Enrico Roselli – Co-founder Digital Fashion Academy. Former CEO of La Martina

Gianluigi Zarantonello – Former Innovation Manager at Valentino

Enrico Fantaguzzi – Co-founder Digital Fashion Academy. Former Manager at Gucci, Tod’s, Yoox

How is AI changing fashion retail?

To create a seamless shopping experience, AI must facilitate a frictionless journey where the customer can move between e-commerce, physical stores, and mobile platforms without being conscious of the channel they are using.

The goal is to make the technology “invisible,” operating in the background to enable users to do whatever comes to mind without technical barriers,.

How can AI help to create a seamless customer experience in Retail

Unified Customer Identity and Recognition

A major hurdle to seamless shopping is the disconnection between online and offline Customer Identities. A customer might be recognised in their local shop but treated as a complete stranger online, or vice versa. AI can bridge this gap by analysing cross-channel data. For instance, AI can leverage a customer’s online browsing history, where they may spend time without buying, to enhance their experience and opportunities when they eventually enter a physical retail store,. This requires building technical capabilities to recognise the customer at various touchpoints, creating a holistic view of their journey.

Hyper-Personalisation via Generative AI

AI is moving personalisation beyond traditional customer segmentation. Previously, retailers might identify a segment but still show the same content to everyone within it due to production limitations. Through generative AI, retailers can now unlock data to produce unique content tailored to specific individuals based on their interactions and intent. This allows for a highly personalised experience that still adheres to the brand’s overall creative vision,.

Frictionless Payments and Loyalty

AI enhances the final stage of shopping, the transaction, by integrating advanced payment systems. Machine learning algorithms contribute to a seamless experience by:

• Scoring transaction reliability to instantly approve valid purchases, thereby improving e-commerce conversion rates.
• Utilising tokenisation and blockchain to create safe connections that integrate directly with loyalty programmes, managing client information securely to further customise the experience.

Agentic AI for Natural Interaction

The emergence of “Agentic AI” changes how users interact with complex software. Rather than forcing customers or employees to navigate rigid interfaces and rules, Agentic AI allows them to use natural language to ask for exactly what they need,. This shifts the paradigm from the user adapting to the software to the software adapting to the user, making the interaction intuitive and removing the complexity of the underlying technology,.

Empowering Physical Retail

Since a significant majority of transactions (around 80%) still occur in brick-and-mortar stores, AI is crucial for enhancing the physical environment. This includes using data to improve store efficiency and potentially integrating robotics for deeper interactions, ensuring the physical space is as responsive and data-informed as the digital one.

How can generative AI produce unique content without losing brand vision?

Generative AI allows companies to produce unique content at scale by unlocking data and production capabilities that were previously limited. Historically, even when retailers identified specific customer segments or “personas,” they were forced to show the same content to everyone within that group due to the difficulty of producing numerous variations. Generative AI solves this by using customer data to create individualised content dynamically.

To ensure this unique content does not compromise the brand vision, the process relies on the following principles:

• Adherence to Core Creativity: The brand vision remains firmly in the hands of the creative director and the company. The content produced by AI serves as variants of this central vision rather than independent creations, ensuring that the foundational aesthetic and message provided to the customer remain consistent.
• Strict Guidelines: Businesses can “teach” the AI to provide outputs that adhere strictly to the exact guidelines of the brand, ensuring that automated content stays within the established identity.
• Human-Centric Finalisation: While AI is excellent for brainstorming and providing “divergent,” out-of-the-box ideas to spark inspiration, human creatives are still considered “much better than AI” at creating the final, polished output. The AI acts as a tool to expand possibilities, not a replacement for the creative mind that defines the brand.

This approach allows brands to move beyond generic segmentation to true personalisation without the fear that the technology will distort the brand image.

How can AI identify customers across online and physical stores?

AI identifies customers across online and physical stores by integrating specific “technicalities” into the customer journey that link data points, rather than relying on “magic” or passive observation.

According to the sources, AI achieves this cross-channel identification through the following methods:

Payment and Loyalty Integration
One of the most effective ways to identify customers is through advanced payment solutions. By applying AI to payments, retailers can use tokenisation and blockchain to create safe, secure connections that link transactions directly to loyalty programmes. This allows the system to instantly recognise the client via their payment method, unlocking their history and preferences to customise the experience without requiring complex manual logins.

Leveraging Digital Intent in Physical Spaces
AI bridges the gap between the two worlds by analysing a customer’s online browsing history to enhance their in-store experience. For example, in the luxury sector, e-commerce conversion rates can be as low as 1%, meaning 99% of visitors are merely browsing. AI can capture this “intent to buy” or interest shown online and make it accessible to store staff when the customer eventually visits a physical location, transforming a “stranger” into a known prospect with specific preferences.

Designing Identification Touchpoints
The sources emphasise that technology alone cannot identify people without a strategy. Retailers must engineer specific “moments in the journey” where the customer is prompted to reveal their identity. Whether through a mobile app interaction, a loyalty scan, or a tokenised payment, these intentional touchpoints provide the data AI needs to connect the online user with the physical shopper.

What are the hidden costs of implementing AI in fashion?

While the surface narrative of AI often suggests a “plug and play” solution, there are several hidden costs and challenges associated with its implementation in the fashion industry.

Unpredictable Financial Costs A significant hidden cost lies in the pricing models of corporate AI tools. While individuals may be accustomed to free versions of tools like ChatGPT, corporate environments face costs based on consumption. This variable pricing makes financial planning highly volatile; while CFOs and boards may request three-year forecasts, it can be difficult for technical leaders to predict consumption costs even one week ahead. Without proper knowledge and management, companies risk wasting money rather than investing it effectively.

Organizational and Cultural Reinvention Implementing AI is not merely a technical update but requires a complete “reinvention of the organization”. The costs involve:

• Cultural Shift: There is a need for a deep change in mindset and culture to adopt these new instruments, as they cannot simply be used from scratch without preparation.
• Education and Literacy: Companies must invest in educating their workforce. There is a risk that without “literacy” (understanding) and “fluency” (ability to use), employees will struggle to adopt the technology.
• Underutilization: A historical hidden cost of technology is that companies often pay for tools with huge potential but only utilize 20% to 25% of their capabilities because they are too complex for staff to use effectively.

Data and Process Preparation Before AI can be effective, significant resources must be spent on the “boring” background work.

• Data Quality: AI operates on the rule of “garbage in, garbage out.” High-quality output requires high-quality structured and unstructured data, necessitating substantial work to clean and organize existing databases.
• Process Mapping: Companies must invest time in mapping their actual processes to understand what they want to achieve, rather than applying AI to obsolete workflows which would result in little more than a faster version of a broken process.

What strategies help manage the consumption-based costs of corporate AI?

To manage the unpredictable consumption-based costs of corporate AI, businesses must shift from a fixed-budget mindset to a value-driven strategy rooted in education and process efficiency. Unlike consumer tools like the free version of ChatGPT, corporate AI vendors typically charge based on consumption, making financial forecasting difficult for CFOs and boards.

Based on the sources, the following strategies help manage these costs:

Invest in “Literacy” and Preparation
The most critical cost-management strategy is education. Leaders must possess “literacy” (understanding) and “fluency” (ability to use) regarding AI technologies. Without deep knowledge, it is impossible to determine whether the organisation is “wasting money or investing money”. Thorough preparation allows companies to unlock actual value, whereas jumping in unprepared leads to financial loss and reduced competitiveness.

Optimise Processes Before Automation
Implementing AI requires a “reinvention of the organization” rather than a simple software installation. A major hidden cost arises from applying AI to obsolete processes; this merely results in a faster version of a broken workflow without generating real value. To control costs, companies must perform rigorous process mapping to ensure the AI is solving a valid problem efficiently.

Focus on Efficiency and Sustainability
Strategies should aim for maximum efficiency across the value chain, from sourcing to logistics and marketing. This approach aligns cost management with sustainability: by using technology to identify and eliminate waste (whether materials or energy), companies can simultaneously reduce costs and meet environmental goals.

