My Photo Editing Workflow

Someone asked me recently about how I work through my photographs after I shoot. When I go out to photograph, I shoot a lot - most people I go out with are usually surprised at the number of frames I produce. It's not uncommon for me to take the kids to the park and come home with 150-300 images.

I find that I can usually edit down 350 images to around 50 in 40-60 minutes and I thought I would share how I go about this.


I have a Lexar FireWire CF card reader, which helps get that number of images into the machine quickly. In the past, I haven't always done much more than just import the images here, but I'm starting to make more use of Aperture's ability to apply certain metadata presets upon import, such as a standard copyright and credit notice. I don't apply keywords at import time.

I use referenced masters, so I'll have Aperture pull the master files from the card and place them in ~/Pictures/Aperture/Masters/YYYY/MM/DD/IMG_xxxx.CR2, which is where I keep them.

I only use Aperture's stacks feature if I've been doing a lot of high-speed continuous shooting. I don't stack images that are just conceptually similar - I use ratings to filter the good from the bad. I only use stacks when the frames were shot so close together that they're practically identical. If I've been shooting like that, I'll usually arrange the stacks before importing with a combination of the automatic stack slider and manual splitting/combining if the automatic stacking doesn't do the trick.

Let's assume for the rest of this article that I'm using stacks, since it brings another editing pass into play that I don't do if I don't have stacks.

Stack Sorting

The first thing I'll do is go through each stack and choose the pick for that stack. This is nothing at all to do with whether any of the images in the stack are any good. My aim in this pass through my shoot is to be able to close all the stacks, be certain that the pick of the stack is the best one in there and basically never open them again.

So, first thing I do is switch my second display into Alternate mode (View > Secondary Viewer > Alternate or Option-A). This way, I have one display showing the main Aperture window, and my other display shows whatever I'm looking at in a larger view. Next, I'll switch the secondary viewer into Stack mode (View > Main Viewer > Stack or Option-T). This causes the selected stack in the grid view to expand and the viewer shows the stack pick on the left and a second selected image on the right.

The question I'm asking myself in this phase is "Is this second image better than the pick?". If the answer is yes, I'll hit Cmd-\ to make the new image the pick. This pushes the old pick down into place #2 in the stack, moves the new pick to the top and selects the next unseen image. By going through a stack in this way, the good ones bubble up to the front of the stack.

I don't usually assign star ratings at this point, but I'll sometimes do one little trick: if I have a stack with a small proportion of good images - say the first 3 of 20 are good - I'll select image #4, hit #9 to reject it and then split the stack at image #4 (Stacks > Split Stack or Option-K). Then, when I select the good part of the stack, image #4 and the other rejects stacked with it will disappear from view.

If the Main Viewer is in Stack mode, pressing the down arrow will collapse the current stack, expand the next stack in the grid and select its pick and first alternate image. This skips any intermediate individual images, which is what I want since I'm just going through the stacks in this pass.

Having sorted my stacks, I'll collapse all the stacks (Stacks > Close All Stacks or Option-;) to make sure I didn't leave any open by mistake.


At this point, I've already cut down the shoot by some reasonable chunk just by collapsing all the stacks. Now I'm confident that whatever images are hidden in a stack don't need to be considered at all, just the picks. Remember that some stack picks might still be worthy of rejection - my stack sorting phase just determines the best image in the stack.

First thing I do is switch the Main Viewer back to Multi mode (View > Main Viewer > Multi or Option-U). Stack is a pretty specific viewing mode that's perfect for the first stack-sorting pass, but is exactly what we don't want for the image-rating passes since it would always be expanding and collapsing the stacks.

Now, I'll set Aperture's search field to show Unrated or Better (Ctrl-`). This hides any stacks that I already split and rejected above. Then I'll select the first image. The question I'm asking here is "Does this image deserve to be kept or rejected?".

I'll use Aperture's 'set rating and select next' keyboard shortcuts Ctrl-= and Ctrl-- (that's Ctrl-Minus). Since I'm starting with all my images unrated, Ctrl-= will set the rating to one star and select the next image. Ctrl-Minus will mark the image a reject and select the next one. The criteria I use in my head for star ratings is as follows:

RejectSomething photographically wrong (overexposed, missed focus, eyes shut)
OneTechnically OK
TwoMight be OK with some processing.
ThreeGood enough for Flickr
FourPick of the shoot
FiveOne of the all-time best

Once I've been through all the images and rated them One Star or Reject, I'll do another pass asking myself "Does this image deserve to be a Two Star or stay a One Star?". Having done that, I'll be looking at some images rated One Star and some Two Star. What I do next is have Aperture hide all the One Star images by choosing 'Two Stars or Better' (Ctrl-2) in the search field, then I do another pass deciding if the image is Three-Star or just Two-Star. As an example of how quickly this can happen, I have a 297-image project in front of me right now that I've edited down to a set of 38 Two-Star photos in less time than it took to write this blog post.

I usually stop these passes at the three-star rating. I always think that I need to let a set of images sit for a few days before I can really tell which are the Four-Stars. I take even longer to decide what gets a Five - maybe some weeks or months.

How do I decide what's Five Stars? Well, they're usually the images that I keep going back to just to look at them over and over again. They're the images that people ask me for prints of. They're the images that get hundreds of views and lots of faves on Flickr.

At this point, I hit the Update Vault button, kick off a Time Machine backup and go get a cup of coffee. I'll come back later and look at keywording (about which I'll blog later) and choose some images that I might want to post to Flickr using a very clever export plugin that someone made for Aperture.

[Update: Fixed incorrect shortcuts for increase/decrease rating and go to next. The shortcuts are Ctrl-= and Ctrl-- not Shift]