The reason it works is straightforward. A first year business has no track record, no recognisable brand, and no institutional credibility, which means the person is the evidence. Somebody deciding whether to hire you is trying to establish whether you seem competent and whether they would want to deal with you, and a minute of video answers both questions in a way that written copy cannot.
Start with something you already explain regularly rather than trying to produce content. The answer to a question customers ask, a short demonstration of how something works, or a walkthrough of a completed piece of work. Those are things you can say without a script, which removes the largest obstacle, and they are the material customers actually want.
Accept that the first several will be poor and produce them anyway. Everybody is uncomfortable on camera initially, the recordings improve quickly with repetition, and the alternative is waiting for a confidence that only arrives through doing it. Recording five and publishing the fifth is a reasonable approach if that makes starting easier.
Prioritise sound over picture, which is the mistake most people make in reverse. Viewers tolerate mediocre video and abandon poor audio quickly, and any inexpensive external microphone is a substantially larger improvement than a better camera. A quiet room matters more than either.
Keep it short and let the length follow the content. A video answering one question should end when the question is answered, and attention drops sharply past the first minute for anything promotional. Somebody watching a demonstration of work they are about to buy will stay considerably longer, which is why the format determines the length rather than a rule.
Look at the lens rather than the screen, which is the single technical adjustment that most affects how it lands. It feels unnatural and it is what produces the impression of speaking to somebody rather than reading. Everything else about presentation matters less than that.
Consider the alternatives honestly if you genuinely will not do it, because a business that decides against appearing has other options. Photographs of your work with your name attached, written material in a distinctive voice, and reviews that mention you by name all establish a person behind the business. Those are weaker and they are considerably better than nothing.
Then use the video where it does the most work, which is on the page where somebody is deciding rather than on a page collecting videos. A short introduction on your contact page or your main service page is encountered by people at the point of choosing, and that placement matters more than the quantity you produce.
Record several in one sitting rather than one at a time, since the setup is the expensive part and confidence improves within a single session. Four short videos in an afternoon is considerably more achievable than four videos across four weeks, and the later ones are noticeably better than the first.
Write down three or four points before recording rather than scripting it, since reading produces a delivery that everybody recognises and improvises badly. Knowing the shape of what you want to say is enough, and the small imperfections of unscripted speech are part of why it works.
Publish the first one rather than waiting until you are satisfied, because the standard you are measuring against is your own self consciousness rather than anything a viewer applies. Customers are evaluating whether you seem competent and reasonable, and both are visible in a video you consider imperfect.