Why rotate at all
A digital screen is not a painted billboard. It changes its picture every few seconds, so pinning one frame to it for a whole flight throws away the one thing that makes it digital. Rotation is how you use that: you loop several versions of your ad on the same screens and decide which version runs when. There are three reasons to do it, and most good campaigns use all three.
Fight fatigue. The first time someone passes your ad it lands. The tenth time, the exact same frame starts to fade into the street. Rotating two or three versions of the same idea keeps it fresh for the people who pass a screen twice a day, so the campaign keeps working on the audience it repeats to instead of wearing out.
Match the moment. The right message in the morning is rarely the right message at night, and a sunny-day offer reads oddly in the rain. Rotation lets one plan say the appropriate thing for the hour, the day and the weather on each screen, rather than one compromise line that fits none of them. If you are new to the format, the what is DOOH guide covers the basics.
Test what works. Because every play is logged, running two versions against each other turns a screen into a live test. You see which line pulls, then move weight toward it. That is the part traditional out-of-home could never do, and it is why rotation is as much a measurement habit as a creative one. See the creative topic hub for the wider picture, and the billboard design guide for making each variant readable.
Rotation by daypart
A daypart is a block of hours with its own crowd, and the simplest rotation swaps the creative between them. Morning commuters, lunch shoppers and the evening crowd are different people in a different mood, so give each block the line that fits it. On Blindspot you set a creative to a window down to the hour, so a "grab a coffee" cut can own the morning while a "wind down" cut takes over the evening on the very same screen.
| Daypart | Who is out | Message that fits | Rotation move |
|---|---|---|---|
| Morning, 7-9am | Commuters, school runs | Quick, functional, a start-the-day offer | Run the morning cut |
| Midday, 12-2pm | Lunch crowds, errands | Nearby, now, an eat or shop prompt | Swap to the lunch cut |
| Evening, 5-7pm | The commute home, after-work | Unhurried, a wind-down or plan-tonight line | Run the evening cut |
| Late evening | Nightlife, diners, leisure | Nearby venues, tonight-only offers | Nightlife cut, in those zones |
You do not have to fill every block. Buying by the hour means you can run only the dayparts your audience is out for and drop the rest, so the budget lands on the windows that matter. The hourly scheduling guide goes deeper on that control, and the rush-hour playbook shows the two-peak version for commuters.
Rotation by context
Time is the easy dimension. The one that makes DOOH feel alive is context: rules that read the world and change the creative to match. Each version you upload can carry a contextual rule, so it plays only when a live condition is true on that specific screen. A rotation built this way is not a fixed loop; it reacts.
The supported triggers are broad. Weather and temperature let a wet-weather cut run only when it is raining, or an iced-drink version only above a set temperature. Air quality, stock and crypto prices and live sports scores can each drive a swap, and a custom live-data API lets you feed in your own signal, stock levels, a countdown, a live count. Set the rule once per creative and the screen does the rest. The weather-triggered advertising guide walks through the most common one end to end.
Day of the week is the quiet workhorse here. A weekend cut, a match-day cut, a payday cut: each is just a rule on a variant, so a single plan can carry the whole week without you touching it once it is live. This is where rotation stops being a chore and starts being a plan that runs itself, and it is a big part of what people mean by programmatic DOOH.
Rules read the world; the screen changes to match.
Rotation by context, in one line
Rotation for testing
Rotation is also the cleanest way to learn what works on the street, because every play is logged with a time and place. There are two moves. The first is a straight A/B test: run two versions of the same ad across the same screens and the same hours, then read the verified plays and any foot-traffic lift, and keep the winner. Because you pay per play, the losing variant only cost you the plays it ran during the test, not a separate buy.
The second is sequential storytelling: instead of testing two rivals, you order three or four variants so the screens play them in turn and build a single message across the flight. A teaser, then the reveal, then the call to act. The same audience that passes a corridor twice a day sees the story advance rather than the same frame repeat, which is exactly what deepens recall.
$0
extra to add a variant, you pay only per play
0M+
screens to test across, in 50+ countries
$0
the self-serve floor, a real test campaign
0¢
roughly the median cost of one play
Keep tests honest by changing one thing at a time: same screens, same hours, one difference in the creative, so the plays tell you about the creative and not the placement. Then feed the winner back into the daypart and context rotation above. Measurement is the whole point; the measurement topic hub covers verified plays and lift, and the 2026 statistics put the numbers in context.
Set a rotation, in 4 steps
Setting a rotation takes minutes. Upload your variants, give each one a rule, order them if you are telling a story, and publish. Blinky, the free AI planner, will draft the whole thing from a one-line brief if you would rather start from a suggestion.
Add each creative you want in the loop: the morning message and the evening message, the sunny version and the wet-weather version, or the two lines you want to test against each other.
Set the window or trigger for each version: a daypart such as morning versus evening, a day of the week, or a live condition like weather or temperature. A creative with no rule simply runs in the normal rotation.
If you are telling a sequence, set the order so the screens play variant one, then two, then three, building a single message across the campaign instead of showing the same frame every time.
Publish, then watch verified plays per variant, the appearances that actually ran, so you can see which creative is working and shift weight toward it.
That is the whole setup. If you want a starting draft, describe the audience, the city and the variants you have to Blinky and refine what it gives you, or follow your first campaign from the top.
The Blindspot way
Everything above rests on one pricing choice: you pay per play, the cost of one ad appearance on one screen, not for a block of time or a modelled thousand impressions. That is what makes rotation efficient rather than expensive. Swapping which version fills a play does not add a fee, and a variant you add sits idle at no cost until its rule fires. You are buying a set of plays; rotation just decides what shows in them.
It matters that this is per play and not per thousand. A cost-per-thousand buy prices a modelled audience average you cannot audit, so a second creative is another line item negotiated into the deal. Per play, the second creative is just a rule and an upload, and every appearance is logged so you can see it ran. That is the difference between rotation as a feature and rotation as an afterthought.
This is why the approach works at any budget, not only a small one. The same per-play discipline that lets a small campaign from $40 carry a morning and an evening cut also lets a national flight run a weather rule across dozens of markets on one plan and one invoice. Nothing about rotation is reserved for big spenders; it is the default behaviour of a self-serve platform priced per play.
When you are ready, open the map to pick screens, read the per-play price on each, and set your rules, or see exactly how a plan comes together in book a billboard. For prices by city and format, the price index and billboard costs pages have the figures.