Stocksy’s proprietary Visually Similar (VS) and Conceptually Similar (CS) search tools come from a desire to create more direct pathways for discovering high-quality, useful stock media. If you’re hunting for content that meets Stocksy’s standards with specific production value, tone, quality, palettes, models, or environment in mind, and you’re also a fan of saving time and energy, these AI-powered modern tools are about to be your new best friend. Learn more about VS and CS and start scouring your favorite blogs and brands for inspirational photos to upload for inspiring results.
Visually Similar
Sometimes, when you are developing a brand campaign, you see the exact photo you need to run your social media ads. It’s the right color, the right lighting, the right subject matter, it’s perfect. But your perfect campaign picture is a photo from the internet and is definitely not for sale.
Here’s the good news: if you can save a copy of the image, you may still be able to uncover the photo of your dreams. Have you heard of reverse image search? Visually Similar is a proprietary Stocksy tool like reverse image search that allows you to upload any image you collect while scouring the internet to then return visually similar images and videos available for licensing.
Tools like Stocky’s Visually Similar solve your problem by using advanced machine learning to find stock photos that look like your reference image in seconds. You can even use images from other stock libraries that are close but not quite hitting the mark to upload to Stocksy’s Visual Search and find the stock photo or footage that ties your project together.

To try Visual Search, simply go to stocksy.com, click on the camera plus icon in the search bar and drop your reference image into the pop-up drop zone to surface 200+ results. That’s it that’s all. Watch the options unfold before your eyes for tonnes of campaign-ready images.
Conceptually Similar
If you’ve tried Visually Similar, you must try Conceptually Similar — they are not the same thing!
Often when developing brand campaigns, the tone or idea behind a visual asset is the most important thing, but finding the right stock media asset to convey your concept with a reverse image search tool can sometimes miss the mark.
While Visually Similar returns stock media that looks like your reference image, Conceptually Similar can detect the mood, vibe, or feel of your reference image and offer alternative options that fit your concept.
For example, if your campaign is about seasonal depression disorder and you upload a reference image of a person wearing black tones feeling the SAD blues, it’s not especially helpful if your visually similar results are of people dressed in black having a great time at a party. The angle of the image may be just what your digital ads call for, but the tone is way off. Stocksy’s advanced machine learning Conceptually Similar tool solves for this vibe shift by detecting the concepts and overall themes rather than a literal visual interpretation.


If we begin with this input of an anonymous woman posing with flowers in cool studio lighting…




If you’ve found a stock photo with the right visual elements but doesn’t quite hot the mark for your message, Conceptually Similar offers a more nuanced take on the atmosphere and subject matter in an image. Get endless inspiration and ideas from Conceptually Similar by scrolling down on any asset details page to see what Stocksy’s AI surfaces.
*Hint — keep scrolling and you’ll get Visually Similar results too!
Deep Similarity
Collecting media assets into galleries with a common theme or aesthetic is a common practice in stock media research. Building these galleries can often take a significant amount of time. That’s where Stocksy’s machine learning and AI can speed up this process.
Our newly-launched Deep Similarity Search feature is driven by the same technology that powers our conceptually similar recommendations. This new feature queries every single asset within the gallery, finds similar assets, and then returns the most similar assets across the whole gallery.
This means that the more assets you add to a gallery, the more the deep search will surface results that share some commonality with all the assets in the gallery. Unlike visually similarly or conceptually similar, deep similarity returns a more diverse set of assets for you to consider.