Street Interview
- Format
- Video
- Category
- Authority & IRL Faces
- Stage
- TOF
- Requires
- Real footage
What is it
An ad filmed in public with a handheld mic: someone approaches strangers, asks a question, and the answers carry the pitch. It works because nothing about it looks arranged, and because the reaction you capture is one the viewer believes you didn't script.
What makes it work
Real people, real location.
Actors reading lines on a fake street is instantly obvious, and so is AI. The whole value of this format is that it can't be faked convincingly, which is also why most attempts at it fail.
Ask in the place where the problem lives.
Outside the bar at midnight for a hangover product, at the gym, at the festival, in the queue. Context does half the work: the viewer sees the need and the moment together.
Let the awkwardness stay in.
The shaky camera, the ambient noise, the pause before someone answers. Cleaning it up removes the exact thing that makes it believable. Handheld mic, street sound, no music bed.
Close the loop on camera.
The strongest version films the result: hand it over, then come back the next morning, or the next week, and show the follow-up. A promise is a claim; a filmed follow-up is proof.
Shoot far more than you need.
Most approaches will be unusable. You're fishing for the two or three genuine reactions that make the ad, and you can't write them in advance.
Reference 1
AI Edit Master Prompt
Produce a {DURATION} vertical 9:16 UGC-style "street interview" ad for {PRODUCT}.
Generate the following sequence:
The Hook (The Problem): A {HOST_PERSONA} with a microphone interviews a {GUEST_PERSONA} in a {NIGHTLIFE_SETTING}. The host asks a {RELATABLE_QUESTION} (e.g., "What do you do for hngvers?"). The guest answers with a {COMMON_SOLUTION} (e.g., Alka-Seltzer), and a pop-up graphic of that solution appears. The guest admits they aren't sure it works and it "might be a placebo," accompanied by a confused animation graphic.
The Solution (The Pitch): The host responds with a {HOOK_STATEMENT} (e.g., "I got something that's not placebo"). The host introduces the {PRODUCT_NAME} and its {CORE_CLAIM} (e.g., "the world's first anti-hngver gum"). A large, colorful graphic of the {PRODUCT_PACKAGING} pops up.
The Instructions & Ingredients: The host explains how to use the product ("chew it while you drink"), with an on-screen graphic showing the {DIRECTIONS}. The host highlights the {KEY_BENEFIT_1} (fresh breath) and {KEY_BENEFIT_2} (stops hngvers). They list the {ACTIVE_INGREDIENTS} (e.g., DHM and B Vitamins), with corresponding graphics appearing for each ingredient.
The Review & Endorsement: The guest tries the {PRODUCT} on camera. They provide positive feedback on the {SENSORY_ATTRIBUTE_1} (flavor) and {SENSORY_ATTRIBUTE_2} (texture). The guest makes a relatable comparison (e.g., it's a good alternative to smoking a cigarette while drinking) and concludes it's "good to have."
Assemble: A fast-paced, authentic-feeling "man on the street" interview edit. The interaction should feel spontaneous and unscripted. Use quick cuts to keep the dialogue snappy, and rely heavily on pop-up graphics to visually reinforce the product's name, ingredients, and instructions, as the audio may be competing with ambient background noise.
Text Overlay: Use a bold, easily readable {SUBTITLE_STYLE} for the captions, syncing them word-for-word with the dialogue. Use asterisks to self-censor potentially sensitive words (like "hngvers") to avoid algorithmic penalties.
Audio: The raw, slightly noisy {INTERVIEW_AUDIO} captured on location to maintain authenticity. No background music is needed, as the ambient noise of the nightlife setting provides the atmosphere.
End Card: No dedicated end card; the video concludes with the guest's positive review of the product.Reference 2
AI Edit Master Prompt
Produce a {DURATION} vertical 9:16 UGC-style "street interview" ad for {PRODUCT}.
Generate the following sequence:
The Hook (The Question): A {HOST_PERSONA} with a microphone interviews a {GUEST_PERSONA} in a bustling {PUBLIC_SETTING} (e.g., a city street or park). The host asks a direct, {PROVOCATIVE_QUESTION} related to men's health (e.g., "When was the last time you woke up with morning wood?"). The guest provides a candid, {RELATABLE_RESPONSE}.
The Product Reveal: The host introduces the {PRODUCT_NAME} and clearly defines its {PRODUCT_CATEGORY} (e.g., "natural testosterone support"). A clean, colorful pop-up graphic of the {PRODUCT_PACKAGING} appears on screen to establish immediate visual recognition.
The Science (Ingredients & Mechanism): The host rapidly lists the {ACTIVE_INGREDIENTS} (e.g., Tongkat Ali, Shilajit, Boron). They then briefly explain the {BIOLOGICAL_MECHANISM}—how these ingredients interact with the body (e.g., "releasing bound testosterone") to deliver the {DESIRED_OUTCOME}.
The Credibility & Close: The host builds trust by stating a {SAFETY_CLAIM} (e.g., "third-party lab tested"). The host then asks for the guest's verdict based on the pitch. The guest provides a {POSITIVE_ENDORSEMENT} (e.g., "I'd be down to give it a try").
Assemble: A fast-paced, highly authentic "man on the street" interview edit. The interaction should feel entirely unscripted and spontaneous. Use quick cuts between the host and the guest to keep the dialogue snappy and engaging. Use the product graphic to anchor the viewer's attention during the pitch.
Text Overlay: Bold, dynamic {SUBTITLE_STYLE} captions synced word-for-word with the dialogue. Use contrasting colors (like orange or red) to highlight the {PROVOCATIVE_QUESTION}, the {PRODUCT_NAME}, and the {ACTIVE_INGREDIENTS} to make the key information pop.
Audio: The raw, slightly noisy {INTERVIEW_AUDIO} captured on location to maintain the authentic "street" feel. No background music is needed.
End Card: No dedicated end card; the video concludes naturally with the guest's final positive reaction. A clear {WEBSITE_URL} overlay appears prominently at the bottom of the screen during this final shot.