Images and content are marketing’s bosom buddies – they go together perfectly in nearly every platform, helping to emphasize one another and underline brand messages. That’s great for the consumer, but what about on the backend, when it comes to Google’s indexing practices? Can Google image index, for example, perform stunts like reading the text overlaying an image, or recognize the content of an image? In short, yes – they have that capability and they’re already using it in many corners of the web. The real question in play, however, is how you can use that to your advantage in your own campaigns.
For companies still working on the way the present images, there’s some good news: the practice of treating images exactly like content hasn’t happened….yet. While some industry insiders have wondered “Does Google index text in images?” the answer is that it’s on their radar but isn’t implemented across the board yet. For the purposes of copyright and intellectual property, it seems inevitable that they’ll pursue this indexing method in time, if only out of protective reflex. Copyrighted text can be hidden against a background image, for example, and that can lead to a lot of legal headaches and takedown notices.
Google already uses web searchers to filter, rank, and refine their results – good traffic and low bounce rates lead to higher placement. For images, however, Google is already crowd-sourcing their visual recognition refinement through a unique method: captchas. Using a security feature that’s necessary for everything from form inputs to banking logins, the search engine giant is tapping into that login momentum to figure out what houses, stores, and even street signs in Google maps look like. It’s not hard to guess where the progression is headed: first, it will be yes/no questions that determine if text shows up as an image overlay, then actual transcription of image-bound letters and words, all courtesy of a global workforce that works for free. Google isn’t the biggest name in the game for nothing.
To over-simplify it, if you’re doing anything you shouldn’t be with images right now, even as placeholders, phase it out. Read the writing on the wall and realize that original, text-free images with robust metadata will serve you well as the updates to visual recognition continue to roll out. Metadata may not be indexed, necessarily, but its presence helps your site in other areas, such as accessibility for the visually impaired. Additionally, follow best practices when it comes to your images – don’t overlay text (this practice will hurt you in Facebook ad buys, anyway) and keep your image sizes and resolutions as consistent as you can throughout your brand presence.
While it’s possible and even probably that Google will eventually read your images as easily as your text-based content, it’s always better to clear an easy path for climbing up in the search engine results pages. The way things like keyword density and content structure affect your standings is already a known factor: so work on optimizing it, rather than worrying about potential upcoming changes. Text in images is often a shortcut, so if you’re currently using text overlay as a method, consider why you’re doing it – why is it necessary to your workflow? Google-read image text may or may not pick up on it in time, so go for the sure thing instead: your metrics will reflect this as a smart choice.