I've ranted on about this topic ad nauseum on this and other stock photo forums since long before I tried my hand at photography. In fact, I first "got into" commercial photography precisely because I got tired of searching for stock photos that didn't exist. I had to take the pictures myself or hire college kids to take them.
I now know that the lack of the pictures I needed was a result of moderation by stock photo reviewers who apply archaic rules as if they came down from Sinai. Let's consider the history of these rules for a moment.
The rule of thirds (or as I like to call it, the "rule of turds") was first postulated by John Thomas Smith, and related specifically to the painting of rural landscapes. These paintings were created and purchased solely for their aesthetic value. Commercial print images were drawings or etchings, photography was still an experimental technology, and of course the Web did not exist.
So back then, there was no equivalent to today's stock market, where designers purchased existing images for commercial purposes. Rather, people commissioned and bought paintings for two reasons: The first was for the pictures' inherent aesthetic value (landscapes, flowers, pictures of religious significance, etc.); and the second was to memorialize people, as in commissioned portraiture.
Now consider this: When someone commissioned a painting of himself or herself, or a family or group portrait, they didn't expect the artist to apply the "rule of turds," the "golden ratio," or any of that other hogwash. They wanted the subjects of the portrait to be centered and to be the focus of attention.
Don't believe me? Try doing a Google image search for "greatest portrait paintings." Here, I'll even give you the link:
Go ahead. Search it. Now tell me, how many of those portraits do you think would get past DT's (or any other stock company's) reviewers? I'd say probably none.
What stock reviewers don't get and will never get is that when all is said and done, commercial designers only want two kinds of pictures, and for only one purpose. We want pictures of concepts, and pictures of things; and our only purpose for wanting them is to sell something. That may sound mercenary, but this is a business.
When we want pictures of concepts, then enforcing the rule of turds and so forth may be relevant or irrelevant, but usually it doesn't hurt the picture's value. We can always crop out the wasted space, and sometimes we actually want that space (for example, if we want to place text on it for a masthead graphic).
I should point out that a "concept" in this sense is a rather broad category. It includes, for example, almost all photos of non blue-collar people working. Office shots, coffee-drinking shots, shots of business people concentrating, and so forth, are all conceptual. The focus in not on the people. It's on the concept they're conveying.
For example, if I purchase a picture of half a dozen business people in suits concentrating about something, I am trying to convey the concept that my client has dedicated people who will work together to solve their clients' problems. No one cares about the people. They're models. The concept is what I need.
Now let's consider the other thing designers need: pictures of actual things. When I need a picture of a thing, then I want the focus to be on the thing. In that case, the classical rules often hurt an image's usefulness by reducing the quality of the part of the image that I actually need (you can only crop so much before quality loss becomes apparent), or by neglecting the compositional quality of the part that I need (due to poor lighting, shadows that run off the canvas, etc.) in favor of observing the rule of turds and the golden ratio.
For example, if I need a picture of a nut or bolt (there are companies do sell nuts and bolts, you know, and those companies do have Web sites), then I need a picture of a nut or a bolt, or a group of nuts and bolts. And I don't want them lined up along one of the grid lines. I want them centered. The nuts and bolts are what I'm selling. I'm not selling a concept. I'm selling nuts and bolts.
Now let's consider another whole universe that stock photo companies ignore: blue-collar service companies. There bazillions of pictures of people working in offices, but not a whole lot of pictures of people who actually get dirty at work. And guess what: Designers need those pictures.
In general, what we need for service business are pictures of problems and solutions. For example, if I need a picture for an auto repair shop, I need a picture of a mechanic fixing something on a car.
The "something" is the problem.
The "fixing" is the solution.
For shots like this, the mechanic (person) is not so important as the object, which is the thing being fixed. You don't go to the garage because you like the mechanic. What you like is the mechanic's skill at solving a problem. So I need pictures of problems and solutions, which in this example means pictures of a mechanic (not the focus) fixing something (that is the focus). That "something" may be a water pump, a shock absorber, a muffler, or whatever. But whatever it is, the ability of the shop to fix that thing is what I'm trying to convey.
Here's what I want: I want a picture of a mechanic's hands fixing that thing, with the thing being fixed the focal point of the picture, at or near the center of the canvas. I don't give a rat's about the rule of turds or the golden ratio. I am trying to sell a very specific service, not hanging the picture on my office wall.
