The Visual Search Cheat Sheet

Jenny Griffiths
Aug 5, 2019 · 5 min read

I’ve been working in the field of visual search now for 10 years. I know it’s hard to believe, but if you look up what I was saying a decade ago it’s pretty much exactly the same as what I say now. And that message?

Visual search has the potential to hugely benefit customer experience for retail brand. However, there isn’t a one size fits all approach for visual search, which is sometimes how it’s perceived in the industry.

When you’re thinking about visual search, here’s my cheat-sheet of the four questions you should ask yourself in the planning phases for working out the most effective way to use it, and really challenging why you might need it.

Why are you considering it?

Are you looking to be perceived a brand who embraces trends and drives innovation, or do you have a key KPI that you want to shift? Maybe it’s a combination of the two.

Understanding the context of “why” rather than “because my manager said so / because it’s an industry trend” really helps signpost:

· What success looks like — to your company, to your consumers, and how you’re being measured personally

· Which platforms you should test and integrate on

· The budget you should be assigning to this

· Do you want a tried and tested SaaS (software as a service) product, to reliably deliver those KPIs, or do you want to invest in a custom build which will push boundaries

So let’s dive into a few of these.

What are your starting platforms?

When one thinks of visual search, they tend to think of mobile. “Just take a photo on your phone and find similar.” This is a fantastic integration for certain brands with millennial audiences, but it wouldn’t be for others, for instance brands who have a desktop-first audience.

First up, you shouldn’t try to fit a new-user behavior (like visual search) into a brand-new platform. It would be like becoming a vegan at the same time as training for a marathon, when all you’ve done for the past 5 years is eaten a burger every day and never run more than 5k. You’re setting yourself up to fail. You should instead enhancea channel that you know your audience is engaging with, and build from there.

The ones that we at Snap Tech encourage our clients to think about:

· Web — you have a good, existing web user base which can also be called to action via other web experiences (e.g: email)

· Mobile — you have a young, engaged audience, who turn to your mobile app more than mobile web

· InStore — you have a shop, great footfall, and want to think this to your growing digital presence

· Social — you have a wonderfully engaged community which you wish to monetise

Needless to say, the above thought process works if you’re looking to innovate and enhance growth. If you want to shake up a particular platform, for instance mobile, you can use visual search as a great hook to do so. You’ll just have to invest more heavily and be cleverer around calls to action and PR.

Image for post
Image for post

Is this a lasting integration, or a one-off innovation?

Sometimes you want to create lasting change on a platform. Other times you want to create a huge buzz around an event or feature.

There’s nothing wrong with either — you just need to be honest with why you’re doing something and that will drive design decisions.

We work with clients on both sides of the coin — from “we need to increase time spent on site by X and conversion by Y” to “we’ve got a big red carpet event and want to create the world’s first Z powered by AI.”

As with any business decision, it’s essential for your internal teams and chosen partners to have concrete goals to pin success to, so that you’re able to work out the product design steps to achieve them. If not? You’re just going to design something that is cool that might not deliver (goodbye future innovation budget.)

What’s your internal team’s actual capability?

Be realistic about your teams bandwidth and capability. Are you a technology company? Are you specialists in visual search? Have you invested years in R&D into this area? If the answer is no, you will need to partner.

I know I’m biased, but we partner with people from hardware provisioning to infrastructure to PR to design — it’s all about finding the people who are best to deliver. If you want to deliver something stable, you need a partner who has been doing it for years. If you want to deliver something cutting edge, find yourself a company who lives and breathes that technology. Beware the mantra that “visual search is commoditized” or “there’s an API for that.” If you look at the brands who have really delivered on it, very few have built solutions in house.

Also, when you’re picking a partner, make sure you see the technology in action. Don’t be taken in by logos, promises and blog-posts (I love a bit of irony). Make sure you see a live demonstration, and make sure you pick an image to search. It’s completely fair if the company can’t do a custom integration just for you for a demonstration (if you see visual search working on one retailers’ data, it will work on yours), but make sure that they haven’t just pre-loaded examples. If you feel like they have, there’s nothing wrong with asking to test it on an image which is a fair test. For instance, when we’re demonstrating our capabilities to find similar results from flatlay images, feel free to pick your own. It gives you piece of mind that it’s not all smoke and mirrors, and it gives us piece of mind that you’re really vetting who you partner with.

So there you have it. There are the top 4 questions you should be asking yourself when you’re considering a visual search implementation on your retail brand, which will help shape your focus, and stop you from getting stung.

I’d love to hear your thoughts in the comments below.

Snap Vision

Snap Vision is a visual search company providing world class image search technology

Jenny Griffiths

Written by

Founder @SnapFashion. Software engineer to #startup CEO. Visual search obsessive and #AI pioneer at

Snap Vision

Snap Vision is a visual search company providing world class image search technology to publishers and retailers

Jenny Griffiths

Written by

Founder @SnapFashion. Software engineer to #startup CEO. Visual search obsessive and #AI pioneer at

Snap Vision

Snap Vision is a visual search company providing world class image search technology to publishers and retailers

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store