Rethinking Representation, One Image at a Time

An interview with Noun Project Co-Founder and CEO Sofya Polyakov from franklyfluent, a creative translation and copywriting house.

This interview was first published as part of a series from franklyfluent, a creative translation and copywriting house. Interview by Genevieve Edwards.

We’re surrounded by imagery – from photos on billboards, to signs telling us to walk or wait, to the app icons on our phone’s home screen. And, according to Sofya Polyakov, founder of Noun Project, every single one of those images plays a part in shaping how we see the world. We sat down with Sofya to talk about what inspired her to build one of the world’s leading icon and photo libraries, and how individual images add up to societal change.

Can you tell me a bit about Noun Project and what makes it different to other image resources out there?

At Noun Project, we’re building a global visual language that unites the world – a language that can be easily and quickly understood no matter who you are, or where you are.

We have a collection of over 5 million icons and photos that can be used for free. All the content is submitted by creators from around the world, and we have over 130 countries represented. Every month, we pay royalties to our creators. In 2021, we paid over $2 million back to the creative community.

What sets us apart is our focus on quality over quantity. We curate every single icon and photo that comes in, which is a lot of icons and photos! We curate it not only for quality, but also for diversity and inclusion, to make sure that people are represented in a light that we would consider to be fair and equitable.

We have 10 million users. They include designers, marketers, creators, educators – anyone who needs to visually communicate their ideas.

What led you to set up Noun Project?

My co-founder slash husband Edward Boatman and I started Noun Project about 11 years ago now.

He was an architect, and he couldn’t find images that he needed for client presentations. But he was aware that there were these great collections of content that, for example, the National Parks Service or the Department of Transportation had put together. And even though they were public domain, they weren’t digitalised at the time.

So we launched a Kickstarter campaign. We digitalised those public domain icons, and provided them for free. The campaign went extremely well. It got the word out, so we immediately had access to a worldwide audience and a really great following; and we raised $15,000, which at the time seemed like so much money. In retrospect, it really was not!

We’ve obviously grown, and in 2020, we launched a new offering – photos. Our goal with photos is to build an inclusive and diverse collection of extremely highly curated art-quality images. We recently looked at the stats around our best-selling images, and 77% of them featured people from underrepresented communities, which I find to be very interesting and very empowering. I think there is a huge appetite for better representation, and it’s just going to keep growing.

“Visual language has the power to shape, reinforce, and change perceptions.”

Sofya Polyakov

What drives your focus on diversity, inclusion, and representation?

Visual language has the power to shape, reinforce, and change perceptions.

Images are omnipresent. They’re all around us. They saturate our physical and digital environments. And the messages that they send can really affect how we see ourselves and how we see others in the world.

Unfortunately, there are still way too many visual resources out there that perpetuate stereotypes, unrealistic expectations, and inaccurate depictions of various communities.

One of my favourite examples is, if you go to Google and search for ‘stock photo of boss’, all the top images are white men, with several engaging in sexualised abusive behaviour.

And then if you search for ‘woman boss’ (luckily, I’ve seen some better results come up recently), but again, predominantly extremely white, very unrealistic, and sexualized. Like, come on – what year are we in?

So I think that there’s a lot of room to put out images that are not just beautiful and useful, but also represent the kind of world that we want to live in.

Prior to a pandemic, we used to hold these initiatives called Iconathons, where we get both designers and non-designers alike from the community to create new icons for civic-minded concepts that needed to be added to the public domain.

The last one we did was called the Redefining Women Collection. It turned into 60 new icons that redefined how women were represented in various roles. So it was everything from a DJ, to a chef, to a CEO, to an angel investor, to a late-night talk show host. It’s been downloaded something like half a million times now, and we’re really excited to see it circulating.

And when we launched photos, one of the collections that we put together was Diversity In Tech. If the people who might want to go into the tech field don’t see themselves represented, I feel like that can hold them back from pursuing that career choice. So we put out a collection showing the diversity that exists in tech, or that we want to exist in tech.

And then we did another collection called Empowered Women. We worked in partnership with Time’s Up and The Female Quotient to create photos of women architects, women film directors, women doing all the things that awesome women do. And again, we donated them to the public domain.

Photo by Avel Shah

It’s great you’re making more representative images available. Are there any other challenges around representation once those images are out in the world?

The other problem is that I think a lot of people are just, unfortunately, not very educated in how to use visual images in an appropriate way.

Let’s say they get a directive from the company: ‘our images have to be more diverse’. So they go and they find an image of a person with a disability. But that image might position that person in such a way that they’re not a person who happens to be disabled – the disability is the focal point. It’s like, ‘disabled person is included in our marketing materials: checkmark.’ That doesn’t help people with disabilities become more part of the normal dialogue of our society. It tokenizes them.

So that’s another way that we’re trying to curate and evaluate the collection that we’re building: is this image tokenizing the person, so that somebody can put a checkmark next to their Diversity & Inclusion efforts? Or is it portraying them in the light that they’d want to be portrayed in?

It feels like there’s an issue with our default cultural idea of ‘Person Neutral’, which is basically a white, fairly young, able-bodied man. If you see an icon of a person with no other defining characteristics, that’s who you automatically cast them as, right?

Yeah. And once you start noticing it, it becomes so prevalent. Like, when my kids and I are crossing the street, the little symbol that you see that allows you to walk – that’s a walking person. It’s not a walking man, because women happen to have the ability to walk as well.

