
You get home at eleven, you are running on catering sandwiches, and the entire wedding exists in one place: a few small pieces of plastic that have spent the day in a bag. What happens in the next hour decides whether the rest of the job is work or panic. This is that hour written out as a routine, so you can run it tired.
Between your last frame and the moment a second copy exists, every photo you shot is stored exactly once. On a card. That has been in a bag, in a car, in the rain, and is now in a coat pocket somewhere. Nothing else in your workflow has that property. A slow edit is annoying. A late gallery is embarrassing. This is the only part of the job where the failure is permanent, and it lasts about an hour.
So the hour has one goal, and it is not a creative one: get from one copy to two, with real names on the files and a note about what happened. That is it. It is not culling and it is not editing. A UK survey of more than 300 wedding photographers found that editing takes 55 percent of a photographer's working time and culling another 11 percent, which tells you those are proper jobs deserving their own planned sessions, not something you start at midnight because the files are right there.
This article is the routine. The reference material behind it, how many drives, what to archive, how long to keep things, lives in our RAW file management guide. Read that once, on a quiet weekday. Run this every time you get home.
Before the coat comes off. Every card out of every body and into one place, a single pouch where shot cards face one way and empty cards face the other, so a glance tells you the state of all of them. Card reader plugged into the computer, not the camera, because a camera on a dying battery is a bad thing to be reading from. Drives awake and mounted.
The point of doing this identically every single time is that you are about to make decisions while tired, and a routine you can run on autopilot is the only kind that survives a fifteen hour day. Photographers who lose cards almost never lose them at the venue. They lose them between the car and the desk, on the one night the routine got skipped.
Do not format anything. Do not open Lightroom. Do not look at the frame you know you got during the vows. That single peek is the start of a two hour detour that ends with you having viewed 400 photos, chosen nothing, and still having only one copy of the shoot.
Copy the card to your primary drive. Copy, not move, and not any option described as import and eject. A move reads and writes at the same time and then clears the source, so an interruption halfway leaves you with a partial copy and no reliable way to tell where it stopped. Copying leaves the card untouched, which is the entire safety net for the next few hours.
Then verify, at whichever level your tools support. The cheap version is a frame count: the camera told you how many frames were on the card, and the folder should hold exactly that many. The real version is checksum verification, where the software reads back what it wrote and compares it byte for byte against the source. That is the difference between believing the copy worked and knowing it did, and it is the main reason working photographers pay for a dedicated ingest tool.
| Tool | What it does in this hour | Price |
|---|---|---|
| Photo Mechanic | Verified copy, renaming and metadata templates applied on the way in. Famous for speed. | $14.99 per month, $149 per year, or $299 once |
| Photo Mechanic Plus | The same, plus a searchable catalog of everything you have ever ingested | $24.99 per month, $249 per year, or $399 once |
| Lightroom import | Copies, renames and applies metadata presets in one step. No checksum verification. | Included with an Adobe photography plan |
| FastRawViewer | Fast RAW browsing for a quick sanity check that the frames are there and sharp | $23.99 once, two computers |
| Finder or File Explorer | Copies files. That is all. Still far better than not copying. | Free |
Prices checked August 2026 from each vendor's own pricing page.
A wedding day's 2,000 to 4,000 RAW frames take real time to move, so start the copy and spend the wait on the next step rather than watching a progress bar fill.
Here is the rule that makes this hour worth having: the card is not a backup, it is the original, and one original plus one copy is the minimum acceptable state before you are allowed to feel relaxed. Two copies on the same drive is one copy wearing a disguise. Two drives sitting on the same desk is two copies with one flood, one power surge and one burglary between them and nothing.
Tonight's requirement is small and specific: a second copy on a different physical device, finished, before the cards go back in the bag. That is the whole ask for the hour. The offsite copy matters just as much but it does not have to complete before you sleep, so start the cloud or network upload now and let it run overnight while you are not watching it.
What that second device should be, how many generations you keep, when a drive gets retired and how the offsite copy actually works are policy questions rather than tonight questions. We work through them properly in our photography backup workflow, and the offsite half specifically in cloud storage for photographers. Decide those once, then this step is just pressing go.
If your ingest tool renamed the files during the copy, this step already happened and you can use the ten minutes to eat something. If it did not, do it now, while the files have still never been opened by anything else.
Names go on in the format that sorts correctly on its own: the date year first, then the client, then the shoot type, then a zero padded number, so a frame reads 20260314_alvarez_wedding_0001.cr3. The argument for that exact shape, and the eight small habits that quietly break it, are laid out in our guide to building a photo file naming system. Then the files land in a job folder built to the same pattern, with numbered stage folders inside it so raw, picks, edits and delivered sort in the order you actually work rather than alphabetically. That structure is covered in our photo folder structure guide.
