Skip to the report summary
Creative intelligence, documented builds

Three ads. Every brief and every result, unedited.

Three AI-generated ads across two campaign types, run on live client spend. Every brief, every creative decision and every data point documented, including the ones that only hold for a specific funnel stage.

Get the full report See what is inside

Three live campaigns · March 2026 · free to read · one short form

3
ads built
2
campaign types
1
client, anonymised
6
sections
Explore QuickAdsAd researchServicesReportsGuidesPricingSign in
The short version

What this report settles

QAFrom the QuickAds ad research library

Three AI-generated ads were run on live client budget and documented end to end. The pattern across all three is that format choice, not design quality, produced the result, and the brief was the document that decided it. Each one is shown with the brief, the reasoning, the outcome and the conditions under which it would not repeat.

Everything on this page is drawn from the report itself. The exhibits are real and the figures are the ones inside it.

The questions below are answered in full. The rest of the study opens as soon as you tell us who you are.

3Ads documented end to end
2Campaign objectives covered
Rs 1.3LLive client spend across the three
2Static builds, one video build
FestiveSeason the campaigns ran in
AnonymisedClient identity, all data unedited

All figures come from three campaigns for a single anonymised jewellery retail client on roughly Rs 1.3 lakh of live spend during a festive window, March 2026. They are documented case history for that account and those conditions, not a benchmark and not a forecast. Every number on this page is an observed benchmark from the ad set described above. None of it is a projection, and none of it is a performance QuickAds is promising your account.

Answered here

Six questions this report settles with a number

Each answer is quoted from the study and names the ad set it was observed in.

  • Do AI-generated ads actually perform?

    Question 01 of six

    In the three builds documented here they did, and the report is explicit that what separated them was brief clarity rather than generation quality: which funnel stage, which audience belief, which single action. From the QuickAds Creative Intelligence Report.

  • Should you run static or video ads?

    Question 02 of six

    It depends on funnel stage, and the report shows both sides. For a warm, intent-driven audience a static removed friction and outperformed the video already running on that account. On cold traffic the same choice would be wrong, because video has to build context first. From the QuickAds Creative Intelligence Report.

  • What is hook rate and why does it matter?

    Question 03 of six

    It measures how many people the opening frame holds. One documented lead-generation static reached 17.1% hook rate on a single client campaign, and the report treats it as the number most dashboards are missing. From the QuickAds Creative Intelligence Report.

  • What should a creative brief for a paid ad contain?

    Question 04 of six

    Three things, stated before anything is made: the funnel stage, the audience belief being addressed, and the single action being asked for. All three documented builds trace back to those. From the QuickAds Creative Intelligence Report.

  • Where in an ad is performance decided?

    Question 05 of six

    In the first three seconds. Everything after the opening frame only has to avoid breaking what the opening earned, which is why the report spends its length on briefs rather than on production. From the QuickAds Creative Intelligence Report.

  • Can these results be expected on another account?

    Question 06 of six

    No, and the report says so. All three builds come from one anonymised client on roughly 1.3 lakh rupees of live spend in a festive window, and each result is stated with the conditions it depends on. It is documented case history, not a benchmark. From the QuickAds Creative Intelligence Report.

Sneak peek

One exhibit, partly open

Part of one of three documented builds. Each carries the brief, the decisions, the reasoning and the caveat.

One build, start to finish

Live rows
StageWhat the report shows
Brief and problemThe objective, the audience and what the account was already running
Creative decisionsOffer-first layout, one call to action, no copy beyond the action
Why it was briefed that wayWarm audiences are deciding whether to tap, not whether to buy
LockedThe results, the caveats and the two remaining builds

The brief is the most important creative document.

What to do about it

Five moves, in order

Drawn from the report. The reasoning behind each one sits in the full study.

  1. Write the funnel stage, the audience belief and the single action before anything is made.
  2. Choose format by funnel stage rather than by production preference.
  3. Add hook rate to the dashboard. It explains results that CPA alone does not.
  4. Spend your review time on the opening three seconds of every asset.
  5. State the conditions attached to any win before you generalise it across accounts.
Who this is for

Built for the people briefing the work

  • CMOs deciding how creative gets briefed
  • Performance marketers reading hook rate
  • Creative directors writing briefs
  • Teams evaluating AI-generated creative
How it was built

Read ad by ad, not modelled

  • Three campaigns for one anonymised client, run on live budget during a festive window.
  • Every brief, creative decision and data point reproduced unedited.
  • Each result is stated with the funnel stage and audience condition it depends on.
  • Documented case history for one account. Not a benchmark set and not a projection.
Get the report

Open the full study

Free to read, every section and every exhibit. Fill this in and the report opens.

Your details are used to fulfil this report request. Read our Privacy Policy.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Report unlocked

Your report is ready. Opening it now. If it does not open, use the button below.

Read it now