It’s about changing the perspective that everything has to be identified as male, which, at least in the English language, has become the default. If you refer to an animal, a bird, it’s always a ‘he’; it’s never a ‘she’. It’s so weird. Even I’ll catch myself doing it, and I’ll have to readjust my language, because it is important.

Are there any groups that you want to do more to represent in icons and photos?

Yes; people of colour.

It’s actually very difficult to do this through iconography. I have a wonderful curator on my team who has written an article about how to represent people of colour through iconography. It’s tricky because icons are black and white, and because you don’t want to depict anything that would signify race or ethnicity in a negative way. There’s a tricky line. And I think that a lot of people are, unfortunately, hesitant about submitting content like that, because they don’t want it to backfire. Which is understandable, but there’s a massive opportunity for representation, and there’s a way to do it well.

Some content that we’re striving to get in photos is: people of colour hiking and doing outdoor activities. Stock photos like that are tremendously underrepresented right now, and as well as older people of colour. That content is hard to find.

That’s so true. We find it hard to get representative images of older people full stop.

There’s actually an amazing British organisation that we have recently worked with to get more of those photos up on our site – the Centre for Ageing Better. They commission photos of people who are older, you know, just living their lives; not sitting in a wheelchair staring out the window. That’s not most people. And the photos are great. They’re the kind of photos that you want to see out there, and that as a marketing person you’d want to use. We worked with them to get that collection up on our site.

They also wrote an article with us about how to shoot those photos and use them appropriately.

Do you encounter differences between the kinds of icons people from different cultures use or submit?

Once in a while our curatorial team will get iconography sent in that they just don’t understand. But then they go and look it up, and they’ll post in Slack, like, ‘Hey, did you know that so-and-so exists?’ and it’s fascinating.

There’s an icon that our team shared the other day – it’s a monster, and its signature feature is that it has an eyeball instead of a butthole. My team had to look it up, and it’s a Japanese monster called a shirime. An eyeball butt monster. We just have the spaghetti monster, which is so innocent and naïve in comparison.

It sounds like your curatorial team have a tough job.

It is challenging, and things change frequently.

For example, we’re about to publish a guide on the language that we use for our tags, because it’s always changing, and it’s important to stay on top of it. Where maybe even five years ago, there would be some repositories that would have called a woman of a certain size the f-word, now that is extremely inappropriate. But what descriptive words can you use instead? Is it ‘body positive’? Is it ‘curvy’?

How do you make sure that people can find what they’re looking for while also respecting the people that are being depicted? And how do you educate your audience in the process? It’s hard.

We use a lot of references from the organisations that are representative of the community. We try and go direct to the source and find out how the people we’re representing want to be referenced. For example, when you’re naming a photo or an icon, [in US English] you shouldn’t say ‘disabled person’; the person ‘has a disability’. You’re putting the person first. But that language is also constantly changing. And the way those within those communities like to self-identify changes too, right? And we’ve got to change with it.

Symbols can carry a lot of power. Does that ever cause issues?

It can cause issues because it’s easy to take things out of context, right? For example, we have a collection of World War II aeroplane icons on the site. And some have swastikas on the wings. We do not provide swastikas as icons in and of themselves, but we will provide historically accurate images. And we did have some users write in because they found it problematic.

Another challenge is that there’s symbology, like swastikas, that has been taken from ancient visual language and then repurposed to represent something nefarious. But if we allow the original ancient visual representation, somebody could potentially use it for a not-so-great reason. Where do you draw that line?

We’ve had people send us emails with tattoos of our icons.”

Sofya Polyakov

Do people ever use your icons in ways that are particularly interesting or surprising?

We’ve had people send us emails with tattoos of our icons, which is interesting, and a lot of commitment!

There are definitely use cases that warm all of our hearts. For example, last year, we received an email from an educator who teaches children with disabilities. And they use something that we’ve built called Thankful Bot, which we set up as a little side project for Thanksgiving.

It’s a platform where you can go in and say, ‘I’m thankful for…’, then pick one icon and another icon. So it could be ‘friends’ and ‘family’ or ‘turkey’ and ‘mashed potatoes’. And this educator said that it was the first time that they were able to do this exercise with the kids in their classroom, because they weren’t verbal communicators. But because they were able to use Thankful Bot, they were able to come up with some really cool printouts of what they’re thankful for.

Do you have a favourite icon?

I do, but it changes week to week. Right now I’m really into this Plastic Pollution icon collection. I also tend to go for the quirkier icons, like the icon for birth in an elevator.

We do something called ‘show and tell’, where our curatorial team both on the photo side and on the icon side will showcase their favourite of the works that have come in that month. It could be anything from cat memes to beautifully-designed Halloween collections.

Sofya Loves:

The Overstory by Richard Powers

Dr. Seuss classic The Lorax

Noun Project Designers and Photographers Elisabetta CalabrittoDenis Sazhin (especially his Cat Commerce and Cat Horoscope collections), Luis PradoAvel Shah, and Anna Ivanova.


About franklyfluent

franklyfluent is a creative translation and copywriting house. Launched in 2014, its founders Jordan Woolley and Catherine Barr are language graduates with advertising agency backgrounds, including DDB, BETC London and M&C Saatchi. franklyfluent’s client list includes innocent Drinks, Polaroid, Alpro, Diageo and citizenM hotels amongst others.

Subscribe to Our Newsletter for More Inspiration

Lindsay Stuart

VP, Brand Marketing & Communications at Noun Project

Related Articles