Skip to content

Guides

The fastest way to clean up 20,000 photos

A big camera roll doesn't need one heroic afternoon. It needs a method that survives being interrupted. Here's one that works.

Updated 4 min read

Twenty thousand photos is not a cleaning problem. It's a decision-making problem. If each photo takes you three seconds to judge, that's sixteen hours. Nobody has sixteen hours, which is why the standard approach — open Photos, scroll to the beginning, start deleting — fails within about nine minutes.

The way out isn't going faster. It's making fewer decisions.

Rule 1: never start at the beginning

Scrolling from your oldest photo forward is the worst possible order. Your oldest photos are your most sentimental and your smallest, so you're spending your best attention on the hardest calls for the least reward.

Start where the decisions are easy and the payoff is large.

Rule 2: sort into piles first, decide second

Deciding is slow because your brain switches modes constantly — a screenshot, a sunset, a receipt, your dog, a document. Each switch costs you.

Batch by type instead, and you make one kind of judgement at a time:

Tier 1 — no emotional cost at all

  • Screenshots
  • Photos of documents, receipts, whiteboards, parking spaces
  • Downloaded images from messaging apps
  • Exact duplicates

Tier 2 — no loss, just size

  • Large videos you've never rewatched
  • Live Photos you don't need as Live Photos

Tier 3 — one real decision each

  • Bursts and near-identical shots, where you pick the keeper
  • Blurry or badly-exposed shots

Tier 4 — leave it alone

  • Everything else

Most people are done after tiers 1 and 2, and most of the space came from there anyway.

Rule 3: ten minutes, not one afternoon

A big session feels productive and is a trap: you burn out, you stop, and the library still looks full so you never come back.

Ten minutes a day for two weeks beats one three-hour Saturday, for a boring reason. Attention degrades. After about fifteen minutes of judging photos you stop judging and start keeping everything, which means you're doing the work without getting the benefit.

Short sessions also mean you're never far from a stopping point, so starting is cheap.

Rule 4: make deletion reversible, then stop worrying

The reason people hesitate over every photo is fear. It's slow, and it makes the whole thing unpleasant.

Remove the fear structurally:

  • On iPhone, deleted photos sit in Recently Deleted for 30 days. It's an undo button with a month-long window.
  • Take a backup before a big session if it lets you relax — the whole point is to stop second-guessing.
  • Use something that separates deciding from deleting, so a wrong swipe costs you nothing until you confirm the batch.

Once you genuinely believe a mistake is recoverable, you'll move about three times faster.

Rule 5: one decision per photo, one gesture per decision

Here's the actual bottleneck. In a standard photo app, deleting one photo is: tap to open, look, tap back, tap Select, tap the photo, tap the bin, tap Confirm. Seven interactions.

The alternative is one photo on screen, one gesture, next photo. That's it. Twenty thousand photos is impossible at seven interactions each and merely tedious at one.

This is why the swipe interface caught on — not because swiping is fun, but because it removes every interaction except the judgement itself.

Rule 6: track the number

Storage cleanup has no natural feedback. You delete 400 photos and the phone looks identical. That's why people give up even when it's working.

Watch a number instead. Space freed, photos cleared, days in a row. It's a small trick and it works, because it turns an invisible chore into something with a score.

A two-week plan

  • Day 1 — screenshots. The biggest, easiest pile. Expect to remove several hundred without a single hard call.
  • Day 2 — large videos. Sort by size, look at the top twenty. Usually the largest single win of the whole fortnight.
  • Day 3 — duplicates. Free space, nothing lost.
  • Day 4 — messaging app saves. Everything WhatsApp, Instagram and Telegram put in your roll without asking.
  • Days 5–10 — near-identical shots, one month at a time. Pick the keeper, drop the rest.
  • Days 11–14 — blurry shots, then stop. Anything left is worth keeping.

Two weeks, ten minutes a day, and the pile is gone.

Then stop it coming back

  • Turn off automatic saving of media in WhatsApp, Instagram and Telegram. This alone prevents most future clutter.
  • Drop your video recording quality to 1080p unless you're actually filming something you'll edit.
  • Delete screenshots the day you take them. It takes four seconds and it's the whole problem, solved at the source.

Where SwipeDrop fits

SwipeDrop is built around exactly this method. It sorts your library into those piles automatically — screenshots, large files, videos, duplicates, similar shots, blurry shots, app saves — and shows you how much space each one would free, so you always know where the ten minutes should go.

Then it gets out of the way: one photo, one swipe. Left drops it, right keeps it. Nothing is deleted while you swipe; your drops collect on a review screen and one confirmation clears the batch. It keeps score of what you've freed, remembers a half-finished session so an interruption costs you nothing, and does the whole thing on your phone — no upload, no account, no tracking.

Frequently asked

How long does it take to clean up 20,000 photos?
At three seconds a photo it would be about sixteen hours, which is why nobody finishes. Working in piles cuts it to a few sessions: screenshots and videos are quick and account for most of the space, and the near-identical shots go fast once something groups them for you.
Should I delete photos or just buy more storage?
If your library is mostly photos you want, buy the storage — it is cheap and your time is not. If it is mostly duplicates, screenshots and eleven takes of every moment, more storage just postpones the same afternoon, at a monthly cost.
What is the best order to delete in?
Easiest decisions first, biggest files first, hardest decisions last. Screenshots, then videos and large files, then duplicates, then similar shots, then everything else. Starting with the emotionally hard photos is the reliable way to give up in nine minutes.