Avoid “FOMO” Driven Adoption
Entrepreneurs are advised to be “bold but not a fool.” Rushing to adopt AI due to the “fear of missing out” (FOMO) often leads to wasteful spending on tools that are not fully understood or utilized. Taking the time to understand the specific value proposition for the business ensures that the consumption costs incurred actually drive return on investment.

How can AI identify waste and improve sustainability in fashion?

Based on the sources, AI identifies waste and improves sustainability in the fashion industry by reframing the concept of sustainability from a regulatory “burden” to an exercise in maximum efficiency.

Here is how AI achieves this across the value chain:

Redefining Waste as Inefficiency
AI helps retailers understand that waste—whether it is energy, materials, or products—is simply something that has not been utilised effectively. Instead of viewing sustainability solely as compliance with regulations, AI allows companies to view it as the elimination of inefficiency. If a resource is “put apart because it’s not useful to anyone else,” it represents a financial loss as well as an environmental one.

Granular Visibility and Root Cause Analysis
To improve sustainability, AI acts as a diagnostic tool that illuminates the “road ahead”. It provides access to data across the entire value chain—from sourcing and logistics to sales and distribution. By analysing this data, AI can:

• Pinpoint Origins: Identify exactly where waste is coming from in the production or supply process.
• Analyse Consumption: Monitor and reduce the usage of materials and energy.
• Forecast Accuracy: Help forecast the future to prevent overproduction, ensuring that what is produced is actually needed.

Maximizing Asset Utility
Once waste or potential waste is identified, AI helps determine how to maximise the value of every item produced. If a product fails in its primary channel, AI can calculate the best alternative path, such as:

• Selling the item in a different market.
• Repurposing it for a different use.
• putting it back into circulation rather than discarding it.

Aligning Profit with Planet
The sources highlight that cost reduction and sustainability are now intrinsically linked. With the costs of doing business in fashion (sourcing, advertising, logistics) skyrocketing, AI drives the efficiency needed to lower expenses. This creates a scenario where the “perfect company in terms of efficiency” is also a sustainable one, fulfilling both business objectives and legal requirements (such as those in the European Union) simultaneously.

Key takeaways from NRF 2026

Based on the discussion regarding the NRF (National Retail Federation) and the state of digital transformation leading up to 2026, the key takeaways focus on moving beyond the “hype” of AI to practical, structural, and “invisible” implementations. The sources suggest that by 2026, the industry focus will shift from simply adopting new tools to a fundamental reinvention of the organisation and the customer journey.

Here are the central themes discussed in the context of NRF 2026 and Fashion AI:

1. From Omnichannel to “Seamless” and “Invisible” Experiences The speakers argue that the industry is moving past the concept of “omnichannel,” which can feel disjointed, toward “seamless” experiences. In this model, the customer moves between e-commerce, physical stores, and mobile apps without consciously thinking about the channel.

• Invisible Technology: A major takeaway is that technology should be “invisible,” operating in the background to enable the user’s desires rather than being the focus itself.
• Unified Identity: A critical component of this seamlessness is using AI to bridge the online-offline gap. For instance, using payment tokenisation and blockchain to recognise a customer in a physical store and instantly access their online browsing history and preferences.

2. The Rise of Agentic AI “Agentic AI” is identified as a dominant trend at NRF, though the sources caution that much of it is currently marketing noise.

• Paradigm Shift: True Agentic AI shifts the paradigm from users adapting to software rules to software adapting to users. Instead of navigating complex interfaces, users (both employees and customers) can use natural language to command the software to perform complex tasks, such as creating a landing page by fetching data from disparate databases.
• Current State: While promising, the sources note that the industry is still “far away from actually adopting the real Agentic AI potential,” as it requires robust data structures that many companies currently lack.

3. The End of “Plug and Play” A recurring theme is the demystification of AI as a simple add-on. The sources emphasise that AI is “not plug and play”.

• Organisational Reinvention: Successful implementation requires a complete reinvention of the organisation, including culture shifts and extensive employee education.
• AI Literacy vs. Fluency: There is a distinction made between “literacy” (understanding the concepts) and “fluency” (the ability to use the tools). Leaders must possess both to avoid wasting money on technologies they do not understand.
• Process Mapping: Before applying AI, companies must map and optimise their processes. Applying AI to an obsolete process merely results in a faster version of a broken workflow.

4. Hyper-Personalisation via Generative AI The sources highlight that Generative AI is finally unlocking the true potential of personalisation, which was previously limited by production capacity.

• Scaling Content: Retailers can now use GenAI to produce unique content variants for specific individuals based on their data, solving the problem of having customer segments but only one piece of content to show them.
• Brand Integrity: This can be achieved without compromising the brand vision. The creative director sets the core vision, and AI generates variants that strictly adhere to these guidelines, ensuring consistency while increasing relevance.

5. Sustainability as Efficiency In the context of 2026, sustainability is reframed not as a regulatory burden but as an exercise in maximum efficiency.

• Waste Reduction: AI provides visibility across the value chain (sourcing, logistics, sales), allowing companies to identify “waste” as simply “inefficiency”—resources that are not being used effectively.
• Profit Alignment: By using AI to forecast accurately and maximise asset utility, companies can simultaneously reduce costs and meet environmental standards, such as those in the European Union.

6. The Financial Reality of AI Finally, a practical takeaway for the roadmap to 2026 is the shift in cost structures. Unlike consumer tools, corporate AI often operates on consumption-based pricing.

• Budget Volatility: This makes financial forecasting difficult for CFOs, as costs fluctuate with usage. The sources warn that without proper knowledge (“literacy”), companies risk “wasting money” rather than investing it, making education essential for financial control.

Watch the video

If you’re a small fashion brand owner or fashion marketing professional juggling 13847 tasks at once – writing ALL your marketing copy on a tiny budget can feel like you’ve been hired to style an entire editorial shoot with just three pieces. Impossible! 

Evelina Kaganovitch

Fashion Marketer, Creative Strategist, and Copywriter | Founder of Evelina Kreative
Melbourne, Australia

Evelina helps fashion entrepreneurs craft messaging strategies that convert into high-performing copy for websites, sales pages, ads, emails, and other marketing assets.

If you’re stuck in the middle of pumping out product descriptions, social media posts, ads, emails, and new weekly blog posts… AND you don’t have a fashion copywriter on call 24/7, then you need to read this. 

Here is everything you need to know about making ChatGPT your new copywriting and marketing assistant. If you learn to use AI in your fashion marketing workflow the RIGHT way – you’ll see that it’s 100x MORE reliable than the *assistant* Miranda Priestly always dreamt of.

This isn’t about replacing human creativity; AI hasn’t mastered the art of a perfectly timed fashion reference yet. It’s about using fashion tech to streamline, scale, and speed up your content creation process, giving you MORE time to focus on what you do best… creating amazing fashion experiences for your customers.

Even McKinsey published a report titled Generative AI: Unlocking the future of fashion – confirming what marketers have been saying that AI is NOT going anywhere.

Why ChatGPT is perfect for fashion marketing

Small fashion businesses are turning to AI tools faster than shoppers rush to a sample sale, and here’s why. When you’re bootstrapping a brand, every dollar counts. Traditional copywriting services can cost hundreds or thousands of dollars monthly (and as a fashion copywriter, I can guarantee you they are 100% worth it, BUT they are not worth going into debt for). ChatGPT costs less than a cup of coffee a day and can still create content that grows your marketing and saves you time.

ChatGPT is SO good at doing the 80% of grunt work that in a fashion agency setting, a more junior marketer or copywriter would do. Think of it as a marketing assistant that brings your vision and strategy to life. 

Need social media captions that capture the vibe of your latest collection? Done. 

Want to draft email campaigns for your seasonal sale? Consider it handled. 

Looking for product descriptions that actually sell? ChatGPT can generate these fast.

The best AI for copywriting shines when you’re dealing with high-volume content needs. Instead of staring at a blank page, wondering how to describe your fifteenth pair of earrings this week, you can generate multiple variations and choose the one that resonates most with your brand aesthetic and voice.

The benefits (good) and limitations (bad) of ChatGPT

Benefits of using ChatGPT in fashion marketing:

Speed is ChatGPT’s superpower. What might take you an hour to write can be drafted in minutes, leaving you more time for strategic planning or creative development. Cost-wise, it’s great for small business content marketing. You’re looking at a monthly subscription that costs less than a single freelance blog post.

