Building a World-Class Chatbot in the Haircare Category: 7 Lessons from a UX Designer

Ravi Lal
Nov 27, 2023
October 15, 2020

We have been designing and developing chat and voice-led propositions for Unilever for a few years now, and wanted to step back and reflect on lessons learned. Our Conversational UX Specialist, Kristina Csanaky, has been at the forefront of this effort; exploring how conversational experiences can drive engagement and build meaningful, one-to-one connections between Unilever brands and their customers across the world.  

Some of the largest haircare labels in the world, like Dove, TRESemme, and Toni&Guy, it’s an online platform offering style inspiration, expert advice, and the latest news and trends from the haircare industry. And now, expert advice from All Things Hair can be accessed via a chatbot called Athena; a virtual hair companion offering personalised hairstyle recommendations and advice via Facebook Messenger and the All Things Hair website in specific markets.

In this post, Kristina reflects on designing Athena and shares her recommendations for developing personalised conversational experiences for brands...

Put your Customer in the Driver Seat

We implemented a nested flow structure to cater to ‘inspiration seekers’ looking for quick ideas, as well as ‘power-users’ that want to deep-dive into haircare and styling. While Athena can recommend hairstyles or products that users can easily browse for inspiration, customers can also access tutorials, and expert advice and they can further customise the product recommendation with their favourite brands. There is no one-size-fits-all solution when building a truly useful chatbot, so we wanted to let our customers decide how much information they would like to consume.

Make it Genuinely Helpful

Content plays a key role in how ‘knowledgeable’ a chatbot is perceived to be. All Things Hair has a wealth of haircare and styling advice the chatbot can draw from, and our challenge was to serve relevant information within the space constraints of the Messenger interface without losing the dynamism of a conversation. When looking for hairstyles for a specific occasion, customers are served 3 to 5 matching styles and have the option to continue browsing, or to refine their search parameters within the  conversation.

Make it Flexible

We moved away from a linear conversation flow to create flexible, customer-led dialogues. The modular design with contextual messages allows users to switch between topics and create their own experience with the brand.

For example, the bot needs to collect information about the customers’ current hair and haircare routine to make accurate style and product recommendations. However, instead of a lengthy on-boarding quiz, users can build their profile gradually, as they explore the bot, only being asked for information that is relevant to their experience. To achieve this, we categorised the profiling questions based on which recommendation they need to be factored in, and how frequently the users might change their answers..

Let your Customers Co-Create

The power of a chatbot lies in combining engaging conversations with rich visual experiences. When Athena recommends a haircare routine, customers can swap the products with a suitable alternative to curate and personalise their beauty routine, without having to make any purchase decisions. Allowing users to interact with the products and the recommended regimen is a highly effective way to build engagement and drive product discovery. Most products are available at Unilever’s local online retail partner for customers to purchase, and for a limited time, users are even offered a discount code at the end of the experience.  

Find your Voice

Although All Things Hair includes a range of brands with a distinct tone of voice and brand proposition, it also have a unique and separate ‘voice of its own’. Chatbots can be very effective at supporting your customers throughout your marketing funnel, and a well-defined persona will help your brand stand out and leave a positive impression.

Athena serves a mostly female audience that is diverse in age and haircare needs - what connects them is a passion for stylish and healthy hair. Like a good stylist, Athena is upbeat, chatty, and friendly, which is reflected in her informal tone of voice, speedy dialogues, and strong emoji game.

Get the Value Exchange Right

No matter how carefully you craft the dialogue, sharing information in exchange for a personalised experience does require effort and commitment from the customers. Make sure you manage their expectation throughout the journey and reward them with relevant, high-quality content.

Learn Over Time

Design is an iterative process, and your customers will always find creative ways to interact with your chatbot. We regularly look at analytics and error logs to identify any content gaps, add new features, and improve the Natural Language Understanding (NPL) of the service.

About the Author

Christina Csanaky is a Senior Conversational UX Designer and Content Creator. She has been working at Voxly Digital since 2018 and has led on some of our most important projects including The Body Coach from Joe Wicks.

She is also very active in the London Design Community and is an instructor at Mobile UX. You can reach her at

No items found.

What’s a Rich Text element?

The rich text element allows you to create and format headings, paragraphs, blockquotes, images, and video all in one place instead of having to add and format them individually. Just double-click and easily create content.

Static and dynamic content editing

A rich text element can be used with static or dynamic content. For static content, just drop it into any page and begin editing. For dynamic content, add a rich text field to any collection and then connect a rich text element to that field in the settings panel. Voila!

How to customize formatting for each rich text

Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.

Have a project in mind? Schedule a 30 minute discovery call and I will at the very least give you some great advice.

Contact Us

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.