Modular Top-of-Funnel Video Testing
Hook viewers early, test interchangeable content blocks, and retarget engagement
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 96%
Shackelford says his team usually prefers video at the top of the funnel because it serves two purposes. Viewing behaviour creates an audience that can later be remarketed to, while the video itself can be assembled from interchangeable content blocks. Marketers can vary the hook, demonstration, proof, or other sections to discover useful combinations. The hook or solution belongs in the first three seconds because he reports average watch times of roughly two to five seconds across the brands his team manages. The method remains empirical: even identical creative may perform differently in separate ad accounts, so the team tests the format rather than assuming video will win everywhere.
Origin
Extracted from The Foundr Podcast as Nick Shackelford described patterns his agency observed while testing top-of-funnel creative across many Facebook advertising accounts.
Core principles
- 01The opening seconds must carry the hook or solution
- 02Video creates both creative-learning signals and engagement audiences
- 03Modular content blocks make winning combinations easier to find
- 04Account-specific evidence outranks a universal creative rule
How to run it
- 1
Define the first-three-second promise
Choose the hook, solution, or striking transition that gives a cold viewer an immediate reason to continue.
Pro tip Make the opening understandable without waiting for later context.
Watch out Do not hide the useful idea behind a slow introduction.
- 2
Create content blocks
Produce interchangeable openings, demonstrations, benefit statements, and proof sections.
Pro tip Keep each block self-contained enough to recombine cleanly.
- 3
Assemble controlled variants
Combine the blocks into variants that reveal which opening or sequence contributes to performance.
Watch out Changing everything at once makes the learning difficult to interpret.
- 4
Launch to cold traffic
Run the videos at the top of the funnel and measure attention and acquisition outcomes in the specific account.
Pro tip Treat each account as its own evidence environment.
- 5
Retarget engaged viewers
Build follow-up audiences from meaningful video engagement and move them into the next funnel message.
- 6
Recycle the winners
Retain effective blocks, replace weak ones, and test new combinations rather than rebuilding every video from scratch.
Pro tip Keep the winning hook while testing a new demonstration or proof block.
Watch out Do not promote an agency-wide pattern into an account-level certainty without testing it.
In the wild
A brand films one clear product demonstration and pairs it with three different opening hooks. Each variant keeps the rest of the video stable. Results identify the strongest opening, and viewers who engage meaningfully enter a middle-funnel audience for benefit and objection messaging.
→ The brand learns which opening earns attention and gains a qualified remarketing pool from the same test.
Common mistakes
Saving the hook for later
The viewer may leave before the ad reveals its solution or reason to continue.
Producing indivisible videos
When no content blocks can be swapped, every test requires a complete rebuild and teaches less about individual elements.
Assuming identical account behaviour
Shackelford reports that the same creative can perform differently across separate accounts, so local testing remains necessary.
Is it for you?
Best for
Paid-social teams with enough creative capacity to test multiple short video combinations.
Not ideal for
Advertisers who cannot produce truthful visual demonstrations or gather enough traffic for comparison.
From the transcript
“it provides us a new audience that we can remarket towards off of video”
“content bricks that allow for us to like slide in and slide out to find a right combination”
“your hook should be or your solution should be is in the first three seconds”
From the episode
363: Everything I've Learned Spending $30M on YouTube Ads