Imagine your hardware brand is about to brief a supplier in China on a new monitor arm. The ecommerce team wants a polished setup film. Support keeps getting a narrower question: will the clamp fit a desktop with a steel rail underneath? Put that question on the shoot list before buying another lifestyle scene.
A limited placement test, a broader creative decision
On November 3, 2026, Reddit announced a search results ad test with a select group of eligible advertisers. It uses existing targeting, and query relevance can be one factor in whether an ad appears. Reddit did not announce universal access or publish sales results for this test. The production decision can be made without access: is there a specific buying question your current assets leave unanswered?
Build one answer a buyer can verify
The ONCE query-to-demo brief has three fields: the buyer's question, the shot that answers it, and the place where the answer can be checked. Start with sales and support logs, then use public discussions to pick up wording or uncover another objection. A popular comment is a lead for research; check it against what your customers actually ask.
For the monitor arm, ask the factory to send the production clamp, not the prototype. Film it sliding onto a sample desk from underneath, with a ruler in the same plane as the clearance being measured. Capture a second setup where the rail blocks the clamp. Put the approved clearance and the affected model numbers beside the installation photo on the product page. The failed fit is useful: it lets the customer rule out an incompatible desk before ordering.
Make the first test small enough to change your mind
Run the question-led creative against your usual benefit-led version wherever you can compare them fairly. Hold the product, offer and destination constant; use an actual split test if your channel supports one. Pick a decision metric in advance, such as add-to-cart rate among product-page visitors, then check orders and fit-related returns when enough data has arrived. A higher click rate alone can just mean the question was interesting.
If the page already answers the question but the ad makes a broader promise, tighten the creative. If the two versions reveal a real product constraint, narrow the fit claim. Commission more demos only after the first one shows a useful reason to expand, such as demand for a second verified fit configuration. Until your account gets access, use an existing channel as the test bed. For the playback check, see ONCE’s guide to keeping the key product proof visible.
