Why We Built Auto Vertical into Clipr (And Why It Took Longer Than We Thought)

The Clip That Started It All

Most pastors and podcasters record in landscape. Sensible decision. Pulpits face horizontal. Cameras sit on tripods framed for human eye contact. The moment you try to turn a 16:9 sermon into a vertical TikTok clip, though, you hit a wall.

Manually cropping is tedious. Zoom and pan tools exist, but they eat time. And you're guessing at composition. Do you keep the face? The hands? The text on the screen behind the speaker? A 45-second clip can take 20 minutes if you want it to look right.

Our Manchester manager was doing this for five clips a week. For a church with two people on the social media team. The maths didn't work.

We'd designed Clipr to save time on the hard part: finding which moments are worth clipping. Turns out we'd built half the solution. The other half was making sure the clip actually played well on the platform where it would live.

Fitting the Frame, Not Cropping the Story

Auto vertical 9:16 reformatting in Clipr isn't just a zoom. That was our first instinct, and it was wrong.

Zooming in loses context. Letterboxing (black bars) wastes the whole point of vertical video. What you actually need is intelligent cropping that preserves the subject while adapting the composition.

We spent time testing how different sermon moments behave in portrait. A wide two-shot of co-preachers needs different framing than a close-up on someone's expression during an emotional pause. A title slide behind a speaker reads differently when it's squashed vertical.

The solution we landed on learns the key subject in each frame and preserves it. During the clip selection phase, the system doesn't just score moments for engagement. When you export, it reformats for vertical while keeping the human at the centre of the frame. You don't lose the speaker's face. You don't sacrifice the visual story.

This is available in Creator tier and above. It ships baked into the output file, not as an overlay or adjustment layer. You download, and it's ready.

The Technical Bit (Without Boring You)

Plenty of video tools offer aspect ratio conversion. Most are clunky. We wanted Clipr's vertical reformatting to feel invisible. You shouldn't notice the work. You should just see a clip that looks native to TikTok or Reels.

The process runs locally on your phone or tablet. Apple's on-device speech transcription gives us the timeline of what's being said. The same scoring system that identifies engagement-worthy moments also tracks visual subjects frame by frame. When you choose export, the reformatting happens in realtime using that data.

Why does this matter? Speed. On-device processing means no cloud upload, no waiting for a server to process your video, no privacy concerns about your sermon sitting on someone's infrastructure. Most UK churches have decent broadband, but not every edit suite does. Three minute sermon. Ten seconds to process. Done.

One early tester asked if we could show what the tool was doing. That led to Pro tier, where you get a faith score explanation per clip. Not because we wanted to sound clever. Because creators deserve to understand why the system ranked one moment higher than another.

What Changed After We Shipped

Launch week was humbling. Half our feedback wasn't about the vertical reformatting itself. It was about how it paired with auto-captions.

A sermon clip with proper captions and proper framing feels like native content. Not like you filmed something in widescreen and desperately tried to make it fit. The two features live together in Creator and Pro tiers. Captions baked in, vertical reformatted, watermark gone.

We also learned that batch processing matters more than we expected. Pro users can process five videos at once. That's not a random number. We watched a church social media manager process 15 clips across three sermons. They were frustrated. Not at Clipr, but at the repetitive clicking. Five at a time means one click, walk away, come back to five finished clips.

The feedback we didn't expect came from podcasters. We'd built Clipr for sermon clips, but podcast clip creators found it just as useful. A 90-minute teaching podcast gets clipped the same way a 40-minute sermon does. Vertical format works the same way. That surprised us, and it's shaped how we're thinking about what comes next.

Where It Falls Short (Honestly)

Auto vertical isn't magic. If you film a sermon with a camera position that's truly terrible (waist-height angle looking up, extreme side-angles, extreme close-ups of notes), no reformatting tool fixes that. It works because most sermon filming is competent. You frame a speaker for a human audience, and vertical reformatting adapts that frame for mobile.

It also doesn't touch your upload. Clipr exports a clip. You still manually upload to TikTok or Reels or YouTube Shorts. We could have built uploader integration, but that felt like feature creep. Export a good clip locally, keep your upload workflow your own. Some creators prefer that flexibility.

Watermark removal is Creator tier and up, not free. That decision split our team. The free tier is meant for people testing the concept. Two clips a month, watermark on them. It's honest. If Clipr saves you time, you'll upgrade. If it doesn't, we don't deserve payment.

The Thing We Still Get Right

Six months after launch, the Manchester church manager wrote again. They'd moved three clips a week to five. Same team, same hours. One person now handles social where two were splitting it before. They're not using Clipr's vertical reformatting in isolation, of course. They use the whole workflow: moment scoring, captions, vertical frame, batch export.

But they called out the vertical feature specifically. Not because it's clever. Because it removed a decision. They don't think about aspect ratio anymore. The clip is the right shape when it lands on their phone. That's the bar. Not impressive. Just reliable.

That matters in ministry. Time saved on logistics is time you can spend on strategy. Why are you clipping this sermon anyway? What story does the congregation need to see? When the tool handles the technical parts, those questions bubble up.