First things first, I should clarify that I'm a big fan of image search. I use it all the time. But lately I've begun to question the real value image search provides businesses. To be honest, I've never really thought in any sort of depth about the art of image optimisation. While I've certainly read my fair share of articles on the topic from respected authors such as Tad, Maki, Liana Evans, Ian Lurie & Chris Pearson, it's never been a traffic generation tactic I've employed. Because while there's little doubt that image optimisation has the potential to increase traffic, the real question is whether that traffic holds any commercial value.

Firstly, let's review how I use image search. In my full time role, I pull together a lot of PowerPoint presentations for client pitches. Joy! My presentation style is predominantly visual and I prefer to emphasise imagery rather than words. Hence the reason I love Google Images. It's a great place to find images that visually convey a core concept, minus the pesky watermarks that most image depositories utilise to protect their IP (legitimately I might add). My intentions when it comes to image search are quite clear – seek & steal.

It might not be the most ethical of intentions, but if I'm completely honest with myself, it's my sole purpose. I can't remember using image search for any other reason. I'm a drive-by visitor and I never pay any attention to the site I'm visiting (let alone bookmark it to return at a later time). The image is simply a commodity. Now… I don't necessarily expect my own behaviour to be indicative of how everyone else utilises image search, but I'm struggling to think of a lot of other practical uses for an image.

The difference between image search and keyword search

To get some context, let's compare it to traditional keyword search. The initial intent of a person running a standard Google search isn't markedly different – seek. They're looking for something. However, it is the secondary intention that is the important differentiator, and that is to inform. They are looking for information. Quality information that meets their specific informational needs. And that certainly isn't a commodity. We only need to look to the SEO industry to demonstrate the point. There's a reason Search Engine Land & SEO MOZ have tens of thousands of subscribers – because they distribute high quality information and expertise. Others try, but in most cases fail to replicate it. Conversely, how hard is it to replicate an image? Not very.

To illustrate the point further, let's think about the way users interact with keyword based search results. While Google's algorithm generally does a good job of qualifying results, it's still necessary to sift through the top 10 results. You may even run an alternative search query if none of the results match your criteria. This search behaviour is necessary to find the best quality of information. Once you find an appropriate source, it's possible (maybe even likely) that you'll bookmark the page, subscribe to the site or perhaps make a purchase. And it is here that I see the big difference. I just can't see see this behaviour being regularly applied to image searches, because an image is a commodity.

Does image quality prompt loyalty?

If you run a search for an image of a fire truck, how different can the results really be. Sure… some may be of a slightly higher quality. Others will provide different angles, models and perspectives. But once you find an appropriate image, do you exhibit the same behaviour as you do for a keyword search? Do you bookmark the site for later reference? Do you have any intent to return? Or do you simply take the image and move on with your journey? I know what I do.

Of course, there are exceptions to the rule. Artists and image libraries in particular rely upon unique imagery to sell their product. And you could certainly argue that there is potential for e-commerce retailers to attract qualified traffic via product images as well. There's no question ecommerce retailers could derive a commercial benefit by attracting people with an predetermined interest in a given product. But beyond that, I struggle to find practical applications for image search.

So tell me… am I being narrow minded? Or is image optimisation truly overrated?