The real cost of culling inside Lightroom

Lightroom does not send an invoice every time you stare at a grid of four thousand thumbnails. It bills you in evenings instead, which is harder to notice and harder to argue with. This is how to put a number on your own first pass, and how to work out what a second tool would have to save you before it is worth buying.

Lightroom never invoices you for the first pass

Start with the part that is not in dispute. Lightroom Classic is very good at deciding between photographs. The Library module compares, surveys, filters by metadata and keeps every choice attached to the work, and Adobe now ships assisted selection inside it at no extra charge, which we go through in our Assisted Culling review. On price, a feature included in a plan you already hold is very hard to beat.

The cost is somewhere else. It is the hours, the machine time, and the specific weight of putting every frame from every job through a catalog before you have decided whether any of them matter. None of that appears on a card statement, so it never gets compared against the $120 a year that a dedicated tool asks for. That comparison is the whole point of this page, and it only works with your numbers, not ours.

Two things this page is not. It is not an argument that Lightroom is bad at culling, because it is not. And it is not a settings guide: if your problem is that the application itself feels slow, that is usually fixable, and why Lightroom feels slow when you cull works through the causes one at a time.

The cost starts before you flag a single frame

A full card import asks your computer to register every image, build previews, write catalog data and make each file available for sorting. That is entirely reasonable when the images have a realistic chance of being edited. It is expensive when a third of the job is burst duplicates, test frames, misfires and shots that were visibly soft at capture.

Put rough numbers on it. A wedding is typically 2,000 to 4,000 RAW frames across an eight to ten hour day, and typically 400 to 800 of them get edited and delivered. Take a hypothetical 3,600 file job with an 800 image delivery. Those other 2,800 files still travel through import, preview generation, backup and review before anything rules them out. No single step costs much. The accumulated delay is what decides when you can start editing, and how responsive the catalog feels on a laptop that is also running a cloud backup.

Previews deserve their own line, because they are the clearest example of paying in advance for a decision you have not made. Standard previews let you move quickly. One to one previews make a focus check trustworthy, and they are also the ones you build for a whole wedding at 1am so you can spend two seconds each on half of them. Smart previews reduce storage pressure and let you work away from the drive, and they are worth understanding properly, but they do not reduce the number of decisions waiting for you. A more deliberate order helps: build working previews while backups finish, make the broad pass, then reserve one to one checks for portraits, formals and genuine close calls.

Six lines that give you your own number

You do not need a spreadsheet. Track three recent jobs from card backup to an edit ready selection and write down six figures for each.

• Capture volume. Original files across every camera, including a second shooter's.
• Catalog volume. Files Lightroom actually had to import, render and display.
• Candidate volume. Images that genuinely advanced to editing.
• Active time. Keyboard in hand, comparing and rating.
• Passive delay. Preview building, catalog lag, file transfers, waiting for a handoff.
• Correction time. Going back through a sequence because a frame was missed or the ratings stopped meaning anything.

The interesting part is the ratio, not the total. If active time is reasonable but passive delay is large, your problem is how files reach the catalog, and a different point of entry fixes it. If passive delay is small and active time is huge, no tool change will help much until your standard is written down, because you are re-deciding what a keeper is on every frame. That is a different job, and writing your criteria down is the cheapest fix in this whole article. Do this on three jobs rather than one, because a single wedding with a difficult reception will tell you almost nothing about your average week.

Turning hours into a number you can argue with

Now price the hours. The rate to use is not what a client pays for an hour of coverage. Use one of two honest figures: what you would pay someone else to do this work, or your own target rate for a non shooting hour. Pick one and write it down, because every decision below depends on it and a vague sense of value will not survive contact with a subscription page.

The arithmetic is deliberately simple. Jobs a year, times hours of first pass per job, times your rate. If the first pass takes you three hours on a wedding and you shoot twenty five of them, that is seventy five hours before a single photo has been edited. What those hours are worth is your figure to supply, and the total tends to be uncomfortable enough that the software question answers itself.

