwise degrading the fine lines in applications and text. For this reason, you should capture, edit, and output at the same resolution. For the most part, whether I'm capturing an application or PowerPoint slide deck, I capture, edit, and render at 1080p. 

Regarding the frame rate (Figure 4, below), most projects work well at 30 fps. From an encoding perspective, screencams are pretty efficient, so there's really no reason to step down to 15 fps. Camtasia supports any frame rate up to 60 fps, although I've never yet had a project that required that frame rate. 

Figure 4. Choosing the frame rate and recording codec

Camtasia 2020 captures all recorded videos in the proprietary .trec format using either the TSC2 or H.264 codecs, with the former delivering superior colour quality but a larger file size. Of course, Premiere Pro can't input .trec files, so, as mentioned, I create my rough cut in Camtasia's editor and output an MP4 file to input into Premiere Pro.