Why banners need targeted copy, not free creativity
A banner has 3 to 5 seconds to capture attention and communicate the benefit. It's not a brand ad — it's a billboard. Brands that get the best ROI on banners don't aim for clever copy, they aim for clear copy: the what, the how much, and the how long.
Banner structure that converts
- Headline (top): the benefit in caps. "30% OFF STORE-WIDE".
- Sub-headline: context or category. "Women's apparel — summer drop".
- Urgency: time or stock. "Ends Sunday" or "Limited stock".
- CTA: 2-3 action words. "Shop now", "Grab it", "See collection".
- Fine print (when needed): exclusions, min spend. Small but not hidden.
The 5 non-negotiables
- One promise per banner. Don't combine two different discounts.
- Most important in caps and on top. The eye scans the biggest block first.
- Contrasting CTA. Button with contrasting color and present-tense verb.
- Mobile first. 70% of traffic is mobile — design for small screens first.
- Real expiry. If you say "ends Sunday", Monday it's gone.
Copy by promo type
- Percentage discount: "30% OFF EVERYTHING" + "Ends Sunday 11:59pm" + "Shop now".
- Free shipping: "FREE SHIPPING" + "On orders from $50" + "Grab it".
- BOGO: "BUY 2, PAY 1" + "On the summer line" + "See collection".
- Launch: "NEW COLLECTION" + "Online only this week" + "Discover".
- Holiday: "BLACK FRIDAY" + "Up to 50% off + free returns" + "See deals".
Mistakes that kill conversion
- Headlines like "Don't miss out!" without saying what's at stake.
- Vague CTAs: "Click here" or "Learn more". The CTA must promise a result.
- Stacked discounts: "30% + 20% + 10% extra" — customer can't tell what they pay.
- Huge fine print. If exclusions take more space than the benefit, you lost credibility.
- Banner copy that doesn't match the landing. Says "free shipping", landing has conditions — trust drops.
Quick A/B testing
If your platform allows, run 2 banner versions for 24-48h. The biggest CTR levers: CTA verb, countdown presence, and whether the % goes before or after the product. One variable at a time, otherwise you don't know which won.