How to repurpose a podcast into shorts: the one-hour workflow

We earn a commission if you sign up through some of our links. It doesn't change what we write — we'd tell you the same thing either way.

You have a 45-minute podcast episode. You want 5–8 vertical shorts across TikTok, Reels, and YouTube Shorts this week. Here is the fastest honest path — the steps we run on a weekly episode, and where the AI will try to trick you into shipping bad clips.

The whole thing takes about an hour per episode once you have one tool account and a basic editor installed.

Who this is for

Weekly (or fortnightly) podcasters recording 30–90 minute episodes, video or audio-first, who want consistent short-form output without hiring an editor. If you already pay an editor who handles this, skip the guide. If your podcast is audio-only with no face or screen recording, this still works but you'll live on the "skip" column for the visual polish steps.

What you'll need

• A source episode — video preferred (multitrack if remote), audio-only works with a waveform visual.
• One AI clipper account. Our default is Opus Clip. Free tier clips with a watermark; Pro is $29/mo monthly or $19/mo annual.
• A caption polish tool (optional but recommended). Submagic Starter is $19/mo (or $15/mo annual) if you care about matching your brand.
• A scheduler (optional). Vidyo.ai Essential at $39/mo (or $20/mo annual) handles 7 platforms if you don't want to upload manually.
• A basic editor you already know. CapCut (free), Premiere, or DaVinci Resolve. Used for one 60-second cleanup pass at the end.
• Budget: $20–$60/mo depending on how many of the above you actually buy. The minimum honest stack is one clipper. Everything else is polish.
• Skill: you've edited a video before. You know what 9:16 means. You don't need timeline chops.

The steps

1. Pick your source: full episode vs segment

Action: Decide before you upload. A 45-minute episode uploaded whole returns 8–15 candidates and decision fatigue. A pre-picked 12–15 minute segment you already know contains the three best moments returns 4–6 tighter candidates you actually use.

Expected output: A single video file (MP4, ideally 1080p). Name it specifically — ep47-segment-ai-ethics.mp4 beats podcast-shorts.mp4 when you have 10 exports to sort.

Common failure: Uploading the full 90-minute episode "to let the AI decide." The AI decides based on keyword density, speech pace, and silence gaps — not on whether the moment is interesting. Curated sources consistently produce better clips.

If you don't know which segment to cut, scrub at 1.5x and mark timestamps where you leaned in. Export those ±3 minutes and upload that.

2. Upload to your clipper

Action: Upload the source to your AI clipper. We default to Opus Clip because the clipping engine is fast (a 45-min upload returns candidates in under 5 minutes on Pro) and the pricing is transparent — 1 credit ≈ 1 minute of source, 300 credits/mo on Pro covers ~5 hours of source. If multilingual captions are a must (Spanish, French, German promoted as first-class), swap in Vidyo.ai. If you want to stay on a free tier while testing, Vizard has a real permanent free tier (60 credits/mo, watermarked 720p) instead of a 7-day trial — see our Opus Clip vs Vidyo.ai comparison and the full clipper round-up for the tradeoff matrix.

Expected output: A processing queue that finishes in 3–8 minutes for a 45-minute source on paid tiers (longer on free). You should see a list of clip candidates, each with a thumbnail, length, and some kind of "virality" or "score" label.

Common failure: Free tier + long source = queueing behind other free users. If your test upload is still "processing" after 20 minutes, you're not broken — you're waiting. Plan to upgrade if this is a weekly workflow.

3. Review candidates — don't trust the virality score

Action: Open every candidate clip and skim the burned-in transcript. Rank them yourself. The virality score is a keyword-density heuristic dressed up as a prediction — it rewards clips with questions, numbers, and surprise words. It does not know whether the clip makes sense out of context.

Expected output: 5–8 clips you'd actually post, marked "keep." Anything less than 4 and you under-filtered your source — go back to step 1 with a longer segment. Anything more than 10 and you're posting too much; your audience will fatigue before your clips do.

Common failure: Keeping the top-5-by-score and shipping blind. We've shipped a clip that started with "...so like I was saying" because the AI flagged it as high-virality. Audiences don't forgive mid-sentence openings.

Every keeper should pass three tests: (a) opens on a complete thought, (b) ends on a line that doesn't beg for context, (c) you can mute it and still know what it's about from the captions. If any fails, trim inside the clipper or cut it.

4. Polish the captions

Action: You have two routes. Route A — leave the clipper's default captions alone, accept they look like every other AI-clipped short on your For You page, and ship. Route B — re-run the clip through Submagic to apply a branded caption style (your font, your color, your highlight rules) and optionally add auto-emojis or keyword highlights that match your show's personality.

Route A takes 0 minutes. Route B adds 5–10 minutes per clip but makes your shorts look like your shorts and not like every other podcast clip. See Opus Clip vs Submagic for why creators move between them.

Expected output (Route B): A polished MP4 with captions that match your brand font and color, keyword highlights on the words you actually care about, and emojis where they make sense. Submagic's Starter tier ($19/mo, $15 annual) caps at 40 videos/month and 5 min max length — fine for 5–8 sixty-second shorts per week.

Common failure: Over-styling. Three caption colors, four emojis per sentence, and a lower-third banner will make a watchable clip unwatchable. Pick one highlight color, one emoji cadence (roughly one every 5–8 seconds), and stop.