The versatility factor can’t be ignored either. ChatGPT handles everything from Instagram captions to email sequences to blog posts about styling tips. It’s like having an LBD for your content needs.

Limitations of using ChatGPT in fashion copywriting:

AI generated copy can feel as generic as fast fashion if you don’t add your personal touch. ChatGPT doesn’t understand the subtle emotional triggers that make your brand unique. It can’t capture the feeling of slipping into that perfect vintage jacket or the confidence boost from wearing a piece that truly represents your style.

As I usually tell my clients… “AI works well for tasks like generating product descriptions or email drafts, but it needs a human touch to ensure the messaging resonates with your audience.”

How to Use ChatGPT for copywriting and marketing in your fashion business

Using AI for Email Campaigns 

Email marketing is a great marketing channel for all fashion business owners and fashion marketing professionals. Add ChatGPT to your email marketing tech stack to brainstorm and organise content ideas, draft click-worthy subject lines, and engaging copy.

How ChatGPT can write Product Descriptions

For many small fashion brands, product descriptions are a big time taker. With new styles dropping monthly, or even every few months, it can feel demanding to constantly think of what to write to win over customers to buy your product. Try using ChatGPT to generate your product descriptions, which can help speed up your workflow. 

The best use of ChatGPT for Social Media content

Social Media can feel like a never-ending grind. Many fashion brands rely on social media as a key marketing channel to connect with loyal customers and attract new ones. This means there’s always a need to create new, engaging social media content across Instagram, TikTok, Facebook, Pinterest, LinkedIn… and the list goes on. Integrate ChatGPT into your workflow to accelerate the process, reclaim your time, and generate content ideas that genuinely engage your audience. 

Speed up Blog Content creation with AI

If you’re creating educational content like blog posts, authority driven pieces, or writing long form articles, that positions you as a fashion authority, ChatGPT can help. Ask ChatGPT to help create topic outlines, draft blogs, and even edit and spell check final versions. You can ask AI to help you with anything from writing guides, seasonal trends, to styling guides, or industry insights that keep your audience coming back for more.

The 80/20 rule of using ChatGPT in fashion and still sounding human

10% – You come up with the overall strategy and direction of the content and give ChatGPT a really insightful prompt to give you the best output. 

80% – The generative AI does the heavy lifting of coming up with outlines, first drafts, and headline ideas.

10% – You come in for a final edit to give it a ‘human’ touch at the end, ensuring the content still carries your brand’s personality, voice, and tone. 

Simple editing tips for making AI sound *On Brand*

The magic happens in the editing phase. Once you have your ChatGPT generated draft, then adjust the tone to match your brand personality. Make sure you’re clear on your brand values and brand voice so all your content across different channels sounds like YOU or your brand. 

The extra hidden layer is sprinkling in personal anecdotes or specific details that only you would know about your products or design process. 

Pay attention to vocabulary/word choices. If your audience consists of eco-conscious millennials, incorporate sustainability language naturally. If you’re targeting luxury consumers, elevate the sophistication level of your descriptions. 

Use ChatGPT’s regenerate feature strategically. If the first version feels too corporate, ask for a more conversational tone. If it’s too casual, request something more polished. Think of it as having multiple drafts at your fingertips.

Remove anything that looks and feels like the typical ChatGPT content

Here’s a quick checklist of the most commonly used AI nuances and words that make it obvious you used ChatGPT to write your content.

Ready to give your fashion marketing and copywriting a glow up?

ChatGPT is a TOOL. It’s not an entire solution. And it’s a powerful tool to have in your fashion tech stack, especially for small fashion businesses and fashion brand owners balancing multiple responsibilities. 

AI won’t replace the nuanced understanding of a seasoned copywriter, but it can be a great starting point if copywriting and content creation are something you find difficult. Look at AI as your fashion copywriting partner, not your replacement. Use it to overcome writer’s block, generate ideas, and handle routine copywriting tasks, then add your unique perspective to make the content authentically yours.

ChatGPT will never replace hiring a strategic high conversion fashion copywriter

A skilled fashion copywriter can work with you to streamline your entire marketing strategy and funnel. They can create copy that performs above industry averages. 

For brands that rely heavily on content to generate revenue, a fashion copywriter is a great investment to scale your business and connect with your audience on a deep, psychological level. 

When the time comes, if you would like to learn more about hiring a copywriter, DOWNLOAD this FREE GUIDE “How to Hire a Professional Copywriter,” and discover WHEN it’s time to invest in human expertise alongside your AI tools. 

Download the guide

Your Step by Step Guide to hire a Copywriter

…that gets your fashion business

Please enable JavaScript in your browser to complete this form.

Name *

Email *

By sending this form I accept the Privacy Policy

Download the Guide

ChatGPT for Fashion Marketing: AI Copywriting Guide 2025 [Download your free guide – created by high-conversion fashion copywriter]

By Enrico Fantaguzzi, Editor-in-Chief, Digital Fashion Academy

May 25th 2025, updated

At Google’s annual I/O conference for developers, Google announced several new advances, including a more personalised multimodal AI search experience for shopping queries, agentic checkout capabilities and an enhancement to its virtual try-on tool that will allow users to see products on their own body.  

What does “agentic” checkout mean?

Agentic checkout, it’s an ecommerce check out that behaves autonomously. I.e. the check out acts like an agent that completes a series of actions without you being involved. For example the agentic checkout could ask you, “Do you want to buy this product?” and, if you say yes, it might be able to complete the purchase on it’s own.

How is the agentic checkout able to complete a purchase on your behalf?

The agentic check out can complete the purchase on our behalf if it has all the necessary information to complete the checkout. First of all it needs the payment method saved – Google Pay wallet – and then it may need confirmation of your shipping address.

Google’s AI virtual try-on (VTO)

Virtual try ons are not new to the digital fashion industry, I personally used virtual try ones and virtual fitting solutions in some of the companies I worked for (Twinset, 7 For All Mankind) with excellent results in terms of conversion rate but recently the improvements in AI have made possible to achieve more efficiency (savings) to implement VTO for fashion products.

At I/O 2025, Google unveiled a significant upgrade to its AI Mode shopping experience: a powerful virtual try-on feature that allows users to upload their own photos and visualise garments on their bodies. Powered by advanced generative AI, the system simulates fabric fit and texture across various body types, promising a new level of personalization in online shopping.

This development has the potential to be a game-changer for fashion e-commerce. But beyond the consumer-facing benefits, what does this mean for fashion brands, especially those navigating the complexities of digital transformation?

Organizational Impact

Fashion brands will need to rethink how they organize their product content. To fully take advantage of AI-powered platforms like Google’s, brands must invest in rich, structured product data, including detailed images, accurate sizing information, and standardized metadata. Product photography will likely need to evolve to include multi-angle, high-resolution, and fabric-dynamic visuals that AI models can effectively process.

Furthermore, cross-functional teams, marketing, merchandising, IT, and customer experience, will need to collaborate more closely to align content creation, technology integration, and user experience. This may drive brands to create or expand dedicated digital innovation teams.

Business Strategy Implications

From a business perspective, these AI tools could significantly impact conversion rates, return rates, and customer satisfaction. By reducing the uncertainty associated with fit and appearance, virtual try-on tools may decrease returns, one of the biggest operational costs in online fashion retail.

Additionally, AI-powered personalization could improve customer loyalty by offering a more tailored experience. Brands that successfully integrate these tools may also gain a competitive edge in SEO and product discoverability, especially if they are early adopters and featured in Google Shopping results.

However, brands must also consider data privacy, AI ethics, and moderation challenges. Reports have already highlighted potential misuses of generative AI in retail contexts, including the generation of hypersexualized imagery. This underlines the need for responsible implementation, clear user safeguards, and rigorous testing before adopting such features at scale.

The Road Ahead

As AI redefines how consumers interact with products online, fashion companies must evolve, not just in terms of tech adoption, but in mindset and internal workflows. This is not just a marketing opportunity, but a strategic inflection point: the brands that adapt their structures, teams, and processes for a digitally immersive future will be the ones that thrive.

