Cloud storage for photographers: stop paying for chaos
Here's a familiar audit: a full Dropbox from 2022, a gallery host quietly storing every delivery, a drawer of orphaned hard drives, and a cloud archive you're not sure still syncs. You're paying for storage three or four times and you still couldn't produce the Hendersons' RAWs by Friday. The fix isn't another subscription. It's deciding, once, what lives where and why.
You have two kinds of files. Treat them differently.
Originals, RAW files and catalogs, are huge, touched rarely after delivery, and irreplaceable. They want cheap, deep, boring storage: written once, verified, and left alone. Deliverables, edited JPEGs, galleries, album spreads, are small by comparison, client-facing, and need to be fast and pretty. They want a gallery platform.
Most storage chaos comes from ignoring this split: paying premium hot-storage rates to keep five-year-old RAWs instantly accessible, or worse, treating a client gallery as the archive. Volume is why the split matters. A typical wedding day produces 2,000 to 4,000 RAW frames, with 400 to 800 edited images delivered (platform-published norms), meaning the overwhelming majority of what you're storing is frames nobody will ever see. Cull first, then archive; the whole system is in our RAW file management guide.
Three sane setups (and one trap)
| Architecture | Strengths | Weaknesses | Fits |
|---|---|---|---|
| All-cloud | No hardware, access anywhere, offsite by default | Upload speed is your bottleneck; costs scale with every terabyte, forever | Low-volume shooters, fast internet |
| NAS + cloud cold tier | Fast local working files; cheap deep archive offsite | Hardware to buy and maintain; two systems to manage | High-volume studios |
| Drives + one cloud archive | Simple, cheap to start | Manual discipline required; drives fail quietly | Newer photographers |
| "My gallery host is my backup" | Feels free | Stores only delivered JPEGs; files vanish with your subscription | Nobody. It's a trap |
The last row deserves its own paragraph, because it's the most common architecture in photography and it isn't one. Gallery platforms are delivery and print-sales channels: good ones, as we cover in Pixieset vs Pic-Time vs ShootProof, but they hold resized deliverables, not originals, on terms that end with your subscription. Use them for what they're for: making delivery day a marketing event, not preserving your life's work.
Whatever you pick, it must satisfy 3-2-1
Three copies of every job, on two different types of media, one offsite. Cloud storage is usually your offsite "1", not the whole system. If deleting one account or losing one drive could cost a client's wedding, you don't have backups, you have luck. The full protocol, from dual card slots to test-restores, is in the 3-2-1 backup workflow.
Five questions that expose the real price
- What does it cost to leave? Egress and retrieval fees are where "cheap per terabyte" quietly reverses. The exit price matters more than the entry price, because you'll retrieve in a crisis, not at leisure.
- How fast can I get files back? Cold tiers trade retrieval speed for price. That's a fine trade for a five-year archive; a terrible one for last month's wedding a client just asked about.
- Is this sync or backup? Sync mirrors your mistakes: delete locally, and the cloud copy follows. Real backup keeps versions and deleted files for a window. Know which one you're buying.
- What happens when I stop paying? Grace periods differ wildly. Ask before, not after.
- How does the price scale? Per-terabyte pricing grows smoothly with your archive; tiered plans hit cliffs. Model your growth: every year of shooting is another slab of terabytes that never goes away, unless your deletion policy says otherwise, which is a clause worth writing into your contract.
Notice there's no vendor shortlist here, prices and tiers in this market shift too often for a list to stay honest. The architecture and the five questions outlast any pricing page. When you do total up the subscriptions, storage belongs in the same audit as everything else: run it through the tool stack cost breakdown.
One practical constraint deserves its own line: your upload speed is part of the architecture. A beautiful all-cloud plan collapses if a wedding's worth of RAWs takes days to leave your office, and during that window, your only copies are local. Before committing to any cloud-heavy setup, time a real upload of a real shoot. If the number embarrasses your internet plan, either upgrade the connection or design around it: local-first with overnight scheduled uploads, or a smaller culled archive going up instead of everything.
The cheapest terabyte is the one you never store
Every question above gets easier when the pile is smaller. A disciplined cull before archiving routinely cuts stored volume by more than half. Those 2,000 to 4,000 frames per wedding become a few hundred selected photos plus a safety margin. That's less to upload, less to pay for monthly, and less to migrate when you inevitably switch vendors. It's also the part of the pipeline we're building Kepla around: an AI culling app (native for iPhone, iPad, and Mac: upcoming, waitlist open, free while we build) so the keep-or-kill decision happens fast, before the storage meter starts running.
FAQ
Is my client gallery host a backup?
No. Gallery hosts store the resized, delivered JPEGs, not your RAW originals, and most galleries expire or get deleted when you stop paying or hit plan limits. Treat gallery hosting as a delivery and sales channel, and keep your archive in storage you control.
Should photographers use a NAS or all-cloud storage?
It's a trade of upfront cost against ongoing fees and upload time. A NAS gives fast local access and one-time hardware cost but still needs an offsite copy; all-cloud removes hardware headaches but makes you dependent on your upload speed and a monthly bill that scales with terabytes. Many working photographers land on a hybrid: NAS for working files, cloud cold storage for the archive.
Should I cull before archiving to the cloud?
Yes. Storage is the second bill your rejects run up. A typical wedding produces 2,000 to 4,000 RAW frames while 400 to 800 edited images get delivered, so archiving everything means paying to store mostly rejected frames, forever. Sort first, archive the selected photos plus a defined safety margin, and put your deletion policy in your contract.
What questions should I ask a cloud storage vendor?
Five that expose the real cost: What does it cost to get my files back out (egress and retrieval fees)? How fast is retrieval from cold tiers? Is this sync or true backup with versioning? What happens to my files if I stop paying? And does the price scale per terabyte or jump in tiers? The exit cost matters more than the entry price.
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.