5. Export vertical

Action: Export each keeper at 9:16, 1080×1920, MP4 (H.264), 30 fps. These are the specs TikTok, Reels, and Shorts accept natively without re-encoding. 60 fps is fine but wastes bandwidth for podcast cuts.

Expected output: An MP4 per clip, typically 5–25 MB for a 45–75 second short. File names that match your segment. Save them to one folder per episode.

Common failure: Exporting at 720p "because the source was 720p." Upscale to 1080×1920 in the clipper's export settings — all three platforms push 720p clips lower because their compression compounds on your lossy file. If the clipper won't upscale, run it through CapCut's 1080p export first.

Quick editor pass: open each clip in CapCut for 60 seconds. Trim 0.5s off the head if it starts on breath. Trim the tail if the last word fades.

6. Upload to TikTok, Reels, and Shorts

Action: Upload to each platform. Manual native posting adds a TikTok cold-start algorithm boost at the cost of 15 minutes per clip. Automated: Vidyo.ai's Essential ($39/mo, $20 annual) or Buffer schedules across 6–7 platforms in one pass.

Expected output: Clip live on each platform with a written caption (90–140 chars), 3–5 hashtags about the moment (not the show), and a cover frame you picked — not the auto-selected mid-blink face.

Common failure: Identical captions across platforms. TikTok favors question hooks, Reels favors descriptive openers, Shorts rewards keyword-heavy titles. Write three variants or accept lower distribution on two of the three.

If this works, what's next

• Pre-edit your episodes with Descript first. Run the episode through Descript before clipping — filler-word removal, silence compression, and Studio Sound clean up the source so the clipper has less garbage to work with. Hobbyist is $24/mo ($16 annual) for 10 hrs/mo. Adds 15 minutes at the top of the workflow, shaves better clips out the bottom.
• Add a scheduler. If you're clipping weekly, the 45 minutes you spend manually uploading across three platforms is more expensive than Vidyo.ai Essential's $20/mo annualized. Opus Clip vs Vidyo.ai breaks down when the scheduler math flips.
• Batch two episodes at a time. Once the workflow fits in an hour per episode, running two back-to-back on Sunday is closer to 90 minutes total than 120 — the mental warm-up cost only hits once.

Honest caveat

This workflow gets you consistent, watchable shorts. It does not get you viral shorts. Virality is a function of your show's hook quality and your audience's existing interest — no AI tool makes a boring moment interesting. If you've been posting for six weeks and clips aren't landing, the problem is upstream, not in the clipper.

Don't let the clipper's pricing pressure you into daily output. Three good shorts a week beat eight mediocre ones from the same episode. The "300 credits per month" on Opus Clip Pro is a ceiling, not a target.

Skip this workflow if your podcast is 10 minutes or less. Pick two moments manually in any editor and export vertical. The clipper tax isn't worth it below ~15 minutes of source.

Tools we recommend

• Opus Clip — primary clipper. Fastest, cheapest transparent price, published affiliate program. Pro at $19/mo annual is the sweet spot for a weekly podcaster.
• Submagic — caption upgrade. Skip if you're fine with default AI caption styles; essential if your brand needs to look like your brand.
• Vidyo.ai — optional scheduler. Essential tier unlocks 7-platform scheduling at $20/mo annualized. Pick this over Buffer if you're already clipping here.
• Descript — optional pre-edit. Transcript-first editing for filler removal and Studio Sound. Only worth it if your raw episodes need cleanup before clipping.

FAQ

How long does this workflow actually take? About 60 minutes per episode end-to-end after your first run: 5 minutes to pick the source, 5 minutes to upload and wait, 15 minutes to review candidates, 10–15 minutes for caption polish (skip for 0), 5 minutes to export, 20 minutes to upload across platforms. First run adds ~20 minutes for account setup and brand preset configuration.

Do I need to pay for a clipper or can I do this free? Vizard has a permanent free tier (60 credits/month, watermark, 720p export). Opus Clip's free tier also clips but adds a watermark. For one test episode, free is fine. For weekly output, the watermark kills credibility on TikTok specifically — Pro at $19/mo annualized is the honest minimum for a real podcast.

Can I skip the editor step at the end? Yes, if the clipper's output opens and closes cleanly. About a third of the time it does. The other two-thirds, a 60-second CapCut pass to trim breath on the head and fade on the tail is the difference between a clip that feels produced and one that feels auto-generated.

What specs does each platform actually want? All three (TikTok, Instagram Reels, YouTube Shorts) accept 9:16, 1080×1920, MP4 H.264, 30 fps. Export once at those specs and upload everywhere. Don't re-encode per platform — you lose quality each pass.

Why five to eight clips per 45-minute episode and not more? Because your audience's tolerance for the same podcast across their feed is finite. Eight clips per episode, posted across the week, puts one clip per day in front of someone who follows you on two platforms. More than that and you train the algorithm to show your posts less. We've tested posting 12 per episode — engagement per clip dropped by roughly half.

What if my podcast is audio-only? The same workflow works with a static image or waveform visual. Most clippers auto-generate a waveform if you upload MP3. The captions and virality detection work off the transcript, not the visual, so clip quality is comparable. You'll lean harder on caption polish because the captions are most of what the viewer sees.