Supporting and integrating with third party applications such as Google AI or ChatGPT is just a first step in a digital transformation process that need to create value for the customer

Enrico Fantaguzzi

However, supporting and integrating third party applications such as Google AI or ChatGPT is just a first step in a digital transformation process that need to create value for the customer. Google’s “Shopping Graph now has more than 50 billion product listings, from global retailers to local mom and pop shops, each with details like reviews, prices, color options and availability”, so that customers have more options to select from, however people don’t just buy a new product everytime they go on a trip or to a party, therefore customers should be allowed to integrate the AI features within their own wardrobe. This would allow people to integrate their current selection of garments with new products that their missing. This step is critical for user experience but also to create a sustainable digital fashion industry,

At Digital Fashion Academy, we believe that understanding and leveraging these technologies is no longer optional, it’s essential. We’ll continue to monitor these developments and support fashion professionals with the knowledge and skills needed to lead in the age of AI.

Sources

• Google Blog: AI Mode and Virtual Try-On Update
  
• The Atlantic: Google’s Shopping Tool and AI Image Ethics
  
• Google I/O 2025 Recap (for context on product announcements)

X2CB4D796ZF

Deep dive into fashion AI: join our live webinar

Here, too, the goal is to identify a north-star metric that can serve as a guiding benchmark for all initiatives aimed at optimizing this phase. As in previous discussions, it is essential to take a holistic perspective, recognizing that the e-commerce platform is just one component of a broader brand ecosystem.

The objective is not to maximize the online channel in isolation but to configure it in a way that enhances the performance of the entire ecosystem. At times, this may mean that certain elements appear to underperform — when in reality, they contribute more significantly to long-term brand success by unlocking greater systemic value. With this perspective, identifying the right KPIs becomes crucial, ensuring that all experiments and optimizations align with the broader brand strategy.

At the lower funnel stage, customers finalize their purchase, completing the transaction for products they previously explored. When it comes to placing an order, they have multiple channel options:

• Physical Stores — Traditional brick-and-mortar locations where customers can buy in person.
• Third-Party Partner Platforms — External websites or apps, such as marketplaces and concessions, that facilitate transactions.
• Customer Support — Purchases made via phone, often using a secure pay-by-link method.
• Brand Website — The brand’s primary e-commerce channel, where customers browse and buy directly.
• Brand Mobile App — A dedicated transactional app, if available, that provides a seamless shopping experience.

The first key difference between the discovery and purchase stages lies in how customers interact with channels. During product discovery, they often engage with multiple channels, taking advantage of each one’s unique benefits. However, when it comes to making a purchase, they typically commit to a single channel and complete the transaction within that environment.

This analysis specifically focuses on website performance, as it remains the primary driver of e-commerce activity. For simplicity, the term “online” will refer exclusively to the brand’s website throughout this discussion.

THE PARADOX OF A PERFECT EXPERIENCE

For most e-commerce businesses, the purchase stage is the ultimate goal — both a key revenue driver and a crucial touchpoint for customer satisfaction. A seamless, intuitive checkout experience not only boosts conversions but also fosters long-term brand loyalty.

However, for luxury brands, the role of e-commerce is fundamentally different. Their customer journey is designed to be store-centric, prioritizing high-touch service and immersive in-person experiences. As a result, defining the right north-star KPI requires careful alignment with the brand’s overarching strategy.

Traditionally, e-commerce success is measured by Conversion Rate (the percentage of website visitors who complete a purchase) and Digital Revenue (total online sales within a given timeframe). While these metrics work well for mainstream online retailers, they can be misleading for luxury brands.

Consider this scenario:

A luxury collection is so captivating that it inspires immediate desire. The website seamlessly provides all necessary details, compelling every shopper within a specific timeframe to visit a physical store to complete their purchase. Once in-store, expert advisors enhance the experience — curating additional selections, deepening brand engagement, and ultimately increasing customer spend.

Paradoxically, in this scenario, the website’s Conversion Rate would register as 0%, and Digital Revenue would appear as zero — not due to failure, but because every shopper opted for the in-store path within that period, driving significant business impact. Does this mean the e-commerce platform fell short? Absolutely not. It achieved its ultimate goal: engaging customers, facilitating discovery, and driving store visits, where the brand’s value is truly maximized.

This reveals a fundamental flaw in how luxury e-commerce performance is measured.

While a frictionless online checkout remains important — some customers lack store access, and certain products are well-suited for digital sales — it should be viewed as a secondary function, not the primary goal. For luxury brands, the true success of an e-commerce platform lies in its ability to drive in-store engagement.

North-star metrics should reflect the brand’s strategic priorities — not be based on a fallback option. To accurately assess luxury e-commerce performance, brands must look beyond traditional metrics like Conversion Rate and Digital Revenue and adopt KPIs that better align with their unique customer journey.

Traditional e-commerce metrics assume that brand websites function like Amazon — where every visitor is expected to complete a transaction online. This perspective fails to account for the distinct objectives of luxury fashion’s upper and lower funnels.

Here’s the flaw in that logic: if a customer visits a luxury brand’s website, finds exactly what they need, and is successfully guided to a physical store to complete their purchase, traditional metrics would still count this as a failed conversion. This artificially lowers the Conversion Rate and shows no contribution to Digital Revenue, creating a misleading assessment. In reality, for many luxury brands, driving in-store visits can be far more valuable than an online sale.

Beyond the multi-channel nature of the upper funnel and the single-channel flow of the lower funnel, another key distinction lies in online behavior. Product discovery is inherently nonlinear — customers move fluidly between product listing pages (PLPs), product detail pages (PDPs), and search results (SERPs), continuously refining their choices. This back-and-forth journey is shaped by inspiration, research, and evolving preferences. In contrast, checkout is a structured, goal-oriented process designed for efficiency and transaction completion. It typically follows these key steps:

• Customer Identification — Also known as the Checkout Login screen, this step can be bypassed via guest checkout but remains crucial for loyalty programs and marketing.
• Shipping Selection — Customers choose delivery methods and preferred locations.
• Packaging Options — A consideration for sustainability-conscious buyers and gift-givers.
• Payment — A range of payment methods accommodate different customer preferences.

Some of these steps can be streamlined with express checkout or pre-saved preferences, further reducing friction. However, while product discovery is an exploratory process, checkout follows a linear path toward transaction completion — a fundamental difference that should be reflected in north-star metrics.

Since product discovery is complex and non-linear, measuring its effectiveness requires a combined metric, such as engagement rate. On the other hand, the structured nature of the checkout funnel makes it better suited for traditional ratio-based metrics.

However, the standard Conversion Rate falls short in this context. While the numerator (completed transactions) is clear, determining the right denominator requires deeper analysis. What constitutes a meaningful conversion opportunity for a luxury brand? Instead of blindly applying traditional e-commerce KPIs, brands must examine user behavior patterns to develop more accurate measures of funnel efficiency — ones that reflect the true impact of their digital experience.

When analyzing users transitioning from product discovery to a purchase decision, we can identify three key categories:

1. Users who haven’t visited the website but have interacted with products through other digital channels or physical stores.
2. Users who browsed the website without adding items to the cart; still they might have shown measurable intent (e.g., clicking ‘Find in Store,’ contacting customer support) or not (taking screenshots, or saving product links).
3. Users who browsed the website and added one or more products to the cart, along with other engagement actions.

Users in all three categories may ultimately choose not to purchase, buy in-store, or complete their transaction through another channel instead of the website. However, only the third group is relevant to this analysis, as adding a product to the cart is a prerequisite for entering the website’s checkout process. This is why traditional e-commerce often considers the Add-to-Cart Rate the most reliable leading indicator of an impending transaction.

Given this, one might reasonably define the Add-to-Cart action as the entry point to the lower funnel and measure the effectiveness of this stage by the ratio of purchasers to users who added items to the cart. This approach provides a better proxy than the overall Conversion Rate, as it excludes visitors who merely browse the catalog without displaying intent to purchase via the website.

However, for many brands, the Add-to-Cart function often serves more as a wishlist than a direct purchase intent signal. The final transaction — whether online or in-store — is driven more by customer preference than by the efficiency of the purchase funnel. It’s common for customers to add products to their cart simply as a way to bookmark them, later exploring and purchasing in a physical store.

