The wedding culling workflow, card to gallery

A faster method at the desk only fixes the part of a wedding cull that happens at the desk. The rest of it is decided in your contract, in your card wallet, at midnight when you offload, and in the five different sets of photos one cull is supposed to produce. This is the whole chain, in the order you actually run it.

A method is not a workflow

There is a difference between a culling method and a culling workflow, and the gap costs photographers whole weekends. A method is what you do at the desk on Tuesday: which question you ask each frame, which key you press, how you handle a burst of forty. We covered that in how to cull 3,000 wedding photos in under two hours, and if the desk is genuinely your bottleneck, start there.

A workflow is the whole chain, and most of it happens nowhere near the desk. It starts in the contract, where the number of delivered photos gets agreed. It runs through the cards in your bag, the clocks in your two bodies, the folder you offload into at midnight, the previews that render while you sleep, the passes you make on Tuesday, and the five separate sets of photos one cull should hand you. It ends when the gallery goes out and you write down the single thing you will change next Saturday.

Look at where wedding culls actually go wrong. Two bodies whose clocks were four minutes apart, so the ceremony will not sort. A second shooter's card that landed in a folder nobody named. No agreed delivery number, so the second pass has no finish line and drifts toward keeping everything. Ninety minutes of a Tuesday evening spent watching previews draw. Not one of those is a speed problem, and not one of them is fixed by pressing keys faster.

Four decisions you make before the ceremony

Agree the delivery number. Typical full day galleries land somewhere around 400 to 800 edited images, or roughly 40 to 60 photos per hour of coverage, per ranges published by the platform Snapeen. Pick your number, write it into the contract, and your second pass has something to converge on. A cull with a finish line ends. A cull without one keeps going until you are tired.

Sync every clock in the building. Every body you will shoot, plus every body your second shooter will shoot, set to the same minute before anyone leaves the house. This is the cheapest thing on this list and the most expensive to skip. Sorting by capture time is what lets you pull two cameras into one chronological story on Tuesday, and four minutes of drift turns the ceremony into a shuffled deck. Our guide to culling second shooter photos goes deeper on the merge.

Carry enough cards that you never reformat on the day. A card that gets reused mid reception is a card whose earlier frames exist in exactly one place until you get home. Cards are cheap. Swap and pocket, and write the block on a piece of tape if you are the sort of person who will not remember.

Settle the folder structure and the file names once, permanently. Not per wedding. One pattern you use for every job, so that the ingest is muscle memory at midnight and the archive is searchable in four years.

The habits on the day that shrink Tuesday

You cannot cull on a wedding day, but you can decide how big the cull will be. Four habits do most of the work.

• Do not judge on the back of the camera. A 3-inch screen in daylight tells you almost nothing about focus and lies about expression. Chimping between moments costs you the next moment, and the frames you would discard on that screen are the ones you would have caught in half a second on a real display anyway.
• Shoot a marker between blocks. One deliberate throwaway frame at the top of each part of the day: the back of your hand, the timeline card, the venue sign. On Tuesday those markers are visible dividers in a wall of thumbnails, and grouping the day into blocks becomes a scroll rather than a puzzle.
• Cover the moment, then stop. Wide, medium, detail, the reaction. Once a moment has those, the eleventh frame of it is not insurance, it is a decision you have promised to make later. The moments where you genuinely need frames per second are few and known in advance, mostly the aisle, the kiss, the toss and the exit.
• Brief your second shooter on coverage, not angles. Tell them which moments they own so you are not both shooting the same three seconds from six feet apart. A five minute meeting over the timeline in the morning is worth an hour of your Tuesday.

Nothing gets judged until it exists twice

The night the wedding ends you do three things and none of them involve looking at a photo.

Offload every card into its own source folder first, then merge. Bride prep card, groom prep card, second shooter's three cards, each into a folder that says whose it is. Merge into the chronological working folder as a second step. If something goes wrong you will know exactly which card to re-copy, and the second shooter's frames stay identifiable when you sit down.

