The question of how to manipulate and store all the data acquired in a shoot is an extremely important one. There is a large number of steps involved in downloading, cataloging, processing and backing up the images, most of them rather repetitive. It really pays off to be disciplined and careful about the whole process, as you will ultimately gain time, will be able more easily to retrieve the good images from any session and will ensure that no image can be irretrievably lost.
Workflows are very personal and I doubt two photographers use the exact same process. They are also constantly evolving, as software or hardware gets upgraded and as new ideas are discovered, tested and eventually discarded or retained. So for some inspiration, here is my personal workflow, from start to finish, in the middle of 2010:
- I shoot with my D90 on 4GB SD cards (less capacity means less exposure to disastrous hardware failure). As soon as the card is full – or the shoot is over, I download all the images on my main portable hard drive, the 250GB Hyperdrive Colorspace UDMA, with all the safechecks activated.
- If this is only a dayshoot and I know that I will go back to a computer before the 12GB total capacity in cards is completely filled, then I simply store the card in the camera bag and switch to a new one. If I forecast needing to use the card again, I will download it first to a secondary hard drive, an older 160GB Hyperdrive Space. This way, I always have two copies of my images at all time, one on the main drive and one on either the card or the secondary drive.
- As soon as I get home, I plug the main drive and use Adobe Lightroom to import all the photos, choosing the option to copy the files to my main external 1To drive. At this stage, I have three copies and can afford to format the memory cards. I only use minimal keywording for location, relying instead on dates to find old images.
- When the import is finished and all the previews are generated, I go through all the files quickly in full screen mode. I use only three labels at that point: red (6), yellow (7) and reject (X).
- Rejects is for the obvious garbage only: out of focus, grossly over or underexposed, face hidden by a telephone pole, blinking eyes, etc. The only exception is for high-speed bursts of extremely similar photos, of which I only keep one or two copies. The percentage of rejects varies a lot, but it is usually pretty low for landscape and street work, and much higher whenever action is involved.
- The red label is for the best pictures which will demand my attention and need to be processed quickly. I try to limit how many red labels I apply, usually 5% or less on an average. Whenever several versions of a same scene seem about as good, I don’t try and make the final decision of which is the best one right away and tag them all with red.
- The yellow label is for photos which show some potential but don’t have as much impact as the red ones. I will often dig into yellow images once I’ve exhausted the red ones. They can represent 30 to 50% of a shoot.
- Finally, unlabeled images are just that: they are being spared the garbage bin but they show little potential and only when I grow quite desperate of finding an image will I go through them again.
- Once I am done with this initial pass, I will delete all the rejects (CTRL+Backspace) and, if needed, export high resolution jpgs of the red and yellow images to email to people who were with me on the day of the shoot.
- A few hours to a few weeks later, I will then revisit the catalog looking for an image to process. I usually decide on a day first, then activate the red filter or, if not enough is left, the yellow one. I will quickly review the different choices in fullscreen mode, then make my choice and go to the develop module.
- I always use the same sequence for processing inside lightroom: crop (if needed), select a camera profile, set white balance, set exposure, set vibrance, set clarity and finally fine tune colors and brightness if needed. At this point, I am either done with the image or I need to do some more work (localized adjustments, black and white conversion, etc), in which case I open the image in Photoshop.
- I work non-destructively in photoshop, never touching the background layer and using adjustment layers as much as possible. One I am done, I save the final .psd file and use an action to export a low-resolution jpg file (the steps of the action are: duplicate the image, flatten the layer, convert to 8bit mode, convert to sRGB colorspace, resize to 900×750, apply smart sharpening and finally save as jpg).
- Back in Lightroom, I now have a new .psd file stacked with the original .nef one. I put a blue (9) label on the original, meaning that it should be stored in the database but is not to be used for any other purpose than redoing the processing from scratch.
- I then put a green (8) label on the final image, either a .nef or a .psd depending on whether it received additional photoshop work and, if needed, export it as a low-resolution jpg for web use.
- Finally, every few months when I go abroad to visit either my family or a good friend of mine, I sync my external drive with their copy of my complete catalog, thus ensuring I have an off-site backup, if slightly out of date.
I do not doubt that there are many points that could be made more efficient, but this process is relatively simple, redundant and allows me to find the photos I am looking for fairly quickly. Don’t hesitate to comment and give me feedback, though!