This makes an Add-to-Cart-based north-star metric vulnerable to the same fundamental flaw as the Conversion Rate: it misclassifies users who engaged in product discovery, added items to their cart, and ultimately completed their purchase in-store as failed conversions.

In reality, the Add-to-Cart action is a stronger indicator of successful online product discovery than of an imminent online purchase. From a holistic perspective, these customers have followed a highly successful and preferred journey, emphasizing the need for a broader evaluation beyond traditional e-commerce metrics.

To eliminate this bias, it is essential to identify a funnel stage that is reached primarily by customers who intend to complete their purchase online rather than in-store. This point should come after the Add-to-Cart action but may vary depending on the website’s flow.

In many cases, based on the checkout process flow outlined above, the best approximation for measuring true conversion intent is the Customer Identification step, also known as the Checkout Login screen. Reaching this stage indicates a clear intent to complete the transaction online, making it a more reliable benchmark for assessing conversion potential.

As a result, one of the most effective north-star metrics for evaluating purchase funnel efficiency is Checkout Conversion Rate — the proportion of customers who successfully place an order after reaching the Checkout Login screen. This metric provides a clearer picture of checkout performance and offers valuable insights for optimization.

At this stage, customers are prompted to log into an existing account, create a new one, or continue as a guest. Encouraging account registration is particularly important for loyalty-driven CRM initiatives, which is why brands often highlight benefits such as easy access to order status to incentivize registered purchases.

While most customers encounter the Customer Identification step, some checkout flows bypass it. For example, registered users who are already logged in will skip this step when entering the purchase funnel, and those using express payment methods may follow a different flow entirely.

To ensure accurate metric calculations, brands must account for these exceptions. The equivalent step in alternative checkout paths should be included in the denominator to accurately reflect the number of users reaching this stage, while all successful transactions — regardless of the path taken — should be counted in the numerator. This approach ensures a comprehensive and precise measurement of checkout conversion rate and overall funnel efficiency.

While the checkout login screen is chosen as the point reached only by customers intending to complete their purchase online, a potential concern with this approach is that some users may enter this stage solely to gather information that influences their decision. Many e-commerce websites observe users progressing into the purchase funnel not necessarily to buy immediately, but to check delivery options, shipping times, packaging details, final costs including taxes, or accepted payment methods.

Although these behaviors may lower the Checkout Conversion Rate, they present an opportunity to improve the product discovery funnel. The upper funnel’s role is not just to help customers find products that match their preferences in terms of shape, material, or color, but also to address broader purchasing considerations. For example, if a customer needs an item by a specific date, delivery timelines and availability should be clearly communicated upfront, without requiring them to enter the checkout process unnecessarily.

In other words, optimizing lower funnel performance may require a more systemic approach that enhances clarity and accessibility across the entire website, ensuring that critical purchase-related information is available at the right stage of the journey. The traffic reaching the checkout should be highly qualified, meaning that customers enter the purchase process with clear expectations and encounter no unexpected issues. This is even more crucial in e-commerce than in physical stores, where in-person interactions can help resolve last-minute uncertainties.

Other times, analysis may reveal that improving the north-star KPI requires removing frictions within the checkout process itself.

However, even when customer journeys successfully result in an order, the purchase funnel provides valuable insights from a broader perspective. Given that online purchases are a secondary goal compared to in-store transactions, every completed order on the website can also be viewed as a missed opportunity for store conversion. By analyzing these transactions, brands can gain insights into which products — and, to some extent, which customer segments, such as those in specific shipping locations — chose to purchase online instead of visiting a store.

In the ideal world envisioned by luxury brands, online checkout would be unnecessary. Customers would explore products online but ultimately visit physical stores, where they could enjoy a superior shopping experience while allowing brands to build deeper, more valuable relationships with them.

However, this vision is not entirely realistic. While the primary goal remains to drive as many customers as possible to stores, it is equally important to provide a seamless and effective online purchase experience. When in-store visits are not an option, ensuring that customers complete their purchases through e-commerce becomes the secondary goal.

Traditional e-commerce metrics, such as Digital Revenues and Conversion Rate, primarily measure the effectiveness of online sales but fail to capture the full complexity of customer journeys that blend both digital and in-store interactions — an essential consideration for luxury brands.

The Checkout Conversion Rate, by attempting to filter out users who engage with the website purely for product discovery, offers a more refined and relevant metric for assessing lower funnel performance. As a guiding north star, it helps shape improvement plans that enhance the online purchasing experience. However, these plans may also require optimizations in the upper funnel to ensure that high-quality traffic flows through the entire purchase journey.

In the next article, we will consolidate insights from both the upper and lower funnel to present a holistic, end-to-end view of e-commerce’s role in the luxury brand customer journey.

M.P.

One significant risk of an approach that relies solely on these metrics is that all costs associated with the online presence are allocated to the e-commerce Profit & Loss (P&L). Meanwhile, a considerable portion of in-store revenue is driven by customers who initially engage with the website, yet these sales are not reflected in the online P&L. This disconnect can distort the true value of the online channel, undermining its role in driving both digital and physical sales.

To address this complexity, a more holistic approach is needed — one that analyzes the customer journey across all channels, with a special emphasis on online interactions. This journey can be visualized as a funnel. The following analysis focuses on the upper part of the funnel, often identified as the product discovery phase.

PRODUCT DISCOVERY CHANNELS

During this stage, customers explore the brand’s product catalog. Success occurs when customers identify products that align closely with their needs, prompting them to move closer to purchasing.

Product discovery takes place through multiple channels:

• Physical Stores: Visiting a store in person or browsing its window displays.
• Offline Marketing: Channels like billboards, magazines, public transportation ads, and similar media.
• Brand Website: A central touchpoint where many e-commerce journeys begin.
• Brand Mobile App: Delivering a personalized, on-the-go shopping experience.
• 3rd Parties Partner Platforms: Third-party websites or apps, such as concessions and marketplaces.
• Other Online Channels: Social networks, digital ads, and other web-based platforms.

In luxury fashion, where careful consideration is a key part of the purchasing process, multiple channels are often involved for a single customer. According to the RACE framework for marketing, some channels align with the Reach (or Awareness) stage, while others support the Act (or Consideration) stage. However, all channels contribute to helping customers identify and select the right products.

It’s important to clarify that, while several channels operate online, this analysis focuses specifically on website performance, as it is the primary driver of e-commerce activity. For simplicity, the term “online” will be used throughout to refer exclusively to the website.

CHALLENGES WITH TRACKING

At first glance, tracking might seem like the solution to understanding how the brand’s website contributes to overall business performance. It offers the potential to seamlessly follow customers across the website, stores, and other touchpoints, such as mobile apps, to build a comprehensive view of their journey and revenue impact. However, this approach comes with significant challenges, particularly for luxury brands.

In the early stages of the customer journey, it is crucial to maintain the high-end experience by minimizing intrusive tracking methods. Preserving the brand’s exclusivity and sophistication is essential to building trust and loyalty. As the relationship develops and a personal connection is established — often through a client advisor — tracking becomes both more practical and less invasive.

Additionally, the luxury sector’s distinct characteristics present further complexities. Aspirational purchases, gifting, and the high-value nature of products mean that luxury brands often deal with a higher proportion of first-time customers. This reduces the effectiveness of conventional tracking strategies and CRM tools typically used in more transactional businesses.

In the absence of precise tracking, it becomes crucial to identify the most effective leading indicators of successful online product discovery. In other words, which KPIs specific to this stage of the customer journey can be monitored to predict future conversions, regardless of whether the transaction ultimately occurs online or in-store?

PRE-PURCHASE BEHAVIORS

When analyzing website visitor behavior, two distinct patterns emerge. Some users show little interest in exploring available products, exhibiting high bounce rates or minimal interaction. Others are more engaged, spending time navigating search results (SERPs), product listing pages (PLPs), and ultimately product detail pages (PDPs). Among those considering a purchase, users typically fall into one of three categories:

1. A2C (Add-to-Cart) CTA (Call-to-Action): These users clearly demonstrate intent to purchase by adding a product to their cart.
2. Other CTAs: These users interact with alternative buttons, such as checking product availability in physical stores, contacting customer support, or adding items to their wishlist. While not directly adding to the cart, these actions indicate interest and movement toward a potential purchase.
3. No CTA: These users do not engage with any buttons on the PDP but may still be considering a purchase. Their actions might include taking screenshots of the PDP, saving the product link, or doing nothing explicit. While there’s no visible interaction, these users may return to the product later, perhaps during an in-store visit or after consulting a client advisor.

