Why a company needs a video at all
People skim website copy but watch a video through — or bail at second five. A company video earns its place wherever a decision rests on impression: a first visit to the site, a meeting with a contractor, a tender submission, a recruitment page.
It has three typical jobs. Remove doubt for a new client by showing that the company exists, with a plant, a warehouse and people. Explain what the business does when the name means nothing. Give a salesperson something to send in a message instead of a link to the About page.
So the first question is who watches and what they should do afterwards. The second is where the video will live: homepage, email, trade show, tender pack. That decides length and format. Without answers you get something handsome and useless.
Three formats that get confused
An image film. Three or four minutes on how the company is put together: people, process, scale. It is watched by people who are already interested — on the About page or during a meeting.
A product video. One or two minutes on a specific solution: what it does, who it is for, how it looks in use. Demonstration matters here, not drone shots of the office.
A tender or investor video. Five to seven minutes with numbers, stages, sites and deadlines. Closer to a deck: captions, labels, restrained graphics, and no music competing with figures.
The same footage often gets cut into versions: full for the site, short for the hero section, vertical for social. Those vertical cuts are made separately — vertical video ads cannot be produced by cropping the sides of a horizontal frame.
The first thirty seconds
The opening decides whether the rest is watched. Three approaches work: show the result, show the customer's problem, show a place competitors do not have.
Show the result — a finished product, a site, a process running. Show the problem — a scene the viewer recognises from their own week. Show the place — the plant, the warehouse, the lab; something that cannot be faked.
What fails at the start: a logo held for ten seconds, a city skyline, generic lines about quality. The viewer has seen all three from a hundred other companies.
A script in five blocks
Block one is the hook: ten to fifteen seconds, an image or a line that explains why to keep watching.
Block two is who you are: one sentence and one visual fact. Not "market leader" but "this line ships X units per shift" — if it is true and can be shown.
Block three is how you work: the process from enquiry to delivery. People on camera, hands, equipment. This is the block that gets shot most often and the one that builds trust.
Block four is proof: sites, testimonials, certificates, timelines. Client names stay out unless cleared — use sectors and task types instead.
Block five is the next step: phone, form, visit. One call to action, not three.
Shoot, generate, or both
Shoot what proves reality: people, production, the product in someone's hands. That cannot be generated convincingly, and viewers notice the substitution.
Generate what is expensive or impossible to shoot: process diagrams, the inside of a product, scenes set in the future, abstract mechanics. AI video production covers those inserts without a location or a studio.
The mixed route is the common one: 70% footage, 30% graphics and generation. In that mode handcrafted videos handle inserts checked frame by frame, while mass vertical videos produce short cuts of the same material for different placements.
If nobody in the company will speak on camera, there is a third path: a video with no faces, built on product and graphics. Cheaper, and no rehearsal required.
How the work runs
Start with a brief and a goal: where it will run, who watches, what should change afterwards. The decision on whether to shoot is made at this stage too.
Then script and storyboard. Signing off before production saves money: changing a line is cheaper than reshooting a production floor.
Then preparation: locations, access, a list of people, equipment, filming permissions. On an industrial site this takes longer than the shoot itself.
Filming usually fits into one or two days. Then editing, graphics, captions, colour, sound and voice.
The last stage is versions for each placement: full, short, vertical, and a silent cut for autoplay.
Timing and cost
A video up to 90 seconds starts at $394 with about a week of turnaround. That is piecework: script, storyboard, generated frames and graphics, editing, voice.
A full company film of three or four minutes is a different volume: more scenes, more approvals, sometimes a shoot day. The quote follows the storyboard, because price depends on scene count and locations.
A series about different business lines is priced per episode on the same basis: each video is produced separately rather than copied with a new title.
Mass formats — short cuts built for testing — run continuously and cost from $2.47 CPM.
If only one version for one placement is needed, the scope shrinks: price tracks scene count, not the number of exported files.
How to judge the result
Skip "I like it" and check three things.
First, retention at the start: do people stay or leave. Second, do they reach the call to action — visible in traffic to the contact page and in enquiries from it. Third, do salespeople actually use it: if the link never goes into a message, it is not doing the job.
Then look at what happens after launch: video versioning extends its life, turning one body of material into cuts for several placements without losing the point.
