# Could a Reel Replace Your Booking App? A Travel Ad Pre-Test

A video ad pre-test for a hotel booking app with 100 young Indian smartphone users: 62% top-tier intent, 39.6% doubting the payout, and a conditional go.

Published 2026-09-04 · Updated 2026-10-12
Canonical: https://hercules.works/blogs/hotel-booking-app-ad-testing-case-study-india

A hotel booking app's video ad was pre-tested with 100 young Indian smartphone users; the report was generated on Hercules on 12 August 2026. 62% rated their likelihood to download the app and book or upload stays at 9 or 10 out of 10, and 55.2% of those answering said the app seemed more trustworthy than Airbnb, Booking.com or Agoda. But 39.6% doubted the payout was real, so the report's verdict was a conditional go: explain payouts before spending on media. With 100 respondents, read it as a directional pre-test.

## The Decision the Test Was For

The question was practical: should this video ad get media spend as it is, or does something need fixing first? The film pitches an app where guests film their hotel stays, upload the videos and earn a 50% booking commission, while other travellers book with that video as evidence.

The report does not name the app, and neither does this page. The figures describe how up to 100 respondents reacted to one film in a survey. They do not measure the app's downloads, and the comparison with Airbnb, Booking.com and Agoda is respondents' opinion after one ad, not a measure of those platforms.

## The Verdict and the Numbers Behind It

The report's verdict was "conditional go for launch": the video ad was strong enough to proceed, but payout doubts needed fixing before full media spend.

| Measure | Result | Scale |
|---|---|---|
| Likely to download and book or upload stay reels | 62.0% (38.0% at 10) | 9 or 10 of 10 |
| Opening hook eye-catching | 75% | 8 or more of 10 |
| Clarity of how uploads earn the commission | 59.4% | 9 or 10 of 10 |
| Credibility of the 50% commission | 69.8% (32.3% at 10) | 8 or more of 10 |
| App seems "much more trustworthy" than Airbnb, Booking.com or Agoda | 55.2% (53 people) | Single choice |
| "About the same" as existing booking sites | 36.5% | Single choice |
| "Less reliable" than established sites | 8.3% | Single choice |

Bases vary. The trust question and the emotion question were answered by 96 people, although the report heads the trust section "based on 100 respondents". In two summary panels it also labels the 62% and 55.2% figures "n=150", which cannot be right for a 100-person study. We use the bases from the question tables.

## How the Test Was Run

The report describes forced single monadic exposure: each respondent saw this one film and no other. The video was time-locked on load so it could not be skipped. A 13-question battery covered attention, clarity, credibility, emotion and intent. Respondents were screened on age, hotel-stay frequency and social video use. Quality checks were straight-liner detection and a filter for gibberish in open answers.

The report records 100 completed surveys against a target of 100, "deployed across urban smartphone users across major Indian metros". How they were recruited is not stated, so treat the sample as non-probability.

| Sample | Share |
|---|---|
| Aged 18-24 | 53.0% |
| Aged 25-35 | 45.0% |
| Men / women | 62.0% / 38.0% |
| Delhi | 15.0% |
| Chennai, Kolkata, Hyderabad | 14.0% each |
| Bangalore | 13.0% |

A single-ad test has no benchmark built in. 62% means something only against another video ad tested the same way, or against norms you already hold for your category.

## Who Responded Most

The 25-to-35 group rated the video ad higher on every measure the report splits by age. Both groups are under the report's 100-person reading threshold, so treat these as directional:

| Measure | Aged 25-35 | Aged 18-24 |
|---|---|---|
| Intent 9 or 10 of 10 | 71.1% | 52.8% |
| Intent 10 of 10 | 44.4% | 30.2% |
| Clarity 9 or 10 of 10 | 73.8% | 46.1% |
| Credibility 10 of 10 | 42.9% | 21.2% |
| Hook 10 of 10 | 42.2% | 28.3% |

In the report's driver model (McFadden R² 0.4625, n=100), clarity of how uploads earn the commission carried 46.2% of explained variance in intent, ahead of how eye-catching the ad was (31.2%) and credibility (22.6%); read it as directional.

