Photographer Storage Math: What a Year of Shoots Costs

The assumptions behind the storage math

None of these figures are industry averages. They are round numbers chosen so the arithmetic is easy to redo with your own camera and workflow. Change any row and the totals follow.

Sizes use decimal units (1 GB = 1,000 MB, 1 TB = 1,000 GB), which is how drive makers label capacity. Your camera's real RAW size depends on sensor resolution and compression, so check a recent folder before trusting the 30 MB.

What one shoot weighs

Multiply each row out and add them up:

The things a client actually looks at, the delivered JPEGs and the review proxies, are under 3% of the footprint. Storage planning is almost entirely a question about RAW files and layered masters. The 0.1% proxy slice is also why a review gallery is a cheap thing to keep online; the expensive thing is the archive behind it. We covered why clients get proxies rather than RAW separately.

A year, five years, and three copies

Scale the per-shoot figure: 54.575 GB × 40 shoots = 2,183 GB, or about 2.2 TB a year. Archives are cumulative, so the five-year figure is 2,183 GB × 5 = 10,915 GB, about 10.9 TB.

The three-copy line comes from the 3-2-1 rule: 10,915 GB × 3 = 32,745 GB. If you follow the approach in a backup strategy that survives a stolen laptop, that third number is the one you are buying hardware or cloud capacity against, not the first.

Where trimming actually pays

Three levers exist, and they are not equal.

1. Cull rejects out of the archive

Suppose you keep RAW files only for the 150 delivered frames and discard the other 1,350. RAW per job becomes 150 × 30 MB = 4.5 GB. The per-shoot total becomes 4.5 + 8 + 1.5 + 0.075 = 14.075 GB, a drop of (54.575 − 14.075) ÷ 54.575 = 74.2%. Forty shoots then come to 563 GB a year instead of 2,183 GB.

This is a business decision, not a free win. Some photographers keep everything because a client occasionally returns asking for a frame that was not selected, and your contract and clients may expect that. Decide it deliberately, and write the retention period into your terms.

2. Prune layered masters

Masters are 14.7% of the footprint. Flattening or deleting intermediate versions once a round is signed off helps, but only if you are sure no later round will need the layers. Version history is useful precisely because it is kept; see version stacks versus folder names for how to keep it navigable.

3. Compress deliverables and proxies

Delivered JPEGs plus proxies are 2.8% of the total. Even a 50% reduction there saves about 1.4% overall. Compression is worth tuning for the client's download speed, not for your storage bill.

Retention tiers that fit the numbers

Rather than one rule for everything, tier the archive by how likely you are to need each class of file again.

Gallery expiry is the cheapest tier to automate, since proxies are small. The reasoning on how long to leave a link live is in how long to keep client galleries online.

Turn the math into a number you can budget

Take the one-copy yearly figure, multiply by years of retention, then by your copy count. For this example: 2.183 TB × 5 years × 3 copies = 32.7 TB. Compare that with the same sum after a rejects policy: 0.563 TB × 5 × 3 = 8.4 TB.

Then price it. We have left dollar figures out on purpose because drive and cloud prices change often; take the current per-terabyte price from whichever vendor you use and multiply. The point is that the multiplier you control, how many frames and versions you keep, is worth far more than shopping for a slightly cheaper terabyte.

Finally, re-measure once a year. Pull the actual size of your last ten shoot folders, divide by ten, and replace the 54.575 GB in this post with your own number. A new camera body with higher-resolution RAW files can change that figure by a lot, and you will want to know before the drives fill up.

How the numbers shift for different kinds of work

The 40-shoot studio above is one shape of business. The same arithmetic behaves very differently for others, and it is worth knowing which end you sit at before buying anything.

A high-volume event photographer might shoot 3,000 frames per job and deliver 600. With the same 30 MB RAW assumption, RAW per job is 3,000 × 30 MB = 90 GB, and the delivered JPEGs are 600 × 10 MB = 6 GB. Layered masters barely matter because few frames are retouched individually. Storage is nearly all RAW, so a rejects policy is the only lever that counts.

A commercial retoucher-heavy studio looks the opposite way. Suppose 200 frames are shot, 60 are retouched at 300 MB per layered master, and 60 JPEGs are delivered. RAW is 200 × 30 MB = 6 GB, masters are 60 × 300 MB = 18 GB, and JPEGs are 60 × 10 MB = 0.6 GB. Here the masters are about 73% of the total (18 ÷ 24.6), and pruning intermediate versions matters far more than culling RAW.

The lesson is not that one profile is right. It is that the first step is always to split your own folder sizes into RAW, masters and deliverables, because the biggest slice tells you where to spend effort.

Common mistakes when planning archive capacity

A few errors recur whenever people size storage by feel instead of by sum.

• Buying to the one-copy number. A 4 TB drive sounds generous against 2.2 TB a year until you remember the second and third copies need their own capacity.
• Forgetting versions. Each retouch round can save a new layered file. If you keep three rounds of 40 masters at 200 MB, that is 3 × 40 × 200 MB = 24 GB per shoot, not the 8 GB in the table above, which assumed one master per frame. Decide whether you keep every round or only the signed-off one.
• Mixing working storage with archive. The drive you edit from needs speed and headroom, not five years of history. Treat the active-jobs set (say three months of shoots, 10 × 54.575 GB = about 546 GB) separately from the long-term archive.
• Ignoring camera upgrades. If a new body raises average RAW size from 30 MB to 60 MB, the RAW slice doubles from 45 GB to 90 GB per shoot and the yearly total grows by 1,800 GB, to about 4 TB.
• Treating the review gallery as the storage problem. It is the part clients see, but at 75 MB per job it is rounding error next to the archive.

One more habit helps: record the date and the folder size when a job is archived. After a year you will have 40 data points, and the average of them is a better planning input than any figure from a camera spec sheet or from this post. It also shows you whether jobs are getting heavier, which is usually the first sign that a retention rule needs revisiting. If one client type consistently produces folders twice the average, price its storage into the quote or shorten its retention window.

None of these needs a spreadsheet more elaborate than the one in this post. Write the inputs in a notes file, redo the sums each January, and adjust retention if the total has outrun your budget.