sixtysteps.co

A 4-dimension ad matrix: 8%+ CTR on several cells, zero sales

By Ryan Richardson · Published 8 October 2026

A 4-dimension ad-copy matrix (hook, angle, identity framing, and copy length) ran 133 ads for A$197.86 on cold traffic. Several cells cleared 8%+ click-through rate. Zero purchases were recorded across the whole matrix.

The question

On cold traffic for a personal brand offer (Nothing / ryanlikes.tech), which combination of 4 copy variables, hook, angle, identity framing (problem vs curiosity) and copy length (short/medium/long), earns the most clicks, using an ad-naming pattern of {hook}-{angle}-{identity}-{length}-{number} to tag every ad for analysis?

Setup

133 ads ran this matrix, 22 Oct 2025 to 7 Oct 2026, each name-tagged on all 4 dimensions so performance could be split any number of ways after the fact.

Results

Within the matrix, individual cells ranged as high as 9.89% CTR (angle 'cant_evaluate') and as low as 4.04% (angle 'effort'); full per-dimension breakdowns are on the related finding pages for hook, identity, length and angle.

MetricValue
Ads133
SpendA$197.86
Impressions33,375
Link clicks2,720
CTR8.28%
Landing page views1,500
Purchases0

What won, and how sure to be about it

8.28% overall CTR on 33,375 impressions is a strong top-of-funnel number by this account's standards, and several individual cells (curiosity framing, short-length copy, the 'cant_evaluate' angle) each cleared 8%+ in their own right. None of it converted: zero purchases across all 133 ads. With no purchases at all, this matrix cannot say which combination of hook/angle/identity/length sells; it can only say which combination gets clicked, and even that comparison is thin once the matrix is split 4 ways across 133 ads.

What we changed next

No documented follow-on decision for this specific test was found in the pulled sources.

How to run this test yourself

A 4-dimension matrix needs either a much larger ad count or a narrower set of dimensions to produce a usable sample per cell; 133 ads split across hook x angle x identity x length produces individual cells too small to read reliably, even though the headline CTR numbers look solid.

The numbers
ClaimValueSource
Ads133Sixty Steps ad account, Meta Ads data
SpendA$197.86Sixty Steps ad account, Meta Ads data
Impressions33,375Sixty Steps ad account, Meta Ads data
Link clicks2,720Sixty Steps ad account, Meta Ads data
CTR8.28%Sixty Steps ad account, Meta Ads data
Landing page views1,500Sixty Steps ad account, Meta Ads data
Purchases0Sixty Steps ad account, Meta Ads data
Ad-name pattern is {hook}-{angle}-{identity}-{length}-{number}confirmedSixty Steps ad account, Meta Ads data
Highest single-cell CTR in the matrix (angle 'cant_evaluate')9.89%Sixty Steps ad account, Meta Ads data
Lowest single-cell CTR in the matrix (angle 'effort')4.04%Sixty Steps ad account, Meta Ads data
Go deeper

This page covers one step. The full method is in the book.

Read the full part
sixtysteps.co
AI-powered growth for what's next. Sixty Steps is Onwards Analytics — a data and analytics firm.
Pages
Home Library The book — $7 About
Start
The Teardown — free [email protected]
© 2026 Onwards Analytics Every claim on this site carries a number, or is marked as reasoning.