How Virtual Try-Ons Could Be the Next Step in Retail Evolution

Virtual try-ons in retailhave been around for a few years now, with varying results and user experiences. While the technology has come a long way over the past few years – there is still a way to go.

In this Disruption Talks live stream, we spoke with Netguru’s own Jinder Kang (Innovation Consultancy Lead) and Grzegorz Mrukwa (Deputy Data Science Manager) to talk more about it.

They share some great insights into the technology, challenges, benefits, and potential future of virtual try-ons.

What is a virtual try-on?

Virtual try-on technology enables customers to try on products such as clothes, shoes, cosmetics, or jewelry using their camera and augmented reality.

One of the first retail brands to use virtual try-ons was Converse back in 2012, which let buyers see what their shoes would look like on them. In the age of the pandemic and the rise of online shopping, the demand for virtual try-ons is growing.

In recent years, virtual try-ons in retail have become much more user-friendly and realistic, but the results can vary greatly.

David Stepaniuk: Do you remember what the initial virtual try-on solutions looked like?

Jinder Kang: For me personally, I remember Converse in 2012 had a way to try on their sneakers. It’s very different from what’s available now, but it was exciting at the time.

Grzegorz Mrukwa: I remember there being quite a manual process where you had a photo and just had to move your face around to fit in the photo. You could try on new hairstyles or glasses, but it was more of a template to fit into.

36% of Americans are interested in trying on clothes thanks to virtual reality. Would you say there’s a boom going on now?

Jinder Kang: Yes, I’d agree. There have been so many predictions of various tipping points in retail, so we’re still waiting for a lot to happen. With virtual try-ons, I’d say we’re not very close to that being a tipping point in retail, though.

We surveyed people to ask what they’d like to try on virtually, and shoes are very appealing to customers now. What do you think should be the next step?

Jinder Kang: We’ve seen that the technology is better suited to certain items. Shoes seem to be a sensible next step but ideally, I’d like to see clothing. The problem is that it’s very difficult to finesse that experience. I think, for now, shoes and sunglasses are the next evolutionary step.

Grzegorz Mrukwa: The challenge is with how these solutions are constructed. Elements that are not closely tied to someone’s body, like jewelry or glasses, not clothes, are the easiest to achieve.

With clothing, there’s a difference in sizing for each item. There are little nuances between each person that would need to be reflected with machine learning or some kind of augmented reality.

Getting clothing right is also about the rest of your outfit and also the feeling of the clothes. What other considerations do you think there are for virtual try-ons?

Jinder Kang: I think another interesting area is cosmetics. This is one sector that’s doing really well at the moment. Chanel recently released a new lipstick app, for example. I think with cosmetics, it’s a bit easier to implement because there are far fewer variables.

Grzegorz Mrukwa: I’d say it’s actually quite challenging because there are a lot of details to consider when you apply a specific shade. There’s the shade itself, the natural color of the lips, the lighting, whether the lipstick is matte or not. It’s pretty complex.

What has your experience been like with luxury brands?

Grzegorz Mrukwa: The challenge is how to make the virtual try-on practical, not just look fun. If you make it just fun, it becomes an entertainment tool, not a sales tool. It might make people talk about your brand, but that’s not what you want to achieve.

If we want an authentic experience, we can use a human model, not just a 3D model. This will be able to replicate certain light reflections and so on.

Would you say that virtual try-ons are a selling tool?

Jinder Kang: I think it depends on how you plan to incorporate virtual try-ons. I think in the future, it will definitely lead that way, and for some already, it’s a channel for purchase. It’s also about the ability to purchase from anywhere. I think rather than a linear journey, the retail purchase experience will combine virtual try-ons and social commerce.

Can you talk about the type of data collection required for virtual try-ons?

Grzegorz Mrukwa: It depends on the solution you want to implement. At Netguru, there are some specific requirements we get from clients. What the client wants influences the data collection procedure. For example, if you need a really high resolution or frame rate, you will need a lot of videos recorded in similar conditions. If you just want to process photos, you may only need to find similar photos on the internet.

There are also different approaches if you need a 3D model. While in this case, it wouldn’t look very realistic, it’s still a viable solution for virtual try-ons.

What do you think the differences are between virtual try-ons and virtual fitting rooms in brick-and-mortar stores?

Jinder Kang: They’re pretty much the same thing because they’re both virtual ways to try things on. I think in a store, you have more of an opportunity to push computational power. You can use body scanners in a fitting room. This could enable a much smoother experience. I think that it could be a potentially better experience that could replace current online methods.

Cloud computing can be used for virtual try-ons, but hardware is a great advantage for virtual fitting rooms. Would you agree?

Grzegorz Mrukwa: I think when talking about the Cloud, we need to consider how much computational power you require for this. There’s usually a very high requirement, so this is why try-on applications have appeared on mobile pretty recently.

These types of apps work great on some of the newest smartphones out there, which will allow retailers to push this computational power requirement onto the user. If a retailer is doing this all itself for thousands of customers, it could be a challenge financially.

If specific apps need to be downloaded, will this affect the customer journey?

Grzegorz Mrukwa: You would need to download the machine learning model to your device, but whether you load it up via a web page or a mobile app, it doesn’t matter. But we need to take that into account because it could be challenging for the customer to download every time.

Jinder Kang: A poorly orchestrated try-on could actually discourage a consumer instead of helping them make a purchase decision. What I've seen a lot of recently is people trying to utilize these new channels, but not really putting in the know-how and the expertise and the experience design side of things.

As a result, you end up with something that is not sympathetic towards a brand, image, tone of voice, or anything like that. That can have a very detrimental effect on your final consumer.

Do you think brands that do not follow the virtual try-on trend will be left behind?

Jinder Kang: It depends on your outlook on the future of retail. I think eventually, virtual try-ons will almost become a hygiene factor.

I think there are a lot of benefits for retailers. It could reduce the number of returns, for example. People would be more confident in their purchase choices. In reducing those returns, you can also avoid unsustainable business models. So that's also worth remembering, regardless of size, it could be beneficial to your business model as well.

I think retailers will need to consider this as they could be left behind in the future, especially as AR is becoming more popular with younger people.

For the last few months, our experts have been researching and analyzing selected VTO solutions. The results of their work will help us understand how consumers perceive virtual try-on, which elements of the experience influence them the most, and how brands can maximize the benefit.

Click here to download the Virtual Try-On Report.