Asked which element did the best job of conveying the offer, 31.3% picked the explanation of the 50% commission, 26.0% the contrast with booking platforms, 20.8% the stay visuals and 20.8% the voiceover text (base 96).

Frequent travellers were the keenest viewers. All respondents with five or more hotel stays a year wanted to keep watching, and 82% of them say they regularly post travel reels. The report gives 77% for the share who see real-time guest video as more trustworthy than standard listings, once for all viewers and once for frequent bookers only. It does not give the size of the frequent-traveller group.

## The Payout Doubts Behind the Conditional Verdict

Asked which feelings the ad triggered (more than one answer allowed, base 96), 51.0% said excitement about earning money while travelling. But 39.6% said they were sceptical the payout was real and 31.3% were confused about how the app works. Women were more often confused (38.9% against 26.7% of men); men were a little more often sceptical (41.7% against 36.1%).

These answers sit beside a strong credibility score. The report's credibility section says the 50% offer does not trigger scepticism "as a 'too good to be true' offer", based on 69.8% rating it 8 or more. That is a different question from whether people believe they would actually be paid, which is what the 39.6% answered. Both can be true at once, and the second is the one the conditional verdict rests on.

The report recommends a short "How Payouts Work" explainer in post-click onboarding, given as 15 seconds in one place and as the first 3 seconds of onboarding in another. It also recommends leading with "Unfiltered Real-Time Video Proof" and recruiting frequent leisure travellers as creators. None of these was tested.

## What This Study Cannot Tell You

- **Downloads or bookings.** Intent after a forced viewing is not behaviour.
- **Feed performance.** Real reels can be skipped; a time-locked survey video measures reaction after full viewing, not whether people stop scrolling.
- **The platforms named.** No Airbnb, Booking.com or Agoda ad was shown, so the trust comparison is opinion after one challenger ad.
- **Older travellers.** 98% of respondents are 18 to 35.
- **Whether the fix works.** The payout explainer has not been tested.

## What to Ask in Your Own Pre-Test

1. **Test two cuts, each with its own sample:** the current video ad and one with the payout explanation. That turns the conditional verdict into a measurable comparison.
2. **Keep the questions identical** between waves so the numbers can be compared.
3. **Measure the specific doubt directly**, for example "How sure are you that you would actually be paid?", not only overall credibility.
4. **Report one base per question** and check summary figures against the tables.
5. **Decide whether you need a forced-view read or an in-feed one.** A survey gives you the first.

Our [ad testing guide](/blogs/ad-testing-platform-india/) covers monadic and sequential designs for video, and the [social media usage study](/blogs/social-media-usage-study-indian-professionals/) looks at when Indian professionals scroll.

## Running This Study on Hercules

Repeating this pre-test at 100 completes costs 100 × ₹150 = ₹15,000 on the Plus plan. Testing the fixed cut against the original as two separate cells of 200 each takes 400 completes, a Pro bundle at 400 × ₹120 = ₹48,000, and lifts each cell above the 100-person reading line.

The [panel methodology](/panel-methodology) page lists targeting by age and region. Screens for hotel-stay frequency and social video use, as in this study, would be questions inside the survey.

## Frequently asked questions

### How do you pre-test an app install ad?

Show your target users the ad under controlled conditions, then ask about attention, clarity, credibility and intent. This study used one film per respondent, a video that could not be skipped, and screens for age, hotel-stay frequency and social video use.

### What is a conditional go verdict?

A decision to proceed once a named problem is fixed. Here the video ad scored 62% top-tier intent, but 39.6% doubted the payout, so the report advised explaining payouts before full media spend.

### Did travellers trust the app more than Airbnb or Booking.com?

After seeing this one ad, 55.2% of 96 respondents said the app seemed much more trustworthy than Airbnb, Booking.com or Agoda, 36.5% about the same and 8.3% less reliable. No ad for those platforms was shown.

### How many respondents does a video ad pre-test need?

Enough for each group you will act on. The report reads groups under 100 as directional; this study's age groups were 53 and 45. Comparing two cuts at 200 each takes 400 completes.

### What would the same pre-test cost?

At list price, 100 completes on Plus cost ₹15,000. A two-cell test of 400 completes on a Pro bundle costs ₹48,000.