Customers who intend to make an online purchase typically fall into the first category, as the Add-to-Cart action is required to enter the website’s checkout process. For this reason, traditional e-commerce often views the A2C Rate as the most reliable leading indicator of an upcoming transaction.

However, research shows that the A2C action is often used to save products for later rather than to proceed immediately with a purchase. According to Baymard, 42% of customers who want to save a product for later consideration go through the A2C button. This makes A2C an unreliable signal of a confirmed transaction. The challenge is even greater in luxury markets, where many in-store transactions occur without a prior A2C action online. In addition, for luxury brands aiming to drive as many transactions as possible in-store, the A2C step is optional in the ideal customer journey they envision.
Consequently, relying on A2C as a primary performance indicator in this industry can be even more misleading.

Referring back to our earlier discussion, it becomes clear that optimizing the product discovery journey solely to maximize the A2C rate risks undermining overall business performance. While online metrics may improve, in-store transactions could suffer — potentially outweighing the online gains. In systems thinking terms, a high A2C rate might represent a local maximum, whereas the true goal is to achieve a global maximum that benefits the entire business.

From a tracking perspective, the first two user categories (A2C and other CTA interactions) provide actionable data. However, the third category, which plays an important role as a bridge between online and offline experiences, occurs without detectable interaction, and therefore it’s neither trackable nor measurable.

All these considerations highlight why the step marking the closure of the upper funnel, just before customers enter the purchase journey, cannot reliably indicate the success of a product discovery activity. Even when expanding beyond the single A2C metric to include alternative behaviors, this stage remains insufficient as a comprehensive measure.

To develop a meaningful metric for upper funnel success and a strong predictor of future revenues, it is essential to step back and examine how customers discover the products they eventually choose to buy. A deeper understanding of this discovery process is key to accurately evaluating and optimizing the product discovery journey.

INTRODUCING THE ENGAGEMENT RATE

Healthy activity in the website’s product discovery area — PDPs, PLPs, and SERPs — is the best signal for upcoming purchases.
As in a mall or a supermarket, the volume of the crowd, the time spent in the lanes or in front of the windows, and the interaction with the products on the shelves indicate healthy conditions for those businesses. Similar observations can be considered for a brand website.
In a single term, Engagement is the behavior to measure, and the associated metric should be used as the north star metric for the product discovery stage of the funnel.

This approach aligns with Google’s recent updates to its Analytics product, which introduced more advanced methods for measuring website activity. Among these updates is a new metric called Engagement Rate, which goes beyond basic metrics like button click ratios. It aims to capture the qualitative aspects of user interactions, such as time on site, actions taken, and deeper navigation patterns, offering a richer understanding of genuine user engagement.

In GA4, a session is classified as “engaged” if it meets at least one of the following criteria:

• The session lasts longer than 10 seconds.
• The session includes at least one conversion event.
• The session involves two or more pageviews or screenviews.

A high Engagement Rate indicates that users find the content valuable and are actively engaging with the site or app. Companies can leverage this metric to optimize user journeys and improve conversion rates.

Furthermore, the individual metrics that contribute to the Engagement Rate definition — such as time spent on the site or the number of pages viewed — can be further analyzed for deeper insights.

Brands should establish a tailored definition of Engagement Rate, either adopting Google’s standard or customizing it to reflect better the unique characteristics of the online product discovery journey for luxury goods. For instance, users interested in luxury products tend to engage for longer periods than the typical duration suggested by Google.

Once defined, this customized metric should act as a guiding north star for upper-funnel activities, informing all experiments related to Conversion Rate Optimization (CRO). As customer behaviors evolve, the definition of this metric should be revisited and refined. The underlying assumption is that a high Engagement Rate signals the brand’s success in capturing and maintaining customer interest, encouraging deeper exploration of the catalog, and ultimately increasing the likelihood of future purchases, whether online or in-store.

IS LONG TIME SPENT ON SITE ALWAYS A POSITIVE INDICATOR?

A common critique of using metrics like time spent on site as an engagement indicator is that longer durations could imply either customers are actively exploring the catalog and evaluating products they plan to purchase or, conversely, that the site lacks relevance, leading to frustration and difficulty in finding desired products. This distinction is why, when introducing the Engagement Rate, the focus was placed on defining the activity as “healthy” to ensure the metric accurately reflects meaningful user engagement.

There are several ways to address these concerns. One effective approach, particularly in this context, is to use a mutually destructive pair of metrics: alongside the Engagement Rate, another metric can be tracked to ensure that customers are deriving real value from the discovery process. This secondary metric, often referred to as a health metric, is not intended for primary optimization but serves as a safeguard. For instance, brands might monitor the A2C Rate or interaction rates with CTAs within the PDP, establishing a threshold to detect a potential drop in engagement.

Another approach is to identify key elements that ensure the intrinsic health of engagement. For luxury brands, a positive customer journey typically involves significant interaction with product galleries, particularly images. This interaction rate can be integrated into the custom definition of Engagement Rate, providing a more accurate measure of genuine engagement.

It’s important to note that the Engagement Rate, as defined above, does not provide an absolute measure of the quality of the product discovery segment in the funnel. Given that different brands may have varying definitions of this metric, it’s not easily applicable as a benchmark against industry standards or competitors. Instead, its value lies in tracking changes over time — specifically, comparing the metric before and after a particular experiment or website update. Drawing from the Toyota Kata approach covered in the previous discussion, the Engagement Rate should act as a measurement system that captures trends and ensures that changes in the website are consistently moving toward the vision.

WRAPPING IT UP

While traditional e-commerce can effectively rely on the Add-to-Cart Rate to measure the performance of the product discovery stage, brands that favour an omnichannel experience cannot rely on it as a sole metric. This is because it doesn’t provide a comprehensive view of customer journeys that span both online and in-store experiences, which are crucial for luxury brands.

The Engagement Rate offers a more sophisticated and relevant metric for assessing the health of the upper funnel, acting as a guiding north star for improvement plans. However, defining and applying this metric requires careful consideration and customization to align with the brand’s specific strategies.

Optimizing the product discovery phase of the website based on the Engagement Rate can deliver holistic benefits, regardless of where the customer ultimately decides to make the purchase — whether autonomously online or by visiting a physical store.

In the next article, we will explore the lower stage of the e-commerce funnel, examining the key differences and commonalities in dynamics between luxury brands and transactional or traditional businesses at both ends of the funnel.

M.P.

For luxury fashion brands, traditional e-commerce KPIs — such as Add-to-Cart Rate, Conversion Rate, and Digital Revenues — are typically adapted from transactional models. This KPI framework works indeed well for companies that are e-commerce-driven. For example, Amazon is built around selling physical products, but its business is fundamentally driven by e-commerce, with KPIs focused on online transactions. On the other hand, platforms like Udemy and Epic Games, which sell digital products such as courses and games, rely entirely on online sales without the need for physical inventory or logistics.

For fashion brands like Zara or MyTheresa, where the in-store experience is less central or the physical presence is minimal, these KPIs are also effective.

However, luxury brands present a more complex picture. For them, e-commerce is just one part of an integrated, immersive brand experience, and conventional KPIs may not fully capture the richness of customer interactions, brand perception, or loyalty — key elements of success in high-end retail.

UNIQUE ASPECTS OF CUSTOMER JOURNEYS

Several factors uniquely shape the luxury customer journey. First, competition among luxury brands is far less intense compared to the mass-market segment. When a customer chooses a specific handbag from a brand, a slightly subpar online experience is unlikely to drive towards another brand. The switching barrier is much higher than in transactional businesses, where seamless online interactions often dictate customer loyalty.

Moreover, luxury brands have long been anchored in the in-store experience. They are neither web-first nor web-only businesses. E-commerce was introduced not to replace but to complement their brick-and-mortar presence, enhancing overall performance rather than serving as the primary sales channel.