The reason this belongs in the first hour rather than on the vague later list is mechanical. Once Lightroom has a catalog pointing at these files, their names and paths are recorded in a database, and renaming them from outside gives you a catalog full of missing images and an evening spent reconnecting folders. Right now, nothing is pointing at them. It is the last cheap moment.
Copyright notice, your name, your email and your website go into every single file, applied as a template during or straight after ingest. It takes no time at all when it is applied to a whole card at once, and it is a genuinely miserable job to retrofit onto four years of archive later. Set the template once and it is free forever after.
On top of that, add the handful of facts that describe the whole job and that you will absolutely not remember in March: the client, the venue, the city, the shoot type. If a second shooter's frames are in this job, put their name in the creator field on their files now, while you can still tell whose card was whose. There is more on which fields actually earn their keep in our photo metadata guide, and on why the descriptive ones matter for anything you publish in metadata and photo SEO.
What does not belong in these ten minutes is per image keywording. That is work you do to a finished set of a few hundred pictures, not to four thousand RAW frames of which most will never leave the drive. Job level facts now, image level detail much later, if at all.
Open a plain text file in the job's admin folder and spend five minutes writing down what tonight knows and next Thursday will not. How many cards, how many frames, and whether those numbers matched. Who shot what. The deadline you promised out loud at the reception. The specific things the couple asked for, the grandmother in the wheelchair, the shot of the ring against her mother's dress. And the problems: the flash misfired through most of the second set, the speeches were badly underexposed, the light in the barn went strange after nine.
That file is the single highest return five minutes in this entire hour. It turns your next session from an archaeology project into a task with instructions.
Then the cards go back in the pouch, shot side up, and they stay unformatted until two things are both true: the second copy has been verified, and the gallery has actually been delivered. When that day comes, format in the camera rather than on the computer. Until then a full card is the cheapest insurance you will ever own.
And then stop. The files are safe, named, described and documented, which was the whole job. Picking your best shots is a different task with a different kind of attention, and doing it at midnight on adrenaline produces selections you will disagree with on Tuesday. Book it as its own session instead, the way we lay out in our wedding culling workflow.
The one honest exception is the pass that happens before you ever get home. That is the problem Kepla is being built for: your best shots picked on your iPhone or iPad on the drive back, with the same shoot opening on your Mac when you need a bigger screen, and every original left exactly where it is. The picking app is still in development. The booking page is live and free today.
Print this, tape it above the desk, and stop making decisions at midnight.
| Minutes | Do this | Done when |
|---|---|---|
| 0 to 5 | Every card into one pouch. Reader and drives connected. Nothing opened. | All cards accounted for, none formatted |
| 5 to 20 | Copy each card to the primary drive. Copy, never move. Verify. | Frame counts match the camera and verification passed |
| 20 to 30 | Second copy to a different physical drive. Start the offsite upload. | Two copies exist on two separate devices |
| 30 to 40 | Rename to your convention. File into the job folder and its stage folders. | One named job folder holds everything |
| 40 to 50 | Apply copyright and contact to every file. Add job level keywords. | No file in the job is missing your name |
| 50 to 60 | Write the job note. Cards back in the pouch, unformatted. | The note answers next Thursday's questions |
| After | Nothing. Picking is a separate, planned session. | You went to bed |
If you are too tired to run the whole thing, run three of them: the copy, the second copy, and the note. Those three are the ones where skipping costs you something you cannot buy back. The rest is tidying, and tidying keeps until Tuesday.
Copy every card to your main drive, verify the copy, then make a second copy on a different physical drive before the cards go back in the bag. After that, rename the files, apply your copyright and job level metadata, and write a short note about the shoot. Leave the cards unformatted. Do not start picking or editing that night.
Always copy. A move operation reads and writes at the same time and then clears the card, so if it is interrupted you are left with an incomplete copy and no simple way to know where it stopped. Copying leaves the original on the card untouched, which is your safety net until a verified second copy exists somewhere else.
When two things are both true: the second copy has been verified, and the gallery has been delivered to the client. Until then, a full card is a free extra copy of the whole job. When that day arrives, format in the camera rather than on the computer, because the camera writes the file structure it expects to read.
Plan for about an hour for a full wedding day, most of which is the computer copying rather than you working. A day producing 2,000 to 4,000 RAW frames is a lot of data to move twice. Your hands-on time is closer to fifteen minutes if your ingest tool handles renaming and metadata templates during the copy.
Usually not. Picking your best shots asks for a kind of judgment that a fifteen hour day has already spent, and selections made at midnight tend to get reversed later. The first hour after a shoot is for making the files safe, named and documented. Book the picking session separately, when you are rested and can work through it in one clean pass.
No, but it helps. Its advantage in this hour is checksum verified copying plus renaming and metadata templates applied in the same pass, which turns three steps into one. Lightroom's import panel does the renaming and metadata but not the verification, and a plain Finder copy does neither. Any of them beats leaving the files on the card.
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.