On screenIf you advertise on venue screens, you will eventually face a choice between two messages. One leads with a discount. The other leads with convenience. Guessing is cheap, but it teaches you nothing. A simple test design can tell you which direction to lean, as long as you keep the test clean and the conclusions modest.
This guide covers how to test ad creative out-of-home, including a/b testing digital out-of-home (DOOH) creative across venue categories such as restaurants, fitness facilities, dealerships and salons. It does not cover how CETV Now rotates ads on a screen, because that has not been confirmed. Ask the team about rotation before you design a test that depends on it.
What can you learn from a creative test?
A test answers a narrow question: given this audience and this setting, did version A or version B draw more of the response I can measure? It does not tell you why people responded, and it does not predict what a larger campaign will do.
Good questions for a first test are specific and tied to a decision:
- Does a percentage-off offer or a free-add-on offer draw more redemptions?
- Does a headline naming the neighborhood work better than a generic one?
- Does a call to action to scan a QR code get more response than one to search your name?
- Does the same offer behave differently in a restaurant than in a fitness facility?
The last question is the one that makes venue categories useful. Venue categories give people different moments and different mindsets. CETV Now runs in auto dealerships, restaurants and bars, medical and dental offices, fitness and wellness facilities, retail, and professional services such as offices and salons. To pick which two or three to include, see how to choose venue categories for in-venue advertising.
Why change only one variable at a time?
If version B has a new headline, a new offer, a new ending and a different call to action, and it performs better, you cannot say which change mattered. You have a winner but no lesson. Changing one variable keeps the lesson attached to the result.
Pick the variable that matches your biggest doubt
| Variable | Example of A vs. B | What you learn |
|---|---|---|
| Offer | A percentage discount vs. a free add-on | Which value framing people act on |
| Headline | Neighborhood name vs. generic benefit | Whether local relevance matters |
| Call to action | Scan the code vs. search the business name | Which response path people use |
| Format | 30-second video vs. static design | Whether motion earns its production cost |
| Opening seconds | Product first vs. person first | Which hook holds attention |
Everything else stays identical: the logo, the length, the voice, the offer terms, the schedule. For help choosing between formats, read static vs motion graphics vs video ads and review the available ad formats. For craft questions, such as how to design for a screen people watch from across a room, see how to design video ads for venue screens.
How do you compare like with like?
Most bad tests fail here. If version A runs in a restaurant at dinner and version B runs in a dental office in the morning, the audience is doing the deciding, not the creative.
CETV Now targeting can be layered by neighborhood, city, metro, zip code, sector or vertical, and individual venue, along with audience type and time of day (daypart). That gives you the controls to hold conditions steady. Read how dayparting works if the term is new.
Two kinds of comparison
- Within a category. Run version A and version B in comparable venues of the same category, such as two groups of restaurants in the same market, over the same dates and dayparts. This isolates the creative.
- Across categories. Run the same version in different categories, such as restaurants and fitness facilities, over the same dates. This isolates the setting. Do not change the creative at the same time.
Do not mix the two in one test. If you want to know both whether version A beats B and whether restaurants beat gyms, run them as two separate tests, one after the other.
Keep the conditions matched
- Same start and end dates, so weather, paydays and holidays affect both versions.
- Same daypart, since a lunch crowd and an evening crowd are different audiences.
- Same market, or at least markets you have reason to treat as similar. Live markets are Phoenix, Dallas, Atlanta, Nashville and Salt Lake City.
- Same schedule intensity, so one version is not simply shown more often than the other. Confirm with the team how your schedule is set up.
If you want to test somewhere else, availability is unconfirmed. Miami is coming soon but is not live. Send an inquiry to ask what is possible.
How do you make each version trackable?
Screens in venues do not produce clicks, so you have to build the response path into the creative. The goal is that each version leaves a different fingerprint. For more on this, see how to measure out-of-home advertising without clicks.
- A promo code per version. Version A says "mention SPRING" and version B says "mention BONUS." Count redemptions of each at the register or booking form.
- A QR code per version. Use two different codes pointing to two different landing pages, or to the same page with different tracking parameters.
- Branded search. Watch searches for your business name during each period. This is a softer signal, since you cannot tie a search to a specific version.
- A simple ask. Train staff to ask how people heard about you, and log the answers by date.
Promo codes and QR codes are the cleanest, because they tie a response to a version. Branded search and staff asks are useful backups, but they are noisier.
If you are in a regulated field such as healthcare, financial services, alcohol or legal services, check your own compliance rules before you put offers or codes on screen. Creative is also submitted to CETV Now for review before it runs.
How do you use a control period?
A control period is a stretch of time with no test running, used to establish your baseline. Without it, you cannot tell whether a spike came from the creative or from something that was going to happen anyway.
Record your normal level of redemptions, bookings, walk-ins or branded searches for a period before the test starts. Then compare the test period against that baseline. The IAB's DOOH measurement guide describes more rigorous attribution approaches, such as synthetic control groups and matched market testing, which are designed to isolate campaign impact. A small advertiser usually will not run those, but the idea is the same: compare against something that did not see the ad. Read more at IAB's overview of the measurement guide.
How should you read the results?
Be honest about what a small test can support. Ad insertion on CETV Now has no minimums and no long-term commitment, so you can start small. The trade-off is that small tests produce small numbers, and small numbers swing a lot.
- If version A gets a couple more redemptions than version B, that is not a winner. Random variation can easily produce that gap.
- A large, consistent gap that holds across venues and weeks is more meaningful than a narrow gap in one location.
- A result tells you about that audience, in that category, in that period. A holiday week may not resemble an ordinary one.
- Redemptions can reflect an offer's appeal more than the quality of the video. That is useful to know, but it is a different lesson.
- Network figures CETV Now reports, such as a 4.2x average campaign ROI, are network-wide reported numbers. They are not a forecast for your test, and individual results vary.
Treat the first test as a way to form a hypothesis. If one version leads, run it again against a new challenger. A result that repeats is worth acting on. A result that appears once is worth a second look.
A simple test plan you can copy
- 01
Write the question
One sentence, for example: does a free add-on draw more redemptions than a percentage discount at restaurants?
- 02
Choose one variable
Offer only. Keep headline, length, visuals and call to action the same in both versions.
- 03
Set up tracking
Create one promo code or QR code for each version. Brief staff on how to log each redemption.
- 04
Record a baseline
Note your normal response level for a period before launch. This is your control.
- 05
Match the conditions
Pick comparable venues in one category, one market, the same daypart and the same dates for both versions.
- 06
Run and leave it alone
Do not edit creative or schedules mid-test. Changing things partway makes the results unreadable.
- 07
Compare, then repeat
Compare each version's responses against the baseline and each other. Note the limits. Then test the winner against a new challenger, or against the same offer in a different venue category.
For a wider planning view, use the DOOH campaign planning checklist and how to set clear goals for a DOOH campaign. If you want a low-commitment starting point, see how to run a small business DOOH test campaign.
When you are ready to scope a test, share your markets, venue categories, creative and budget. CETV Now maps the plan to real-world audiences and provides a quote, since pricing is quote-based and depends on market, venue category and schedule. Start from the advertisers page or the how it works overview.
Figures reflect CETV Now network data and are provided for informational purposes. Individual campaign results vary; no specific outcome is guaranteed.
