The File Format That Promises Everything and Delivers Nothing on Its Own

I shot my first newspaper assignment on a Canon 10D and filed JPEGs straight from the camera. The photo editor never complained. The images ran. That was the job. When I moved into commercial work and started hearing photographers talk about RAW like it was the Holy Grail, I understood the appeal but also saw what nobody was saying out loud: a RAW file is a liability until you process it. It’s an unfinished product sitting on your drive, eating storage, and going nowhere until you build a real system around it.

If you’re shooting RAW without a defined processing workflow, you’re not being professional. You’re being slow.

What a RAW File Actually Is (and Why It Can’t Be Ignored)

A RAW file is sensor data. Uncompressed, uninterpreted, proprietary to your camera manufacturer. A Canon CR3, a Nikon NEF, a Sony ARW, these are not images yet. They’re instructions waiting for a renderer to make decisions about white balance, tone curves, noise reduction, and sharpening. Your camera’s JPEG engine makes those decisions automatically. When you shoot RAW, you take on that job yourself.

The upside is meaningful. You’re working with 12 or 14 bits of data per channel instead of 8. That’s the difference between 4,096 tonal values per channel and 256. In practical terms, it means you can pull two to three stops of detail out of a blown highlight or a crushed shadow without the image falling apart. For commercial work, where a client might need a sky swapped or a product detail held in difficult lighting, that latitude is not optional. It’s what you’re selling.

The downside is equally real. RAW files are two to four times larger than JPEGs from the same camera. A single 45-megapixel Sony A7R V RAW file runs about 83MB uncompressed. A full-day commercial shoot at 500 frames becomes roughly 40GB before you’ve touched a single slider.

The Actual Processing Workflow: From Import to Delivery

My import process starts before I even leave a shoot location. I use a dual-card setup on every camera, writing simultaneously to two cards. At the end of a session, one card stays in my bag as a field backup until I’ve confirmed a successful import and two off-site backup copies exist. That’s not paranoia. That’s the standard I work to.

For processing, I run Adobe Lightroom Classic on a Mac Studio with 64GB of RAM. Lightroom Classic, not Lightroom cloud, because Classic keeps your catalog local and gives you full control over file structure. I organize everything by client, then by date: Client Name / YYYY-MM-DD - Project Description. Every shoot gets a unique folder, no exceptions.

My default import preset applies a baseline correction to every file: lens profile enabled, Chromatic Aberration set to remove automatically, sharpening at 70 / radius 1.0 / detail 25 / masking 50, and luminance noise reduction at 25. That preset alone eliminates about 20 minutes of repetitive clicking on a 300-frame shoot. From there I cull in Lightroom using the X key to flag rejects and the P key for picks. I target a 25 to 30 percent selection rate for initial delivery. For a 400-frame shoot, that’s 100 to 120 files going into the editing queue.

Color grading starts with white balance. I shoot a WhiBal card at the top of every lighting setup change. In Lightroom, I click that card with the white balance selector and sync that setting across the batch. Then I adjust exposure globally, set my black point by holding Option while dragging the Blacks slider until clipping just appears, and set the white point the same way. After that, everything is image-specific.

For commercial deliverables, I export at full resolution, Adobe RGB color space, 16-bit TIFF for retouching handoffs, and sRGB JPEG at quality 90 for web and client proofing. I use a second preset in Lightroom for each of those export scenarios. The whole export process takes two clicks and whatever time the machine needs to render.

When the System Saves You From Yourself

My twins are eleven. A while back, one of them was using my office computer to look something up and managed to drag a folder of client proofs into the trash and empty it. We’re talking 340 finished JPEGs. I keep a locally attached 8TB working drive, a second 8TB drive that mirrors it hourly via Carbon Copy Cloner, and a third drive that runs a nightly backup. I had those files restored from the mirror drive in about 90 seconds. My kid’s panic lasted longer than the data loss.

The point isn’t that I’m clever. The point is that the system I built around my workflow, not the RAW files themselves, is what protected the deliverables. File format doesn’t save you. Process does.

Choosing a Raw Processor That Matches Your Volume and Output

Lightroom Classic is my primary tool, but it’s not the only answer. Capture One Pro handles color, especially for Fujifilm and Phase One files, with more precision, and its tethering is more stable than Lightroom’s for studio work. The subscription runs about $24 a month or $299 per year. For photographers doing heavy volume work with consistent studio lighting, the color fidelity is worth the cost difference. DxO PhotoLab 7 is worth a serious look for high-ISO work. Its PRIME noise reduction is genuinely better than Lightroom’s for files shot at 6400 ISO and above, though the workflow is less fluid for large batches.

The tool matters less than using it consistently, learning it completely, and building presets and export settings that remove repetitive decisions from your day.

The Real Deliverable Isn’t the RAW File

Every RAW file you shoot is a bet that your processing workflow is good enough to cash in on the data your sensor captured. The photographers who build real businesses are not the ones with the best files. They’re the ones whose clients receive clean, consistent, correctly profiled images on deadline, every time. That is the product. Build the system that produces it, and the file format takes care of itself.