Walmart Connect: videos launch muted, so lead with music, $0.125 a bed

Walmart Connect's creative guide says its videos launch on mute, so it prefers upbeat music rather than voiceover (Walmart Connect, read 2026-10-05). For a holiday spot that means the first thing a shopper sees is text and picture, and the bed only matters after they unmute. On Sume the bed is one POST /v1/music-router/generate job at $0.125, with the length written into the prompt.

What the page recommends

The same page recommends 6 to 15 seconds for food and beverage, and it asks for simple, easy-to-read text overlays that tout product benefits. It gives category notes too, such as showing food and beverage both in and out of its packaging, and storytelling through visuals rather than verbal messaging for fashion.

Writing the bed

curl -X POST https://api.sume.com/v1/music-router/generate \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: walmart-holiday-bed-001" \
  -d '{
    "model": "sume/music-auto",
    "prompt": "Upbeat holiday pop, 118 BPM, G major. Sleigh bells, bright plucked guitar, handclaps. A 15-second track with a clean ending. Instrumental, no vocals."
  }'

Why the bed is not the message

If the video launches muted, the bed cannot carry the offer. Treat it as the second layer: the first layer is the picture and the burned text, and the music is what the shopper hears if they choose to unmute. That ordering is why Walmart prefers music to a voice track, since a voiceover that nobody hears is wasted audio and a caption that nobody can read is a wasted frame.

Keep the bed simple. A bright, steady rhythm with no vocal lead leaves room for a second language of text overlays, which matters if the same spot runs under several product names. Write one prompt per mood rather than one per product, and reuse the bed across the set.

Prompt tips from the Sume docs apply: name a tempo as a number, a key, two to four instruments with a texture, and one named moment in the arc. A 15-second track gives you room for exactly one such moment, so pick the beat where the price appears.

Cost and caveats

One bed is $0.125; four candidate beds for the same spot are $0.50. Pick one, mix it under the cut in Timeline ($0.10 per started output minute), and burn the offer as captions at $0.20. That is $0.125 + $0.10 + $0.20 = $0.425 of post-production on a single spot.

Walmart's page does not list file specs or loudness limits, so check the Ad Center for those before upload.