Verify, then get to two places, then three. Copy to the working drive and a second drive, and start the upload to whatever offsite copy you use before you sleep. Judging photos on a single copy is the one genuinely irreversible mistake in this whole workflow, and a backup workflow you will actually follow is not a long document.

Render previews overnight. The largest hidden cost in culling is not deciding, it is waiting. Kick preview generation off before bed and Tuesday's frames appear instantly instead of drawing a quarter second at a time, several thousand times. A repeatable version of all three steps lives in our post shoot ingest checklist.

Three passes, and what each one hands you

At a system level the cull is three passes with three different questions. Keeping them separate is the entire trick: mixing them is what turns a two hour job into three evenings, because you end up asking "is this the best frame of the moment" of photos that are out of focus.

Pass one is where most of your frames leave, and it is the pass that asks least of you, which is exactly why it goes first while momentum is cheap. Pass two is the part that is actually your job. Pass three takes ten minutes and pays for itself all week.

Two notes that matter more than they sound. Nothing is being thrown away in any of these passes: frames that do not get picked keep sitting in the working folder, and every one of them is one click from coming back. And pass two goes far faster block by block than in one long chronological scroll, because near identical frames end up next to each other and your job becomes comparison, which people are good at, instead of absolute judgment, which people are bad at.

One cull, five sets of photos

Most photographers run the cull to produce one thing, the gallery, and then go back through the wedding three more times over the following month to find the other four. Every one of those trips is a second cull you already paid for once.

1. The gallery. Your contracted number, the reason you sat down.
2. The sneak peek. Ten to fifteen frames, starred during pass three, edited first and sent while the couple is still checking their phone every ten minutes. This set is worth more to your referrals than any other twenty minutes of your week.
3. The blog and social set. Thirty to fifty frames that carry the story in order, chosen while the story is still in your head rather than reconstructed in November.
4. The album shortlist. Different criteria from the gallery: spreads, pairings, room to breathe, verticals that face the right way. Picking it during the cull instead of after the design fee is agreed is the whole argument in album selection versus gallery selection.
5. The vendor set and the feedback set. A handful of frames for the planner, the florist and the venue, which is free marketing with your name attached, plus the honest notes for your second shooter about what worked.

All five come out of the same passes if you flag them as you go. Star the standout when it makes you pause. Label the vendor frame when the flowers are the subject. It costs seconds during the cull and saves hours after it.

From picks to gallery, and the loop back

The cull hands off to editing as a selection, not a new folder full of copies. Export the picks as flags or a collection your editing software understands, keep the originals where they are, and start editing the sneak peek set first so something reaches the couple this week.

Then answer the question every photographer eventually gets asked: what happened to the other 4,000? Nothing happened to them. They are still in the working folder, still in your backups, unpicked rather than gone. Say that plainly when a couple asks, because the honest answer is reassuring and the vague answer is not. If you want a policy rather than an instinct, we wrote one in what to do with the photos you do not pick.

Close the loop while it is fresh. Write down two numbers after every wedding: how many frames you shot, and how long the cull took. Six weddings in you will know whether your Tuesdays are getting worse because you are slowing down or because you are shooting 1,200 more frames than you did in May, and those are opposite problems with opposite fixes. Turnaround expectations, and what to promise in the first place, sit in the wedding photo delivery timeline.

Where a machine fits in the chain

Look again at pass one. Focus misses, blinks, blown frames, near identical bursts. That is pattern matching, and it is the one part of this workflow that software is genuinely good at. Pass two is taste, and taste is the job. The shape that works is simple: the machine proposes the technical pass, you approve or overrule it, and you keep every editorial decision that matters.

The context for why this is worth automating at all: a UK survey of 300 plus wedding photographers, run in 2020 before AI tools were common, found culling took 11 percent of total working time, against 55 percent for editing and 4 percent for actually shooting. Culling is also the gate in front of editing, so a slow cull quietly becomes a six week gallery.

Before you hand a real wedding to any tool, audit it. Run its free trial on a wedding you have already culled by hand, then compare its picks against yours: the panning shot it called blurry that you would deliver proudly, the technically perfect frame it kept that says nothing. Ten minutes of that tells you more than any feature list.

FAQ