← Blog

Interactive demo, video, GIF, or screenshots: which one does your job need?

Four formats for showing how software works: what each one costs to make, where it works best, how it fails, and whether readers can skip to a specific step.

Most of the time this question gets answered by whichever tool someone already has a licence for. Which is understandable, and usually wrong, because the four formats fail in completely different ways.

Here is the question that decides it, and it is not about production quality:

Can the reader skip to step seven?

If they need to — because they are stuck on one specific thing, or because they have done this before and only forgot one part — you need a format with an index. Screenshots and interactive demos have one. Video and GIFs do not.

Everything else follows from that.

Screenshots and text

What it is. Numbered steps, an image each, a caption underneath.

Best for. Reference documentation. Help-centre articles. SOPs. Anything someone arrives at with a specific question already formed.

Why it wins. It is scannable, searchable, indexable by Google, translatable, printable, readable on a phone with no sound, and accessible by default if you write the alt text. A reader can find step 7, read it, and leave in fifteen seconds. No other format on this page can do that.

How it fails. It is the most tedious format to keep current — twelve screenshots means twelve things that can go stale — and it is bad at conveying motion. Anything involving a drag, a transition, or “watch what happens to the sidebar” is poorly served by stills.

Cost to make. Low, with a real time floor. The method is here, and it takes about twenty minutes for a ten-step guide done properly.

Interactive demo

What it is. Captured screens of your real product that the viewer clicks through themselves, one step at a time, with the next thing to click highlighted.

Best for. Product tours on a landing page. Sales walkthroughs sent before a call. Onboarding. Anywhere you want someone to experience the product rather than read about it.

Why it wins. The viewer is doing something, so they stay engaged past the point where they would have stopped watching a video. They set the pace. It embeds on a page and works without sound, which matters more than people expect — a lot of demo viewing happens in an open-plan office. And it has an index: most players let you jump to a step.

How it fails. It is the wrong format for reference. Nobody wants to click through six screens to re-read the one instruction they forgot. It also carries a maintenance expectation: because it looks like your live product, a demo showing last quarter’s UI is more jarring than a stale screenshot.

Cost to make. Low to record, and it is the format most helped by tooling — the tools run $38 to $50 per seat per month, which is the real cost, not the recording time.

Video

What it is. A screen recording, usually with narration.

Best for. Explaining why, not just how. Announcements. Anything where tone, enthusiasm, or a human voice does work that text cannot. Complex motion.

Why it wins. Highest bandwidth per second of any format here. You can say “and this is the part everyone gets wrong” while pointing at it, which is difficult to convey in stills.

How it fails. No index. A viewer who needs step 7 must scrub a timeline hunting for it, and scrubbing is a miserable experience on mobile. It is expensive to update — changing one step means re-recording narration, or living with an obvious edit. It is not searchable unless you also publish a transcript. And it requires sound to be useful, which rules out a meaningful share of your audience at the moment they are looking.

Cost to make. Highest. The recording is quick; the retakes are not. Everyone underestimates how many times they will restart because of one stumbled sentence.

GIF

What it is. A short silent loop, usually a handful of seconds.

Best for. One interaction. “Drag the handle.” “The menu appears here.” Embedded inline in a document or a support reply, right where the sentence needs it.

Why it wins. Autoplays, no controls to click, no sound to enable, works everywhere HTML does. For a single motion it is unbeatable — it answers the question before the reader has finished asking it.

How it fails. Terrible at anything longer than a few seconds. The loop restarts and the viewer loses their place, there is no pause, no scrub, and no way to slow down. File sizes get large fast, and a large autoplaying GIF is the fastest way to make a documentation page feel slow. It also has no text, so it is invisible to search and to screen readers unless you describe it.

Cost to make. Low, if you keep it to one interaction. The discipline is in keeping it short.

Picking, in one table

Skip to step 7? Works without sound Easy to update Good for motion
Screenshots Yes Yes Tedious but reliable No
Interactive demo Yes Yes Yes — edit one step Partly
Video No No No Yes
GIF No Yes Yes Yes, briefly

The two-format combination that covers most needs: screenshots or an interactive demo as the primary artefact, with a GIF inline for the one interaction that stills cannot explain. Video is worth its cost when the message is not purely procedural.

The format you might not have considered: both

The formats are not mutually exclusive, and this is where the maintenance argument bites.

If you record a walkthrough once and your tool can export it as video, you get two artefacts from one recording session — the clickable version for your landing page, the MP4 for the places that want video. More importantly, when the UI changes you fix one step in one place and re-export, rather than maintaining two independent assets that gradually disagree with each other.

That is a practical argument for a tool over a screenshot folder, and it is worth checking on a pricing page: video export is a paid feature in most of this category. MP4 and GIF export require Growth at Arcade; MP4 export requires Pro at Guidde.

Demonstratio is designed to render MP4 at 720p or 1080p in your browser, with per-slide timing you set. It is still in development, so this describes the planned product rather than something you can use today. See our planned approach to pricing.

Demonstratio is in development

We're building a focused way to record a workflow in your browser and share it as an interactive demo. Join the waitlist for launch updates.

Keep reading