A 1970s split-screen grammar rebuilt for software: hard ink gutters divide the frame into simultaneous views. The hook drowns in six dashboards at once, the drop slams one gutter aside for Acme Pulse, the proof runs three synchronised cameras (Pulse, the chart, the on-call human) on the same timecode, and the gutters close for the lockup.
Panels pop in one by one, left to right, top to bottom, one per eighth note, until six dashboards fill the frame; the headline panel slides up from the bottom and “today?” lands on beat 1 of bar 4.
All six panels collapse into the left strip and the centre gutter slams left (50% → 33%) exactly on the drop; the teal panel wins the frame and the mark scales up from 0.8 inside it.
The three panels open in sequence (A top-left, B top-right, C bottom) with matching amber timecodes: Pulse’s coral row flags, the chart dips at the same 09:42:07, and the human’s cursor hits Roll back on the next downbeat.
“You ship.” lands on the left panel, “Pulse watches.” on the right a beat later; then the gutters slide shut toward the centre and the lockup band rises with Start free on beat 3.
Split screen shows cause and effect at the same instant, which is exactly what a monitoring product promises: the metric, the alert and the human decision are all visible together. It’s also cheap to build from real UI captures. Risk: on a phone at 1:1 more than three panels becomes noise, and a split that doesn’t say where to look loses the viewer.
Adapt to your product: Put the before-state (your users’ current juggling) in many small panels and your product in one big one. For the proof, split the frame into the three roles your product connects, for example the signal, the system and the person, and keep the same timecode in each.
References: https://en.wikipedia.org/wiki/Split_screen_(film) (split screen history and definition) · https://www.premiumbeat.com/blog/split-screen-editing-and-composing/ (composing split screens to guide attention)