Post-production bottlenecks: find yours first
Almost every article on this subject hands you ten tips and lets you guess which one applies. That is backwards. A workflow has one bottleneck at a time, and speeding up anything else changes nothing you can feel. This is a diagnostic: how to find yours in one job, what the fix is for each, and how to tell when it has moved.
The slowest stage is not always the bottleneck
Two words get used interchangeably and they are not the same thing.
The slowest stage is the one that takes the most hours. For most photographers that is editing. A UK survey of more than 300 wedding photographers, reported by PetaPixel, put editing at 55% of working time against 11% for culling and 4% for actually shooting. That was 2020, before the current wave of AI tools, but the shape has not changed much.
The bottleneck is different. It is the stage that sets your delivery date. It is where work piles up in front and starves everything behind. It is the stage where, if you gave it one more hour, the gallery would go out a day earlier.
Sometimes they are the same stage. Often they are not. Editing can eat 55% of your hours and still not be your bottleneck, because editing runs at a steady rate once it starts and delivers as it goes. The bottleneck might be the two weeks the card sat there because you could not face the first pass. That stage took twenty minutes of labor and three weeks of calendar time.
The test
Do not ask which stage takes the longest. Ask: at any given moment, which stage has unfinished work waiting in front of it? That queue is your bottleneck. Everything else is just work.
One job, one page, five timestamps
You cannot find a queue by remembering. Memory reports the painful stage, not the constraining one. So take one real job, ideally the next one, and write down five moments. Nothing more elaborate than a note on your phone.
- Card out of the camera.
- Files on disk and backed up.
- First pass finished, picks marked.
- Editing finished.
- Gallery link sent.
Then write two numbers against each gap: hands-on hours, meaning time you were actually working, and elapsed days, meaning calendar time from one timestamp to the next. Those two columns are the entire diagnostic.
| What the gap looks like | What it means | Where to look |
|---|---|---|
| Hands-on hours high, elapsed days close to it | You are working flat out at this stage. It is genuinely slow | The work itself: technique, tools, batching |
| Hands-on hours low, elapsed days high | The work is not the problem. The starting is | Scheduling, dread, decisions, waiting on someone |
| Both low, but the next stage stalls anyway | You are handing off something incomplete | The join between the two stages |
The second row is the one nobody expects and the one that catches most people. A stage with twenty hands-on minutes and eleven elapsed days is not a speed problem. No faster software fixes it. Our piece on culling time benchmarks gives you numbers to compare your hands-on column against once you have it.
The five places it usually is
Once you have the two columns, one gap will be obviously worse than the rest. Match it against these.
1. Ingest and backup. Symptom: high elapsed days at the very start, because copying takes hours of wall-clock time and you keep putting off starting it. This one is genuinely solvable with hardware and a fixed routine rather than skill. If this is your gap, our post-shoot ingest checklist is the whole fix.
2. The first pass. Symptom: low hands-on hours, high elapsed days, and a familiar feeling of not wanting to open the folder. This is not laziness. It is thousands of near-identical judgment calls stacked back to back, which is a specific kind of tiring. We wrote about the mechanics of that in decision fatigue for photographers.
3. The handoff. Symptom: both stages look fine on paper but you keep redoing work, re-sorting in Lightroom, or finding ratings that did not arrive. Ratings and color labels travel between programs. Flags and collections do not, because they live only in the Lightroom catalog. If that sentence is news, it is very likely your gap, and exporting selections to Lightroom or Capture One has the mechanics.
4. Editing. Symptom: high hands-on hours and high elapsed days together, in proportion. This is the honest case where the work is simply the work. The lever here is almost never speed per photo, it is how many photos reach this stage. Cutting a gallery from 900 to 700 is a bigger win than any keyboard shortcut, which is the argument in how culling and editing fit together.
5. Delivery and the client. Symptom: everything is finished and the calendar still runs. Upload time, a gallery you rebuild by hand every job, a client who takes nine days to pick their album images. Some of this is yours to fix and some of it is a communication problem wearing a workflow costume.
Why fixing the wrong stage makes things feel worse
This is the part that justifies doing the measuring at all.
Suppose your real bottleneck is the first pass, sitting untouched for ten days. You read an article about editing speed, buy a preset pack, and get 20% faster at editing. What happens? Nothing changes for the client. The gallery still goes out on the same day, because it was never waiting on editing. You have spent money and gained nothing you can point to.
Now suppose you speed up a stage that sits before the bottleneck. Say you buy a faster card reader and ingest drops from ninety minutes to twenty. Now jobs arrive at the first pass faster than you clear them. The pile in front of the real bottleneck grows. You have not made anything worse in any absolute sense, but the workflow now feels worse, because the visible backlog is bigger and the finish date has not moved.
That is the whole reason the diagnostic comes first. Effort spent anywhere other than the constraint is either invisible or actively demoralizing. Effort spent on the constraint shows up in the delivery date immediately.