It’s also essential to recognize that these brands don’t sell digital products; their physical products are meant to be part of a larger experiential journey. The nature of luxury goods makes the in-person element particularly crucial. Customers want to feel the fabric, admire the craftsmanship, and immerse themselves in the brand’s ambiance. These tactile interactions are a vital part of the decision-making process, much like test-driving a car before purchase. In this context, a luxury purchase transcends a simple transaction — it becomes an emotional and immersive experience.

A central figure in delivering this immersive experience is the client advisor, whose role extends far beyond facilitating a purchase. Acting as brand ambassadors, client advisors are instrumental in establishing meaningful relationships with customers, understanding their unique preferences, and providing personalized recommendations. Through attentive and knowledgeable guidance, they help transform each visit into a memorable experience, fostering loyalty and encouraging repeat visits. In luxury retail, the client advisor’s expertise and attentiveness are not just service enhancements; they are pivotal to achieving lasting success.

Finally, the high price point of the goods often makes customers hesitant to buy online without first experiencing the product in person. They want to ensure that key aspects such as size, color, and material meet their expectations, while also gaining confidence in the overall quality and feel of the item.

Taken together, these factors make it clear that in the luxury industry, online and offline experiences are deeply intertwined, making it difficult to assess e-commerce performance in isolation with conventional KPIs.

IS “E-COMMERCE” THE RIGHT WORDING?

These observations highlight that customers often prefer to include a store visit as part of their purchase decision-making journey. Brands, in turn, actively encourage this by steering customers toward the in-store experience, where deeper, long-term relationships can be cultivated, rather than prioritizing a quick online sale.

Transaction data underscores the importance of in-store engagement: despite offering the same products, metrics like Units per Transaction, Average Unit Retail, and, consequently, Average Order Value are consistently higher in stores, while Return Rates are lower. This highlights how online and offline channels play distinct, complementary roles, creating a seamless, integrated luxury shopping experience.

Given the close integration of online and offline experiences, it’s worth reconsidering whether “e-commerce” is even the right term. A more fitting description might be “online presence”, as digital channels in luxury retail aim to complement, rather than replace, the immersive in-store experience that defines the brand.

E-COMMERCE KPIs AND SYSTEM THINKING

When revisiting the discussion of primary e-commerce KPIs, it becomes evident why traditional online-centric metrics fall short in capturing the full complexity of the luxury brand ecosystem. While metrics like Add-to-Cart Rate, Conversion Rate, and Digital Revenues can still provide useful insights in certain contexts, they are not suitable as “north star” metrics for a strategy that emphasizes the offline experience.

A well-regarded approach to optimizing system performance is the Toyota Kata method, which focuses on assessing the current state and conducting iterative experiments to gradually move toward target conditions aligned with the broader vision. The success of this approach depends on a measurement system that captures trends and ensures that changes are consistently moving in the right direction.

Take the example of a racing car: reducing the car’s weight might seem like an easy win for increasing speed, but if the measurement system focuses solely on weight as a leading indicator for velocity, replacing the engine with a smaller one might be viewed as an improvement. However, this could result in a slower car due to reduced power.

Similarly, adopting e-commerce Conversion Rate as a “north star” metric poses the risk of optimizing for a local maximum. Enhancing the online experience could inadvertently draw attention away from store visits, leading to diminished overall business performance. While improved digital metrics may align with better company results in some cases, this is not guaranteed. To use a fitting geographical analogy, relying on such a metric is like navigating with a compass that occasionally points north and occasionally doesn’t — hardly the kind of tool you’d trust to guide your journey.

WRAPPING IT UP

At this stage of the discussion, a crucial question emerges: if the traditional e-commerce KPI triplet falls short, what is the better alternative? Unfortunately, there’s no simple answer. The challenge with conventional KPIs is that they attempt to isolate the digital segment, whereas for luxury brands, the online and offline experiences are deeply intertwined. Any improved approach must offer a holistic perspective, though this inherently makes such KPIs more complex to calculate than straightforward metrics like the online Add-to-Cart Rate.

In the next article, we’ll take a deeper dive into the e-commerce funnel and introduce new indicators that are better suited for a comprehensive analysis.

The good news is that while the north star metric should be holistic, the action plan can still focus on specific areas. However, for optimal results, the strategy will likely need to integrate digital actions, in-store initiatives, and a combination of both.

M.P.

Maria, Social Media and Content Creator

How do the big fashion and luxury brands guarantee an optimal User Experience for users and customers? It all starts from a solid testing strategy

In recent years the Fashion & Luxury sector has undergone a significant revolution, triggered by the adoption of digital technologies, by the change in
behaviour and habits of consumers and of ever-increasing centrality of digital experiences.

The growth of e-commerce in the sector is rapid and constant. For example according to Statista, fashion ecommerce is expected to grow by 16% in Italy and by 11% in Spain from 2023 and 2027.

These numbers contain a series of challenges and opportunities for brands and industry professionals.
Always characterized by strong competitiveness and the importance of experience and of customer service, fashion & luxury today finds itself having to adapt to the rhythms and necessities of digital processes. Values such as excellence, luxury, uniqueness and exclusivity have always had a central importance in the boutiques of the global brands of luxury; today the challenge is to be able to transfer them to online shops and digital experiences of all the world.

Companies today must be able to adapt to the changing needs of customers, e.g. aspire to offer the same level of excellence and exclusivity in their online stores that customers live in boutiques. And here is the first and most important challenge: how to achieve these levels of excellence for e-commerce?

The importance of testing for fashion & luxury e-commerce

Creating a flawless experience for online customers in the fashion industry it goes beyond the visual appearance of a website. While companies invest considerably in the design of their website and in the care of the brand image, there are factors not immediately visible that can compromise both the image and functionality of the sites. Furthermore, fashion e-commerce have to face a series of specific challenges related to their digital nature:

• the request for continuous updates, both of the catalog and of the platforms;
• the international (and often global) dimension of e-commerce;
• the lack of direct feedback from customers, which is instead present in physical store

The impact of technology changes on user experience

The continuous updates of the e-commerce platforms, the changes in content and the implementation of new features on sites are frequent causes of links not working, 404 errors, problems viewing sites on mobile devices, broken payment systems and so on.

A complete and effective response to this need comes from methodologies of testing.

Testing methodologies as a solution to user experience problems

The practice of testing for e-commerce – not only in the world of fashion – is established in recent years as one of the most solid ways to guarantee a level of a satisfying experience for the demanding fashion & luxury clientele.

The tests are carried out both before the launch of an e-commerce (for example on a navigable prototype, or in a staging environment) both later, with the site already active and public, coinciding with the release of a new version, of an important update or within a testing strategy that involves recurring cycles of testing and maintenance.

These are tests that are carried out by both expert and qualified personnel – Quality Assurance professionals, with skills relevant to the world of development and programming – both with the involvement of real users and they test e-commerce in real contexts and with real devices, simulating navigation and the purchase.

What problems does website and e-commerce testing help solve?

This type of activity allows you to highlight operational defects that would otherwise not be possible visible, but which have a profound impact on crucial aspects:

• Site Performance: A website that loads slowly or has technical errors can frustrate visitors, regardless of beauty of design. Accurate tests must be conducted to ensure that the site is responsive and fast on a variety of devices and internet connections.
• Browser and device compatibility: Visitors can use a wide range of browsers and devices, both mobile and desktop, for navigate. A design that looks great on a browser might have problems on another. Tests must cover different platforms for ensure uniform viewing.
• Security and Privacy: With the growing concern about security online, ensuring the protection of customers’ personal information is essential. Safety tests must be carried out to identify and resolve any vulnerabilities.
• User Journey: An attractive design can lose its meaning if the Site navigation is complex or the purchasing process is short intuitive. User journey tests are crucial to optimizing customer journey.
• Bugs and operational defects: Broken links, pages that don’t open, filters that do not work properly. These are all aspects that have an impact negatively impacting the browsing and purchasing experience not on user satisfaction, but also on metrics business.

To better understand the importance of these aspects, let’s consider a customer who visit a fashion site. If the site loads slowly, the customer may abandon the experience, negatively influencing the overall impression of the brand (the 60% of visitors expect the site to load in less than 2 seconds, otherwise they leave page). Also, if the site is not compatible with the customer’s preferred browser could generate frustration, undermining trust in the reliability of the site and therefore of the brand.