For scale on the other side of the ledger, The Knot's 2025 study of more than ten thousand US couples puts the average wedding photographer at $3,000 a booking, with quartiles at $1,500, $2,800 and $4,700. And it is worth checking that you are optimizing the right stage: a UK survey of 300 plus wedding photographers, reported by PetaPixel, put culling at 11 percent of working time against 55 percent for editing. That survey ran in 2020, before any of these AI features existed, so treat it as a shape rather than a current measurement. If editing is your 55 percent, fix that first. If your whole software stack is the question rather than one line of it, the cost calculator does that arithmetic.

What a second tool has to save to pay for itself

Here is the break even, using each vendor's own published price and a placeholder rate of $50 an hour. Substitute your own rate from the section above and the last column moves with it.

Prices checked August 2026 from each vendor's own pricing page.

Two honest readings of that table. The first is that the bar is low. If a tool saves you an hour on each of three weddings, it has paid for a year, and most working photographers clear that on volume alone. The second is that the bar is only low if you shoot enough. At four or five jobs a year, a subscription will not pay for itself, and the sensible answer is the feature that came free with the plan you already have. There is a longer version of that argument in do you still need a separate culling app.

Two costs get left out of this arithmetic and should not be. One is the handoff: moving files to a second application, then bringing the selection back, takes time that comes straight off the saving, and on a small portrait job it can cancel it entirely. The other is learning and maintenance, which is real in the first month and near zero after that. Neither is a reason to stay put, but a tool that saves twenty minutes and costs fifteen in transfers is not the win it looked like on the pricing page.

The cost that no timer records

Culling fatigue arrives before editing fatigue. By the time you have compared eighty near identical dance floor frames, the next genuine moment gets less attention than it deserves. Research on decision fatigue suggests repeated choices degrade the quality of later ones, and parts of that literature are debated, but the version photographers report is consistent enough to plan around: late in a session you make safer, more repetitive picks. A gallery chosen that way can be technically tidy and emotionally flat.

Context switching costs the same way. Every wait for a preview, a filter or a sync is either a stop or a half attention start on something else. Those minutes are fragmented, so they never show up in a time tracker, but they are the reason a first pass that should take one evening spreads across three.

This is the part of the bill that argues hardest for changing where the work happens rather than how fast you click. An hour of picking on a full battery is worth more than two hours at midnight, and the difference lands in the gallery the client sees.

The catalog you have to live with afterwards

A large catalog is not automatically a broken catalog, and nothing here is an argument for purging your archive. But a live working catalog holding every frame of every job from four years gets slower to back up, noisier to search and heavier every time another event lands in it. Photographers running one laptop catalog for everything feel this first.

The practical fix is separation rather than subtraction. Finish the project, verify delivery and two backups, then decide whether the originals you did not select belong in a cold archive rather than the daily working catalog. Keep the folder structure and enough metadata to find them again. Retention is a policy question with client obligations attached to it, and what to do with the frames you did not select covers it properly. What matters for cost is that current work stays responsive.

Whatever you change, do not let a faster route make file handling fragile. Two verified copies before anything else touches the card, no formatting until delivery, and the original capture folder treated as read only in spirit even when the operating system does not enforce it. Speed is only worth having when it does not put a wedding at risk.

Where the first pass actually belongs

The deciding question is not which application is better. It is where you trust yourself to make the first reduction, and how long it takes to get a job into a state where you can think about it at all.

If you love the compare and survey views, and your close calls turn on crop room or how a skin tone will recover, keep the final pass in Lightroom. That is where the information is. If getting a 4,000 frame job into a workable state is what pushes your evening past midnight, move the broad triage earlier, whether that is a fast browser, a dedicated culling app, or scene based collections you apply the assisted filters to one at a time. Do not change systems to own another app. Change the point of entry that is holding up the whole job.

That earlier point of entry is the bet we are making at Kepla. The photo picking app is still being built for iPhone, iPad and Mac, so it is not an option for your August, and we say so plainly in our comparison with Assisted Culling. The idea is a first pass in the gaps of the day rather than at the end of it, with the Mac for the frames that need a big screen, and nothing ever deleted, moved or renamed. Until it ships, the honest advice on this page is the arithmetic above: measure three jobs, price the hours, and buy only what clears the bar you calculated.

FAQ