Main content

Augmented Marketing is the future as the function gets a gen AI makeover at L’Oréal

Stuart Lauchlan Profile picture for user slauchlan March 26, 2024
Summary:
Generative AI is combining with human creativity to revolutionize marketing, says Asmita Dubey, Chief Digital and Marketing Officer at L’Oréal Group.

marketing

With generative AI we are at a new frontier of technology. We are at a point where creativity and technology are coming closer together like never before.

A bold declaration from Asmita Dubey, Chief Digital and Marketing Officer at global beauty firm L’Oréal Group, as she contemplates how gen AI tech is revolutionizing marketing. It’s the latest stage in an ongoing process, she suggests: 

Beauty marketing has been getting augmented by multi-sensorial experiences in the physical, digital and virtual worlds over the last few years, Beauty marketing is also getting augmented by data used across business labels. And now beauty marketing is getting augmented by generative AI.

Augmented Marketing is the future of marketing. What we are doing is we are exploring Augmented Marketing with generative AI technology across different parts of marketing, across content creation. We are bringing a new generation of beauty services, both offline and online, to our beauty consumers.

Dubey concurs with the thesis proposed by WPP's Stephan Pretorius about enthusiasm for generative AI among creatives

L'Oréal is a strong creative company. There are very good marketers in our company. What I find fascinating is how creativity and technology are coming closer together, how we need a new generation of marketers who will be as apt at using the left brain and the right brain, as apt in doing data sets for the brands versus thinking creatively about what their brand values and brand proposition should be. I feel this opens a great opportunity in training and creating and having a new generation of marketers.

She adds:

Through this whole creative journey, we are definitely using a combination of man and machine to create beauty that moves the world with generative AI. Technology is sparking creativity like never before in L’Oréal Group. We are ready to embrace creative augmentation on one side and content automation on the other side. We are a leading beauty tech player. We believe that that beauty tech is something that is going to help us to bring beauty to each of us. 

Responsible

Given generative AI’s capabilities to power deep fakes, L'Oréal is taking its responsibilities seriously, says Dubey, with the firm having set up a generative AI taskforce to lay down guidelines and set out its dos and don’ts: 

At L'Oréal Group we have decided that we will not use AI -generated face, skin, body or hair to enhance our beauty products in our external representation of beauty. Yes, we will use AI-generated face hair, skin and body when it comes to to inspiration, new codes of beauty when it comes to sparking imagination, when it comes to internal storyboarding. But we will not use it to enhance our external representation of beauty. 

A lot of thought has gone into this, she adds, and at the highest level: 

What we said was that the beauty is something that we use on ourselves and it is something that enhances us as people, men and women. That's why, with our CEO, we took the decision that we are not going to use AI-generated lifelike faces or hair or skin or body to enhance the product benefits. 

You won’t put an AI image of a man or a woman and say, 'Look your skin is so much better'. That  would mean that you're enhancing a product benefit on something that is not real, whereas we would want to show that the product product benefit is on someone who is real and lifelike and you can identify with. So that is the position that we have taken as a leader in beauty.

But the firm is keen to experiment with gen AI’s potential and has set up a unit called CreAItech as a safe space for that purpose, she explains: 

Over the last six months now we have been working together with a lot of partners.  We have been working with Nvidia, with WPP Open and several Large Language Models and we have created more than 800 beauty images. These images have been created with dozens and dozens of workshops with our brands. We have used about 20 different gen AI techs and we are exploring and learning from all of this…There are pros and cons. At the same time what it requires is strong, creative direction. It needs prompts that have imagination and that have very strong brand cues for the images to start becoming relevant for us to use it to augment marketing. 

Shared vision 

This augmented creativity journey began in April 2023, she recalls, with an internal gen AI summit inside the company, supported by technical and creative third parties. Partnerships with the likes of WPP and NVIDIA were formed and CreAItech was set up. Then, says Dubey, testing could begin and learnings gathered. Building on a shared vision is key: 

Building on our shared vision means custom brand models.  I think we are taking kind of three insights and learnings in three different buckets - the first one on tech readiness, the second one on tech modularity, and the third one on data value. In terms of tech readiness, we [have] experienced that there is an uneven tech readiness in terms of outputs, but also in terms of the input capabilities because they keep emerging and there are new capabilities coming all the time. 

In terms of tech modularity, if you're to have a diverse and modular gen AI tech stack, that could be helpful for a large corporate like ourselves because there's so much innovation happening in this space that we'd like to keep it modular. And in terms of data value. we need effective management for brand data sets of our training data. Once we have effective management of the brand data sets we can be more assured in terms of the quality of the output. We can also be more assured in terms of compliance.

Nearly 12 months into this journey Dubey says L'Oréal is still in a 'test and learn' phase, but is seeing results: 

We are in a controlled experimentation phase today and in that controlled experimentation phase, we are saving time, saving almost 200% to 300% in time generating  product images. And then you can assume that time is money. 

But there’s a need to think about service models and scalable models of adoption for the Group overall in the coming six months:

We're trying to assess efficiency. Do we reduce or save time and resources? Effectiveness? Do we improve time to market do we improve volume of production? Do we improve performance and quality of production overall? Our assessment is also on are we creating value? Are we increasing consumer engagement, conversion to more satisfaction or overall ROI? At the same time we are asking ourselves a lot of questions on IP rights. We are asking ourselves what are the challenges that are ahead for us when we have to scale the technology. 

Scalability is of particular interest:

In the CreAItech lab, we have generated 800+ images over the last six months. We have, of course, in our countries and zones, some product detail pages on e-commerce that we are working on with our partners. So there will be thousands of images there, but still the number is today is a few thousand, whereas if you look at our assets, we have hundreds of thousands of assets every year, because we produce a lot of content. We have to think about what is the model of scaled adoption and that model of scale adoption will need brand data sets, will need tech modularity, will need a technology architecture.

The firm is also beginning to think of [AI] for research and innovation, she adds: 

AI in general helps in the formulation in understanding different hair types. It understands it much faster. So there are different use cases there.  One use case is simply to understand diverse hair types or diverse skin types, which a person or human scientist will take a certain amount of time to do it, but the AI can do it much faster and then it helps on the formulation.

But it's on the creative front that gen AI is having its most visible impact. Dubey asserts: 

Technology is sparking creativity and it is sparking creativity amongst copywriters. It is sparking creativity amongst 3D artists. It's sparking creativity amongst all of our marketers, and therefore we are trying to shape the future of beauty.

Loading
A grey colored placeholder image