Improve your visuals

Your visual is what the eye sees first when scrolling. It determines whether the ad registers at all. A poor visual means your carefully crafted hook and copy never get read.

Most visuals fail because they're generic stock photos, too complex to understand at a glance, or don't work on mobile (where 60%+ of viewing happens). The visual needs to support your hook, not compete with it.

This chapter shows you how to test image formats (single, carousel, data visualisation), video structures (problem-solution-proof, testimonial, demo), and platform-specific visual rules. You'll learn when to use professional photography versus casual screenshots, and how to ensure everything works at small screen size.

Single images are the simplest format. One image, one message. They work when you have a clear visual that supports your hook.

Product screenshots work for tool-chooser segments (comparing features). Show your interface doing the thing they care about. For cybersecurity training: screenshot of the behaviour tracking dashboard (proves you measure impact), screenshot of a training module (shows what it looks like), screenshot of the compliance certificate (proves it satisfies auditors).

People images work for culture-focused buyers. Show employees engaged in training (not stock photos of people pointing at screens). Real photos beat stock every time. For cybersecurity: employees completing a gamified module, a team discussing phishing results, a CISO reviewing behaviour metrics.

Data visualisations work for analytical segments. Show a graph that proves your point. For proactive segment: before/after graph showing click-through rates dropping 47%, ROI calculator showing payback in 6 months, comparison chart showing your results versus industry average.

Test professional versus casual photography. Professional photography (studio-lit, perfectly composed) signals premium but can feel distant. Casual photography (iPhone photos, real offices, real people) signals authentic but might feel cheap. Different segments respond differently. Enterprise buyers often prefer professional. SMB buyers often prefer casual.

Mobile-first rule: will this image work at small size? If the key element (graph axis labels, interface text, person's face) isn't visible on a phone, the visual fails. Test every image on a phone screen before running it.

Carousels let you tell a story across multiple cards. They work when you need to show progression, comparison, or multiple proof points.

Step-by-step processes work for implementation-worried buyers. Show the timeline from setup to results. For breach-reactive segment: Card 1 "Upload your users", Card 2 "Send training invite", Card 3 "Team trained by tomorrow", Card 4 "Track completion in real-time". Four cards showing speed and ease.

Before/after comparisons work for proactive segment. Show the transformation. Card 1 "Before: 23% click-through rate", Card 2 "After 30 days: 12% click-through rate", Card 3 "After 90 days: 6% click-through rate", Card 4 "47% reduction, £180K breach risk avoided". Four cards building the ROI case.

Feature showcases work for tool-chooser segment. Show different capabilities. Card 1 "Gamified training employees actually complete", Card 2 "Phishing simulations that teach", Card 3 "Behaviour tracking that proves ROI", Card 4 "One platform, complete security culture". Four cards covering feature differentiation.

Economics breakdowns work for price-sensitive segments. Show the maths. Card 1 "Average breach cost: £2.4M", Card 2 "Training cost: £15 per employee", Card 3 "Breach risk reduction: 47%", Card 4 "Expected savings: £1.1M". Four cards making the affordability case.

Keep carousels to 3-5 cards. Fewer than 3 and you should use a single image. More than 5 and people don't swipe through. First card must hook. If card 1 doesn't stop the scroll, cards 2-5 are irrelevant.

Videos are more engaging than static images but require more production effort. Use them when motion and demonstration add value, not just for the sake of using video.

Problem-solution-proof structure (30-60 seconds) works for cold traffic. Show the problem (employees clicking phishing emails), introduce the solution (gamified training), prove it works (results data). Keep it tight. For compliance-driven: 10 seconds showing boring annual training, 20 seconds showing engaging gamified alternative, 30 seconds showing completion rates and compliance certificates.

Customer testimonial videos (45-90 seconds) work for warm traffic. Real customer, real voice, real results. For proactive segment: CISO explaining "we needed to prove ROI", showing behaviour metrics dashboard, stating "reduced incidents 63% in 6 months, justified the budget to the board". Authentic beats polished.

Product demo videos (30-45 seconds) work for hot traffic. Show exactly how it works for tool-chooser segment. For breach-reactive: "Here's how fast you can deploy" then screen recording showing upload users, send invite, done in under 2 minutes. Speed and simplicity demonstrated, not just claimed.

Video must work without sound. 90% of people watch with sound off. Use captions for all spoken content. Use text overlays for key points. Test your video on mute to confirm it makes sense.

Choose format based on segment and stage. Cold traffic needs problem-solution-proof (education). Warm traffic needs testimonials (social proof). Hot traffic needs demos (capability proof). Don't use the same video for all segments.

Different platforms have different visual norms. What works on LinkedIn often fails on Facebook. Match your visual style to platform expectations.

LinkedIn wants professional, less flashy, more credible. Use interface screenshots, data visualisations, professional headshots. People are in work mode. Stock photos of people in suits feel appropriate. Casual iPhone photos can feel off-brand. Text overlays work well. Charts and graphs perform.

Google Display wants simple, clear value proposition at a glance. You have less attention than social platforms. Use bold headlines overlaid on images. Minimal text (5-7 words maximum). Clear visual hierarchy (what's the one thing to see?). Product screenshots often fail (too complex). Simple icon + headline + CTA works better.

Facebook and Instagram want more casual, lifestyle context, less corporate. Professional photography can feel too stiff. Real employees, real offices, real situations perform better. Bright colours work. Text overlays common. Memes and casual humour appropriate (if it fits your brand). Video performs especially well.

Test the same message across formats adapted to platform norms. Don't just run identical creative everywhere. The LinkedIn proof-focused single image might need to become a Facebook casual video to work.

Track engagement rate by platform and visual type. If LinkedIn engagement rate is 4% but Facebook is 0.8%, your creative isn't adapted correctly (or Facebook isn't the right channel for your segment). If single images outperform video on LinkedIn but video outperforms images on Facebook, you've learned platform preferences.

Visuals determine whether ads get noticed at all. Test systematically: single images (product screenshots, people images, data visualisations), carousels (step-by-step, before/after, features, economics), videos (problem-solution-proof, testimonials, demos).

Match visual type to segment and awareness stage. Tool-choosers need screenshots showing features. Implementation-worried need step-by-step process. Proactive need data visualisations proving ROI. Breach-reactive need videos showing speed.

Everything must work on mobile at small screen size. Test on a phone before running. If key elements aren't visible at small size, the visual fails.

Adapt visual style to platform norms. LinkedIn wants professional and credible. Google Display wants simple and clear. Facebook wants casual and authentic. Don't run identical creative across all platforms.

Next chapter: improve your calls-to-action to match segment readiness.