To highlight the practical use of the proposed-driven strategy, a prototype of Bloom was developed to illustrate how users interact with the chatbot within the Parsley & Twine website. The chatbot was created using Landbot, enabling the design of structured conversational flows supported by conditional logic and personalised pathways (Huang & Rust, 2018). This prototype reflects a guided shopping experience, where users are led through a series of tailored questions to identify their needs, preferences, and style (Rose et al., 2012).

​

The chatbot begins with a friendly introduction, where Bloom presents itself as a virtual retail assistant and invites users to begin their journey. Users are initially given two primary options: “Find a Gift” or “Style My Look.” This early segmentation allows the chatbot to immediately tailor the experience based on user intent, ensuring relevance from the outset (McKinsey & Company, 2023). The use of limited, clearly defined choices reflects Hick’s Law, which suggests that reducing the number of options available decreases decision time and improves user experience (Hick, 1952). By simplifying the initial interaction, Bloom reduces cognitive effort and encourages users to engage further with the journey.

​

For users selecting the gifting pathway, Bloom asks whether they need inspiration or already have a specific idea in mind. The conversation then narrows further by identifying the recipient’s style, such as pastel and neutral or colourful and bright. Based on these responses, the chatbot provides curated product suggestions, positioning them as “inspired picks” to create a more personalised and engaging experience (Puccinelli et al., 2009). This progressive narrowing of choices enhances usability by breaking down complex decisions into smaller, manageable steps, supporting a smoother and more intuitive user journey (Sweller, 1988).

​

Similarly, users selecting the styling pathway are guided through a series of questions about their personal style preferences. They can choose between options such as “Inspire Me” or “I have a style in mind,” followed by style descriptors such as “Chic and understated” or “Vibrant and bold.” Bloom then responds with tailored styling suggestions, reinforcing the feeling of a personalised consultation (Bleier et al., 2019). This approach strengthens perceived relevance and aligns with user expectations for personalised digital experiences, increasing engagement and satisfaction (McKinsey & Company, 2023).

​

A key feature of the prototype is the integration of lead capture within the conversational flow. After delivering recommendations, users are prompted to provide their details in order to receive exclusive deals, updates, or personalised suggestions. This creates a clear value exchange, where users are incentivised to share their information in return for continued personalised engagement (Baird & Parasnis, 2011). Embedding lead capture within the interaction, rather than presenting it as a separate or disruptive form, reduces friction and aligns with best practices in user experience design (Gefen et al., 2003).

​

The overall design of the chatbot prioritises simplicity, clarity, and personalisation. Each step in the flow is structured to progressively narrow user choices, reducing decision fatigue while maintaining engagement (Iyengar & Lepper, 2000). The use of predefined buttons ensures ease of interaction, while conditional logic enables Bloom to deliver relevant outcomes based on user responses (Huang & Rust, 2018). This structured yet flexible design enhances both usability and performance, ensuring a smooth user navigation experience with minimal effort.

​

However, the effectiveness of the prototype is dependent on the quality and relevance of its conversational design. If the pathways are too rigid or recommendations fail to align with user expectations, the experience may feel limited or impersonal. Additionally, some users may prefer independent browsing rather than guided interaction, highlighting the importance of offering the chatbot as an optional support tool rather than a mandatory feature (Lemon & Verhoef, 2016).

​

This prototype demonstrates how Bloom can be effectively implemented using Landbot to create a dynamic and interactive shopping assistant. When integrated with HubSpot Free CRM, captured lead data can be stored and used for follow-up communication, while Google Analytics can track user behaviour and conversion rates to support ongoing optimisation (Shankar, 2021). Overall, the prototype illustrates how the chatbot transforms passive browsing into a guided, personalised journey that supports both user experience and lead generation (Verhoef et al., 2021).