How long does culling actually take? Do the math, then beat it
Ask ten photographers how long they spend culling and you'll get ten shrugs. Almost nobody times it, the cull just bleeds into the evening until it's done. But culling time isn't a mystery. It's a formula with two inputs, and once you know yours, you can attack both.
Culling time = frames × seconds per decision (+ overhead)
Every cull reduces to the same arithmetic: how many frames you shot, times how long each decision takes, plus the overhead nobody counts: waiting for previews to render, zooming to check focus, and flipping back to compare near-duplicates.
The seconds-per-decision number is where photographers differ wildly. A disciplined culler on a binary pick/reject system with auto-advance spends a second or two on obvious frames. A perfectionist who zooms to 100% on every shot spends ten or more. Run the math on a wedding and that difference is not minutes. It's an entire evening:
- 3,000 frames × 2 seconds = 100 minutes of pure decision time
- 3,000 frames × 4 seconds = 200 minutes, over three hours
- 3,000 frames × 8 seconds = 400 minutes, your whole Sunday
And that's decision time only. Add rendering waits and re-reviews and the real number climbs further. This is why two photographers shooting identical weddings report culls of "an hour and a bit" and "most of a day": same frames, different seconds.
Worked examples, plug in your own numbers
The best-documented volume is weddings: industry norms put a full 8 to 10 hour day at 2,000 to 4,000 RAW frames captured, with 400 to 800 images delivered. Other genres vary too much by shooter to pin down, so treat the non-wedding rows below as illustrative inputs, swap in your own frame counts.
| Shoot | Frames (input) | At 3 sec/frame | Reality check |
|---|---|---|---|
| Full wedding day | 2,000 to 4,000 (typical range) | ~1.7 to 3.3 hrs | Doubles fast if you pixel-peep |
| 1-hr portrait session (example) | say, 400 | ~20 min | Feels like an hour once you add comparisons |
| Mini-session day (example) | say, 8 galleries × 150 | ~1 hr total | Eight separate client contexts. Switching costs add up |
| Headshot session, tethered (example) | say, 200 | ~10 min | Often near zero, clients pre-select at the screen |
Notice the pattern: the cull scales with frames shot, not hours booked. Which means shooting discipline, fewer sprayed bursts, more deliberate frames, is itself a culling strategy. So is handing the first pass to software, which we'll get to.
One more wrinkle the table hides: context switching. A mini-session day isn't one 1,200-frame cull. It's eight small culls, each with its own client, light, and story to reload into your head. Photographers consistently report those days feeling slower than the raw frame count predicts, so batch them: cull all eight in one sitting rather than one gallery per evening for a week.
What "11% of your working time" really means
The most useful hard data we have comes from a UK survey of 300+ wedding photographers (2020, worth noting that's before the current AI-tool era). The time split: 55% editing, 18% admin, 11% culling, 7% communication, and just 4% actually shooting.
Eleven percent sounds small until you convert it. On a 40-hour week, that's about 4.4 hours of culling, every week. Respondents averaged 28 weddings a year: call it 28 evenings of sorting, per year, before a single edit happens. Full breakdown of where the rest goes in our teardown of the survey.
And culling punches above its 11%: it's the gate in front of the 55%. Every frame that limps through the cull "just in case" becomes editing time downstream. A tighter cull is the cheapest editing speedup there is.
Time yourself once, seriously
Next cull, start a timer and note your frame count. Divide. That's your seconds-per-decision baseline, and it's the single most useful number in this article, because now every change, flags instead of stars, previews pre-rendered, an AI first pass, has a measurable before and after. Most photographers guess their cull takes "about an hour." Most are off by 2× or more.
Where the hours hide, and how to claw them back
Rendering waits, the silent tax
If frames take a beat to sharpen on screen, you're paying that beat thousands of times. Build previews on import and cull only when they're done, the walk-away trick that anchors the fast Lightroom workflow.
Decision speed, go binary
Keep or reject. Two buckets, auto-advance, hands never leaving the keyboard. Star scales force a five-way judgment on every frame and quietly triple your seconds-per-decision.
Re-reviews: one pass, forward only
Second-guessing is the sneakiest multiplier: cull the same wedding one-and-a-half times and your "3,000 frames" was really 4,500 decisions. The fixes live in our culling mistakes rundown, and the full two-pass method is in how to cull a wedding in under two hours.
The first pass, delegate it
Blinks, motion blur, hopeless exposures, and burst duplicates are exactly what AI culling tools catch reliably. Let software make the obvious calls and your decision count drops before you've judged a single frame. You review a shortlist instead of a card dump. For the complete picture of the craft, start with the complete culling guide.
FAQ
How long does it take to cull a wedding?
A typical wedding produces 2,000 to 4,000 RAW frames. At a realistic three seconds per decision, that's roughly 1.5 to 3.5 hours of pure decision time for a manual cull: before preview rendering, breaks, and second-guessing add more. Photographers with a strict two-pass system land near the bottom of that range; perfectionists culling in edit view can double it.
Is culling or editing the bigger time sink?
Editing, by a wide margin, a UK survey of 300+ wedding photographers found editing takes 55% of working time versus 11% for culling. But culling sits upstream: every extra frame that survives the cull becomes editing time, so a tighter cull shrinks both numbers.
Does AI culling actually save time?
It compresses the first pass: flagging blinks, blur, and near-duplicates so you review a pre-sorted set instead of judging every frame from zero. You still make the final calls, so the honest framing is that AI turns hours of decisions into a shorter approval session, not zero work.
How can I cull faster without buying software?
Three changes: render previews before you start (never cull against loading spinners), switch to binary pick/reject flags with auto-advance instead of star ratings, and do a fast rejects-only first pass before choosing what to keep. Most photographers who time themselves find these alone cut the session dramatically.
Never cull alone at 1 AM again.
Kepla for Mac clears the obvious misses from a card, names the reason on every frame it sets aside, and leaves the choosing to you. Nothing is ever deleted, moved or renamed. Free through the private preview · the first hundred photographers keep it at $99 a year.