Primary Text Testing with Controlled Creative: Advanced Facebook Ad Creative Testing

Quick answer

Test primary text in isolation by locking the visual and headline. Run 5 variants spanning length, angle, and structure. Length matters most: short (0-90 char), medium (200-400), long (600+). The right length depends on category — there is no universal winner.

Why primary text is misunderstood

Primary text is the body copy above the image. On Feed it is collapsed after 125 characters with a "see more" link. On Reels it is barely visible. Most advertisers either skip it (lazy) or write essays (vain). The right approach is to test it deliberately and let conversion data decide.

The framework, step by step

1. Lock the visual, headline, and CTA. Pick your evergreen winner.
2. Write 5 primary text variants:

- V1: 1-line short hook (under 90 chars) - V2: 3-line medium (200-300 chars) - V3: Long-form story (600-900 chars) - V4: Bullet list (3-4 bullets) - V5: Question-led

1. Launch as iso-variants in one CBO ad set. Run 4-5 days.
2. Read CTR, CPA, and link click depth. Long-form often has lower CTR but higher CPA-quality clicks.
3. Promote the winner. Save the structure pattern for future briefs.
4. Re-test by category and audience — long-form wins B2B; short wins ecom impulse.

Example test matrix

Variant	Length	Structure	Example opening
V1	75 char	Hook only	Stop wasting Fridays on reports.
V2	240 char	Hook + proof	Stop wasting Fridays on reports. We tested this with 200 marketers — 4 hours saved.
V3	700 char	Story	Last year I spent every Friday on report cleanup…
V4	180 char	Bullets	• 4 hours saved\n• 200 marketers tested\n• Free trial
V5	120 char	Question	What if Friday afternoons were yours again?

Pitfalls to avoid

• Writing all five at the same length. The point is to span the length space.
• Front-loading the brand name. The first 90 characters should hook, not introduce.
• Burying the offer below the "see more" cut. Critical info goes in the first 90 chars.
• Comparing primary text variants on different creatives. Lock the creative first.
• Confusing engagement with conversion. Long primary text gets shares but not always sales.

5 FAQs

What is the average best length? Mixed by category. Test it for your account.

Should the primary text repeat the visual? No. The visual carries one message; primary text adds context.

Are emojis effective in primary text? Yes for ecom, mixed for B2B, no for fintech.

Should I include a CTA in primary text? Yes — even if there is a CTA button, a verbal CTA in copy lifts conversion 5-10%.

Does long-form ever beat short? Yes for SaaS over £100/month, courses, and consulting.

Test text without rebuilding ads

Pix-Vu lets you swap primary text variants on the same creative in seconds. https://pix-vu.com.