The Blink Test: Putting Photosensitivity Review on the Delivery Critical Path

There’s a particular kind of silence in a studio when someone asks, “Has this been flash-tested?” It’s not quite panic. It’s the sound of a delivery schedule quietly growing a new branch. For motion and VFX teams, photosensitivity review often arrives late, usually after the hero cut is locked and the client has already fallen in love with a three-frame inverted flash that happens to sit right on the beat.

That’s the wrong order. In 2026, with campaigns shipping to broadcast, BVOD, social, retail screens and stadium boards from the same master, photosensitivity testing should be a creative checkpoint, not a last-minute fire drill. It doesn’t have to sand the edge off the work. Done properly, it makes the work more deliberate — and much easier to defend.

Why UK delivery teams can’t treat this as a footnote

UK broadcast has a long-standing expectation that flashing content is assessed before it goes out. Harding-style flash testing is the familiar shorthand: a frame-by-frame analysis of luminance and red flashes, with thresholds that vary by screen area, frequency and contrast. The detail matters less than the habit. Broadcasters and clearance teams want evidence that someone looked, measured and made a judgement.

Online is messier. Social platforms have their own accessibility guidance and automated systems, but they are not a substitute for your own review. A cutdown that looks fine on a calibrated monitor can behave differently when compressed, cropped, autoplayed and watched on a phone at arm’s length. Add in the usual UK mix of iPlayer, YouTube, Meta, TikTok and a retailer’s in-store screen, and one innocent strobe can create several different risk profiles.

And this isn’t an edge case. Photosensitive epilepsy affects roughly one in 100 people, and many more are sensitive to flicker, high-contrast patterns or rapid movement without having a diagnosis. For a brand, that’s not just a compliance issue. It’s a audience issue.

Where risky motion usually hides

Most problematic sequences aren’t horror-movie strobes. They’re ordinary motion-design moves doing their job a little too well. Kinetic typography is a regular offender: white letters snapping on and off against a dark background, words inverting on every kick drum, or a fast scroll of bold type with hard cuts between frames. Transition effects are another. Whip pans, glitch wipes, camera flashes, lightning reveals and rapid zoom-punch cuts can all stack luminance changes in a way that reads as flicker.

Then there are the quieter culprits: moving stripes, checkerboard patterns, high-contrast textures scrolling behind a product, or a red-on-black graphic that pulses with the music. Red flashes deserve special attention because the eye processes them differently, and automated tools often weight them separately. A sequence can pass a simple brightness check and still feel aggressive in the edit.

VFX work adds its own wrinkles. Muzzle flashes, explosions, energy pulses and sci-fi UI elements are obvious candidates, but so are particle bursts, shimmering shields and rapid camera shake. The question isn’t “Is this a flash effect?” It’s “Does this sequence create repeated changes in brightness or colour across a significant part of the screen at speed?”

A simple pre-delivery workflow

The goal is to catch risk while there’s still room to adjust. Start earlier than you think. Run a first pass on the animatic or previs, not just the final grade. At that stage, you can change timing without rebuilding a whole shot.

1. Flag, don’t guess

Ask the editor or motion lead to mark any sequence with rapid cuts, strobing, flashing, high-contrast pattern movement or red pulses. A simple marker in the timeline is enough. Don’t rely on memory. The person who animated the sequence is often the least able to see it objectively.

2. Test the right version

Run automated flash analysis on the actual delivery master — not a preview export with different compression or frame rate. Test each aspect ratio and each cutdown separately. A 16:9 hero film and a 9:16 social cut are different pieces of motion, even if they share the same assets. If the work is going to broadcast, use the broadcaster’s preferred tool or a recognised Harding-style analysis. If it’s social-first, still document what you checked and how.

3. Review at speed and at rest

Automated reports are useful, but they don’t replace eyes. Watch the sequence at normal speed, then step through frame by frame around flagged moments. Check on a phone as well as a monitor. Compression can turn a soft glow into a harder edge, and small screens change the proportion of the frame that’s flashing.

4. Adjust with intent

You rarely need to remove the moment. Reduce the luminance jump, spread the change over more frames, soften the edge of the flash, desaturate the red, or hold a frame instead of strobing. Sometimes a quick blur or a short ramp is enough. The aim is to keep the punch while removing the repetition that causes trouble. If the creative idea depends on a flash, make it a single strong beat rather than a machine-gun sequence.

5. Document the pass

Keep a short record: version number, timecodes, tool used, result, adjustments made and who signed it off. This doesn’t need to be a legal dossier. A one-page delivery note is fine. It saves arguments later and gives the client something concrete to point to when someone asks whether accessibility was considered.

Don’t let the test flatten the work

There’s a temptation to treat photosensitivity review as a licence to dull everything down. That’s a mistake. The most effective motion often lives near the edge, and audiences respond to contrast, rhythm and surprise. The point is to control the risk, not remove the craft.

A flash that lands once can be more powerful than a strobe that lands twenty times. A red accent that holds for a beat can hit harder than a pulsing red field. Testing gives you permission to make those choices deliberately, with the confidence that the delivery won’t be held up by a sequence nobody checked.

For UK studios, the practical answer is simple: put the blink test on the critical path. Flag early, test the real masters, adjust with intent and write it down. Then send the work out with the same energy it was made with — and one less thing to worry about at 5pm on delivery day.