Also, I may or may not want the model's face in the picture. For a chain shop, I might, because no one expects to see the actual guy in the picture fixing a car. But for a local shop, I only want the mechanic's hands. Now, as long as the compositional and technical quality of the hands fixing the problem are good, I don't mind cropping. But if you really care about my needs as a designer, offer me both shots. Offer the big picture of the mechanic, as well as the picture of his hands fixing the problem.
And while you're at it, offer me picture of skilled, hard-working hands in different colors. My client the mechanic may have white, black, brown, red, or yellow skin, you know.
The same general rules of problem -> solution go for almost any blue-collar service company's site. For a carpet cleaning company, I need pictures of things like food spills, dog puke, wine stains, baby pee and poop... those sorts of things. The pictures should convey frustration, fear ("This stain will never come out!"), and even disgust. I usually need the pictures centered, and I always need them detailed.
Remember: I'm not a curator at MoMA looking for pictures to put on exhibit. I'm a mercenary Web designer trying to sell something to the masses.
Oh, and another thing, while I'm at it: You may find this surprising, but designers really don't like isolated shots all that much. We buy a lot of them, but usually it's because that's the only way to get quality shots that focus on the objects we need, rather than wasting two-thirds of the canvas.
Consider this: isolated shots are very popular. They also completely ignore the rule of turds and the golden ratio. Those rules are irrelevant because the background is uniform, so the designer can simply match that background and enlarge or crop the canvas as he or she desires.
Or to put it more bluntly, the reason for the popularity of isolated shots is precisely because they ignore the 18th Century rules and give us what we want: the object as the focus of the picture.
Some day, maybe stock companies and reviewers will realize this; and maybe someday it will dawn on them that the very popularity of isolated images -- which ignore the archaic rules that they hold so sacred -- says something about the relevancy of the rules themselves in the modern world.
I'm not holding my breath.
Here's another little designer secret. Many times, when we buy isolated images, we convert the backgrounds to transparent and re-save them in some other format (usually PNG). Why? Because 99.9 percent of isolated shots are isolated against either FFF or 000, and we don't want our designs to be limited to white or black. Sometimes we want to use a different color, or a textured background, or to superimpose the subject over some other graphic.
If stock companies actually cared about what designers need, they would offer all isolated images in PNG format with transparent backgrounds. It would save us several tedious steps when we need transparency, and we also can assign any background color we like with a mouse click or two.
But it's a rare thing to find shots like that on stock sites. JPG doesn't even support transparency, but that's the only format that most stock sites will accept. So much for what designers need.
Now let's see... should I tell you the biggest designer secret of all? Okay, I will.
The biggest designer secret of all is that somewhere out there in Designer Land exists a designer sitting at a computer who needs a picture of literally anything that exists in this world.
You name it.
We need pictures of anything and everything. As long as a picture of practically anything is clear, sharp, and centered, one of us. somewhere out there, needs it.
But here's the problem: We probably won't go to DT or other stock photo sites looking for those pictures, because we know from experience that stock photo reviewers routinely reject pictures like that.
So what we do is search sites like Flickr or Picasa, where anyone can upload anything, without reviewers and their rules of turds getting in the way. And then we try to contact those photographers to purchase the rights directly. (At least, that's what those of us who are honest do. The dishonest designers simply steal the pictures.)
It costs a bit more to deal directly with amateurs, but that's a nice segue for yet another designer secret: We really don't care. The client's paying for the pictures, not the designer; and most amateurs will sell non-exclusive rights to their pictures for ten or twenty bucks, just because they're so thrilled that someone wants their photo. To both the designer and to the client, paying that ten or twenty bucks is a lot cheaper than spending endless hours searching microstock sites for shots that don't exist there.
You see, microstock's advantage for us is convenience, more so than low prices. But that advantage evaporates when the pictures we need just aren't there. I'd rather just pay an amateur ten bucks to use a picture, than waste hours of my billable time on a wild goose search of microstock sites.
Speaking of wasting time, I'm pretty sure that I just wasted a lot of it. No one will listen. For creative people, the degree to which stock photo reviewers worship a set of 18th Century rules borders on religion, and I don't expect to change that. But hey, it's Sunday, so what the heck.
Nikon D5100, Nikon L810