Latest / Video Marketing with Fexingo: Short Form, Long Form, and Video-First Strategy / Why Your Video Strategy Needs a Thumbnail-First Approach
Transcript
- Lucas: So I want to start with a number that kind of stopped me this week. YouTube says that ninety percent of the top-performing videos on their platform have custom thumbnails. Not auto-generated grabs from the video — custom-designed images. Luna: That is a staggering stat. I think a lot of creators still treat the thumbnail as an afterthought — they spend hours on the edit and then just pick a frame. Lucas: Exactly. And I think that's the disconnect. If this conversation has already made you rethink how you approach thumbnails, that's exactly why we do this show. The way we keep these episodes ad-free is listener support — buy me a coffee dot com slash fexingo. It's a small thing that lets us keep digging into angles like this without any sponsor interruptions. Luna: Yeah, and it really does make a difference. We hear from listeners who say these specific tactics have changed their workflow. So if you've gotten value, that link is the simplest way to keep it going. Lucas: Alright, back to thumbnails. So that stat — ninety percent — tells me that the people who are winning treat the thumbnail as a separate creative asset, not just a screenshot. Luna: What does that actually look like in practice? Like, if I'm a brand with a small team, how do I start thinking about thumbnails differently? Lucas: I'd say start with a simple three-layer framework: clarity, contrast, curiosity. Clarity means the viewer knows exactly what they're going to get in under a second. Contrast makes the thumbnail pop in a feed full of similar content. And curiosity leaves a small gap — not a confusing gap — that makes you want to click. Luna: So clarity first. Can you give me an example of a thumbnail that gets clarity wrong? Lucas: Sure. Think of a tech review where the thumbnail is just the product on a messy desk with no text and no expression. I have to squint to figure out what it is. Compare that to MKBHD — he always has the product clearly centered, a clean background, and maybe two or three words of text like 'This is NOT good.' You know exactly what the video is about. Luna: Right. And contrast — that's about color, but also about brightness, right? Like a bright orange text on a blue background? Lucas: Exactly. Complementary colors work really well. But contrast isn't just color — it's also about subject separation. If your face blends into the background, you're losing that pop. A lot of top creators use a subtle drop shadow or a rim light in the thumbnail to separate the subject. Luna: And the curiosity gap — that's the tricky part. How do you do it without being clickbaity? Lucas: I think the line is that the thumbnail has to deliver on what the video actually contains. Curiosity shouldn't mean misleading. So instead of 'You won't BELIEVE what happened,' it's more like showing a surprising result from an experiment you actually ran. MrBeast does this really well — his thumbnails often show a dramatic moment that genuinely happens in the video, but framed in a way that makes you want to see how they got there. Luna: I want to push back a little on the idea that this only applies to YouTube. On Instagram Reels and TikTok, the thumbnail is the first frame of the video — or you can upload a custom cover image. But I feel like a lot of brands just let the algorithm pick the first frame. Lucas: That's a really good point. And actually, Instagram and TikTok both show static previews in the feed before the video starts playing — especially on slower connections. So even if you're doing short-form, that static image matters for whether someone scrolls past or stops. Luna: Yeah, I've noticed that on TikTok, when I'm scrolling quickly, the thumbnail is often what catches my eye before the auto-play kicks in. So if it's a blurry frame from the middle of the video, I'm already gone. Lucas: Exactly. So the same principles apply. You want that first frame to have clarity, contrast, and curiosity — even if it's just a still from the video that you carefully select. Luna: What about tools? Is there a way to A/B test thumbnails without a huge budget? Lucas: Absolutely. YouTube Studio has a built-in thumbnail testing feature now — you can upload up to three thumbnails and it'll show each one to a portion of your audience, then pick the winner. It's free and it's incredibly useful. For social platforms, you can test manually by posting the same video with different thumbnails on different days, or use a tool like Thumbnail Test — though that's more for YouTube. Luna: Let's talk about text on thumbnails. I see a lot of debate about whether you should use text overlays at all. Lucas: I think text can be powerful, but it has to be minimal. Three words max, ideally two. Large, bold, sans-serif font. And it should complement the visual, not repeat it. If your image already shows a person crying, you don't need text that says 'they cried.' The text should add new information — like '3M views' or 'banned' — that creates that curiosity gap. Luna: What about faces? There's a lot of research showing that faces with strong emotions drive higher click-through rates. Lucas: Absolutely. Eye tracking studies show that when we see a face in a thumbnail, we look at the eyes first. So if the face is expressing surprise, shock, or joy, that emotion transfers to the viewer. That's why you see so many thumbnails with exaggerated expressions — it works. But it has to feel authentic to the content. If your video is a calm tutorial and your thumbnail looks like you just won the lottery, that's a disconnect. Luna: Netflix is actually a great case study here. They've talked publicly about how they A/B test thumbnails for every piece of content — sometimes hundreds of variations. And they found that a thumbnail featuring a single character with a strong emotional expression outperformed group shots or landscape shots almost every time. Lucas: That's fascinating. And it shows that even with a massive budget, the same principles apply. They're not reinventing the wheel — they're just rigorously testing what works. Luna: So for a small brand, where do you start? If you're doing one video a week, how much time should you spend on the thumbnail? Lucas: I'd say allocate at least twenty percent of your production time to the thumbnail. If you spend four hours on a video, that's almost an hour on the thumbnail. That might sound like a lot, but when you consider that the thumbnail is responsible for most of the click-through, it's actually under-invested. Luna: I've also seen some creators use what they call 'thumbnail-first scripting' — where they design the thumbnail before they even write the script. That way the video is built around delivering on the promise of that image. Lucas: That's smart. It forces you to crystallize the value proposition of the video into a single frame. And it prevents you from making a video that's too broad to capture in a thumbnail. Luna: Alright, so let's say I have a thumbnail idea. What's the final step before I upload it? Lucas: Check it at thumbnail size — literally shrink it down on your screen. Does it still read? Can you tell what's happening? If you have to squint, it's too cluttered. The best thumbnails are simple enough to be effective even on a smartwatch. Luna: That's a great rule of thumb — no pun intended. I think a lot of creators design their thumbnails zoomed in and then are disappointed when they look messy in the feed. Lucas: Exactly. So to wrap it up: think of your thumbnail as a billboard for your video. It's not an afterthought — it's the most important frame you'll create. Use the clarity, contrast, curiosity framework. Test relentlessly. And don't be afraid to spend real time on it. Luna: I'd love to see a future episode where we actually walk through creating a thumbnail from scratch — maybe even live on the show. Lucas: That's a great idea. We'll put that in the pipeline. For now, go look at your last few videos — are the thumbnails working as hard as the content? If not, you know where to start.