Lead-In Lab
Test the opening line. Keep the rest of the ad.
Rewrite only the first 10 to 15 words of a proven ad, grouped by hook type, then test the new line against the original.
Lead-In Lab, Generate · Northwind Sleep demo workspace, fictional data.
01 Hook types
Eleven ways an opening line can work.
- 01 Testimonial
- 02 Transformation
- 03 Unexpected result
- 04 Conflict
- 05 Curiosity
- 06 Specific number
- 07 Contrarian statement
- 08 Founder story
- 09 Customer quote
- 10 Discovery
- 11 Problem
The 11 hook types on the Generate tab’s own picker · Northwind Sleep demo workspace, fictional data.
Testimonial and customer-quote alternatives may only use words a real customer actually said, and a specific-number alternative may only use a number you have explicitly approved. Every alternative carries a plain rationale and a flag for whether it uses only your supplied facts.
02 The triplet test
A 36% lift, well past the 15% bar to call it a winner.
A hook type has to beat the runner-up by at least 15% before Lead-In Lab calls it a winner. Over the demo workspace's Lead-In Test window (Aug 11 to Sep 9), Variant B's 1.92% CTR beat the control's 1.41% by 36%, clearing that bar with room to spare.
| Variant | Lead-in | CTR | Spend | Conversions |
|---|---|---|---|---|
| A, control | Three weights, one cooling layer, so you finally pick based on your body instead of guessing. | 1.41% | $903.53 | 24 |
| B, winner | I almost returned my first weighted blanket because of the heat, then someone told me to try this one instead. | 1.92% | $915.86 | 31 |
| C | What if the reason your weighted blanket feels wrong has nothing to do with the weight? | 1.48% | $892.97 | 25 |
Lead-In Lab, triplet test · Northwind Sleep demo workspace, fictional data.
03 See it
Alternatives, grouped by the hook type they test.
Generate finds or accepts a lead-in and writes alternatives across the hook types you pick. Performance reports lead-in and hook-type efficiency from the account's own synced ads, plus the status of any lead-in experiments already run.
A promising alternative becomes a guardrailed test in Creative testing →
Fig. 4 · Lead-In Lab, Generate · Northwind Sleep demo workspace, fictional data.
Test your own opening line.
Connect a Meta ad account and pick a proven ad to rewrite, or load the Northwind Sleep demo workspace and look around first.