Security is another critical aspect: imagine if a customer were to encounter security issues during the payment process. This might not only damage the reputation of the brand but could also lead to loss of customers.

Finally, a complicated user interface can lead to a high abandonment rate cart, although the design is aesthetically pleasing. Test and optimize the The user journey is therefore crucial to maximizing conversions.

Testing of e-commerce in fashion & luxury: ROI and metrics from keep an eye on

Evaluating the Return on Investment (ROI) of Quality Assurance is one step crucial to determining the effectiveness of dedicated efforts. When measuring ROI, several key factors are taken into consideration:

• Bug Reduction is one of the main indicators. Calculate the number e the impact of fixed bugs after implementing QA practices provides a clear metric of effectiveness in improving stability e the reliability of the system.
• A second important element is the Increase in Conversions. Monitor from close how many additional conversions occur after user experience (UX) optimization allows you to establish a direct link between the QA activities and commercial success. Under the acronym CRO (Conversion Rate Optimization) all those practices and methodologies are collected, across the board between User Experience Design, research and programming, which allow you to increase the conversion rates of a page or website.
• Furthermore, it is essential to consider Positive Customer Feedback. Collect and analyzing the positive reviews, along with customer satisfaction, offers a direct indication of how Quality Assurance is contributing to the appreciation of end users. In this case, metrics like CSAT (Customer Satisfaction) and NPS (Net Promoter Score) are essential to track good QA and UX optimization work.
• Last but not least, the saving in terms of time for the development team and Quality Assurance – considering the average cost that such qualified professionals have within a company – is another fundamental element for measuring the positive impact of a strategy testing.

In summary, measuring QA ROI is not limited to quantitative metrics, but encompasses also qualitative indicators that reflect the overall customer experience and the value added by quality control activities.

Functional and experiential tests

There are numerous types of Quality Assurance tests, which verify just as many various aspects of e-commerce functionality. The most frequent and useful ones in the world of luxury are:

• Copy testing: Ensures correct display and contextualization of text information, in different languages. Often, for large e-commerce sites, it is created on a global scale by specialized companies (for example with the methodology of crowdtesting), which involves involving native speaker testers.
• Payment testing: Verify that payment methods are integrated correctly and work smoothly.
• Device compatibility testing: Ensures the site fits and works correctly on a wide range of devices and operating systems.
• Checkout testing: Optimize the checkout process to maximize conversions.
• Non-regression testing: a type of test (actually a real one methodology) created by professional testers, which allows testing the features of a site introduced with a new release, without having to re-test the “core” elements of the architecture.

To these tests, which are defined as functional in that they allow the correct functionality of a site to be verified, are added experiential tests, which they have a profoundly different nature, methodology and objectives. Indeed they come usually created with the help of a team of User Researchers, professional figures with specific skills, which deal with analyzing interactions between users (who also in this case are testers) and the e-commerce or site under consideration. This type of test aims to highlight usability problems of the site, which technically they are defined as clutches.

In the world of fashion & luxury the most relevant experiential tests are:

• End-to-End Testing of the Purchase Journey: Covering the entire journey online purchase, these tests ensure that every phase, from navigation to payment until return, is optimized for a customer experience without hitches.
• Usability testing: These tests allow you to create an engaging experience for eCommerce customers. Through consolidated methodologies and analysis in-depth, they focus on optimizing brand touchpoints to maximize the impact of the customer experience. They aim to optimize the UX, improve conversion rates and reduce cart abandonment.
• Phygital experience test: With this type of test, that is verified purchasing path, increasingly frequent in large e-commerce sites brand, which starts with online browsing but then ends with collection boutiques.

To carry out these types of testing it is essential to have access to a pool of tester that has characteristics as close as possible to those of the end user.

Testing for fashion & luxury sites: UNGUESS solutions

There are several companies that offer very advanced testing services, with specific skills and expertise in the Fashion & Luxury field. One of them is UNGUESS, a crowdtesting platform founded in Italy in 2015, which among its own success stories presents the one with Canali. This renowned fashion house men needed to refine and optimize the online shopping experience in various international markets. UNGUESS played a crucial role in identifying and resolving challenges that limited the customer experience on the site Canali e-commerce.

Methodology:

• Bug Hunting: Detailed testing has been carried out to locate and resolve bugs and defects in the functioning of the Canali website. This process has ensured that ecommerce was efficient and reliable, improving quality overall shopping experience.
• Analysis of the purchase and return experience: UNGUESS conducted an in-depth analysis of each phase of the online purchasing journey on different platforms. From navigation processes to product selection, from payment to returns management, every aspect has been examined and optimized to ensure a smooth customer journey
• Evaluations of texts and translations: The accuracy and effectiveness of Product descriptions, as well as translations into several languages, have been carefully evaluated to ensure a clear and consistent message for customers of different nationalities.

The testing project carried out by UNGUESS involved 6 different countries (Italy, Germany, France, Spain, United States, United Kingdom) supporting Channels in obtaining an international dimension for your e-Commerce.

Through the work carried out by UNGUESS, Canali has seen improvements significant in the overall user experience on your e-commerce site.

The collaboration between Canali and UNGUESS has demonstrated the fundamental value of
thorough testing, direct customer feedback and constant optimization to ensure an impeccable online shopping experience.

Download the case Study in Italian >>

Download the case study in English >>

The impact of an experiential test: the Decathlon Case Study

If a good testing strategy involves the combined iteration of functional tests and experiential – in order to guarantee perfect functioning of e-commerce, together with offering a satisfactory User Experience – it is indeed possible take care of only one of the two aspects, for example if one of the two parts is already managed through internal processes and activities.

This is the case described in the UNGUESS Case Study with Decathlon. Decathlon Italia turned to UNGUESS to make its online experience smooth and memorable users, at a time when – following the pandemic – the institutional site has registered. It was a digital infrastructure that was already perfectly functional and effective, but still lacking external, objective and objective validation.

This approach, guaranteed by UNGUESS, allows you to obtain reliable insights from translate into practical and actionable actions on the structure of the online store.

Decathlon initiated the tests with the aim of optimizing the shopping experience on its eCommerce. In particular, the work of the UNGUESS team and the community was aimed at understanding the real user’s perception of the experience of exploration of the catalog and understand if the navigation and information are clear enough to allow the user to find what they are looking for and enter the desired products in the cart.

A pool of testers was then built within the UNGUESS crowd, with
characteristics and purchasing habits targeted by Decathlon, with a series of tasks very specific to be created on the site, both desktop and mobile:

• homepage display
• product search using filters and guided search
• interaction with the product sheet
• cart navigation
• checkout and registration

The product purchases have actually been completed, so that the experience was as natural and realistic as possible.

Once the paths to be tested had been established, the methodology applied was that of Thinking Aloud. This method involves asking users to think aloud high as they interface with the test subject, explaining their movements, actions, impressions and difficulties. Participants independently record a video while they carry out the test steps with their own devices in their real-world environment.

Knowing the sensations and blocks that real users have encountered, it is possible overcome all the assumptions of the decision-making process and put into action corrections based on objective data produced by end users. The recommendations generated by this experiential test – which was also carried out in just 3 days – have allowed Decathlon to direct design choices, increasing Conversion 15% installments in the year following implementations.

Access the case study in Italian >>

Access case study in English >>

Prenatal

The example of Prenatal is no less relevant. Once developed by an agency external, Prénatal eCommerce has been tested by the UNGUESS community in pre-release phase.
The engagement of real users, as in the case of this test, indicates a high level of engagement attention to the customer and a user-centric Prénatal strategy.

The community’s work was aimed at: reporting bugs on the Prénatal site in a test environment starting from the homepage until the purchase is completed (with the use of gift cards and choice of payment method); identify malfunctions that are annoying to users or that block the user from completing of your purchase to ensure a smooth and enjoyable experience.

Thanks to tests carried out with UNGUESS, Prenatal identified 138 unique bugs, of which 26% malfunction bugs – which affect the correct functional flows of the site – e 35% usability issues during checkout.

Download case study in Italian >>

Download case study in English >>