|The perfect stock photo for this article|
Everyone involved in the creation of a website understands this on some fundamental level because we (as the creators of a site) know where the images come from and know that the images we choose are just a paltry handful of a thousand others that represent or show the same thing. But do our end-users notice this? And, if they do, does it have any kind of effect on their ability to use or engage with a site? Of course it does! I wouldn't have asked those questions if the answer wasn't a solid "yes".
People don't engage as much or as well with a site when the imagery doesn't line-up with the content or their expectations. To use a stupid example, it would be inconceivable to think that a site for a pet store would do as well if all of the imagery involved anything but fluffy, cute animals. If I replaced every puppy photo on the site to show a 1970 AMC Gremlin, the site's usefulness would come to a screeching halt and engagement would flat-line.
Of course no web designer worth their salt would ever disassociate the content and imagery to that extreme, but when it comes to real world scenarios it's not the extreme examples that are the problem; it's the subtle and often overlooked commonalities that people pick up on. Is that businessman on the home page a member of the organization? Is that lady with the headset really a customer service representative? Is that building shot tied in any way shape or form with the company itself? In spite of what some people may say, the end-user definitely picks up on these common threads and subtle incongruities when they look at a website, even if they can't fully articulate what the problem is.
One of the biggest issues with stock photography is that people just "know" when something isn't right. This is especially true when it comes to shots of people. Ever notice how the lighting in a photograph is just a little too clean or how the subjects' teeth are a little too white? Is the the angle of the shot just a bit off-center but not by so much that it would ruin the composition? These are all subtle signifiers that scream "fake" to users.
I won't get too far into the details as Jakob Nielsen over at UseIt.com did a more than suitable job of articulating this point 2-years ago in his study of photo content.
In Nielsen's own test, he found that when people were presented with biographies with both image and text content, users spent 10% more time examining employee portraits than they spent on reading biographical text. This is in spite of the fact that in terms of volume and space used, the biographical text was the predominant element. By comparison, generic stock photography was largely ignored to the point that it's placement and subject matter was largely irrelevant. Suffice it to say that if it isn't "real", it's not meaningful, and if you're a business that takes their web presence seriously, then it should concern you if large sections of your website are being wasted or ignored.
Blow it up, start over, and really think about the real estate that's being wasted on your site. In an age when illustrators, vector artists, and freelance photographers are plentiful, there's no rhyme or reason to saturate otherwise good web content with generic, ineffective stock photography.
Even in instances where budget is a concern and designers have to bite the bullet, there's absolutely no excuse for not putting in the time and effort to find stock imagery that "works". Whether it's because a particular image hasn't already been purchased by a thousand other people, or because it provides a better thematic fit, being able to deliver a final showpiece with imagery that works speaks volumes about the time and care that went into the initial build of a site.