Fix one stage, in the smallest way that works
Once you know your gap, resist the urge to redesign everything. Pick the smallest change aimed squarely at that one stage and run it for three jobs.
| Your gap | The smallest useful change |
|---|---|
| Ingest and backup | One written routine, run in the same order every time, started before you take your shoes off |
| The first pass | Change when and where it happens, not how fast you click. Break it into two short sessions rather than one long one |
| The handoff | Record picks as star ratings or color labels, and set the XMP write and read settings once |
| Editing | Reduce the number of photos that get here. Then batch by lighting condition instead of shooting order |
| Delivery and client | A saved gallery template, and a date in the contract for client selections |
Three jobs, then re-measure the same five timestamps. Two things can happen. The gap closed and the bottleneck has moved somewhere else, which is success and also your next project. Or the gap did not close, which means you misdiagnosed and the queue was actually in front of a different stage. Both outcomes are useful. Neither is available if you skip the measuring.
The bottleneck moves, and that is the point
A workflow is not something you fix once. Every constraint you clear promotes a different stage to bottleneck, which is why photographers who solved culling three years ago now complain about client selections, and why photographers who solved delivery now complain about editing. Nothing went wrong. The queue moved.
Two things follow from that. First, the diagnostic is worth repeating once or twice a year, not once ever. Second, be skeptical of any advice, including ours, that names your bottleneck without asking you a question first. A tool that fixes a stage you were not stuck on is a cost, not an improvement, however good the tool is.
Season matters too. In peak season the bottleneck is usually capacity, not any single stage: the jobs arrive faster than you clear any of them, and everything queues. In quiet months the same workflow has no bottleneck at all and every stage looks fine. Measure in the busy weeks. That is when the constraint is visible, and it is the only time it costs you anything.
The bottleneck that is not a stage
Sometimes the two-column exercise produces a result that no software addresses. The hands-on hours are reasonable everywhere. The elapsed days are enormous everywhere. Nothing is queued in front of one stage, because everything is queued behind everything.
That is not a workflow problem. That is a volume problem, and it has three honest answers: take fewer jobs, charge more per job so you can take fewer, or hire someone. Buying software at that point buys you a slightly faster version of a schedule that does not work. Our guide to pricing photography is more use than any app in that situation.
Where tools genuinely do help is the second suspect on the list, the first pass, because that is the stage most likely to show low hours and high days. It is a stage that takes little time and a lot of willpower, which is exactly the combination that stalls. That is the stage Kepla is built for. It does the first pass on your iPhone or iPad, so a wedding can be shortlisted on the drive home instead of waiting for a free evening, and the same shoot opens on your Mac when you need to compare two frames properly. It never deletes, moves or renames a single file. It marks which photos it picked, and every frame it set aside is one tap away.
The picking app is still being built. The booking page is live and free today, and if your measuring points at the client end of the chain rather than the file end, that is where to start.
FAQ
What is the biggest bottleneck in photography post-production?
There is no single answer, which is why generic tip lists rarely help. Editing takes the most hours for most photographers, but the stage that sets your delivery date is often the first selection pass, because it takes little working time and tends to sit untouched for days. The only way to know which one is yours is to record hands-on hours and elapsed calendar days separately for one complete job.
How do I know which part of my workflow to fix first?
Track five timestamps on one job: card out of camera, files backed up, picks marked, editing finished, gallery sent. Then note two numbers for each gap, the hours you actually worked and the calendar days that passed. The gap where elapsed days are far higher than hands-on hours is your bottleneck. Fix that one and ignore everything else until it improves.
Why does my workflow feel slower after I sped part of it up?
Because you probably sped up a stage that sits before your real bottleneck. Work now arrives at the constraint faster than you clear it, so the visible backlog grows while the delivery date stays where it was. Nothing got worse in absolute terms, but the pile in front of the stuck stage is bigger, which feels worse. Speeding up the constraint itself is the only change that moves the finish date.
How much of a photographer's time goes on editing versus culling?
A UK survey of more than 300 wedding photographers, reported by PetaPixel in 2020, found editing took 55% of working time, admin 18%, culling 11%, communication 7% and shooting just 4%. Those figures predate the current generation of AI tools so treat them as a shape rather than a current benchmark, but the ranking has held up: post-production dwarfs everything else.
Is new software the answer to a slow post-production workflow?
Only if it targets the stage that is actually constraining you. Software that speeds up a stage you were not stuck on is a recurring cost with no visible benefit. Measure first. If the diagnostic shows long elapsed days at every stage rather than one, that is a volume problem instead, and the honest fixes are fewer jobs, higher prices or extra help.
How often should I check for bottlenecks?
Once or twice a year, and always during your busiest weeks. Every constraint you clear promotes a different stage to bottleneck, so the answer expires. In quiet months no stage looks like a problem because nothing is queuing anywhere, which makes the exercise useless. Measure when the work is stacked up, since that is the only time the constraint costs you anything.
Never sort photos 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.