Katrina Watson & Farida Khalaf: Author Interview #20, AI Everywhere, Volume 1

AI Everywhere, Vol. 1, Ch.18: AI in Retail and e-Commerce. How AI levels the playing field for small business and why transparent, ethical personalization is the key to building customer trust

Welcome back to the She Writes AI interview series, where we talk with the authors behind AI Everywhere, Volume 1 exploring their personal journeys and professional expertise to understand how we can build a future for AI that is truly ethical, responsible, and human-centered.

This is the 20th episode in our series, and today I am joined by Katrina Watson. Representing the Business section of our book, her Chapter 18: AI in Retail and e-Commerce addresses a critical shift in the digital economy. Katrina launched her first company at 22 with a simple goal: to make the world better one website at a time. Based in Australia, she has dedicated her career to helping “underdogs”, the small and medium businesses, compete against corporate giants who have long used AI to dominate the market at “insane rates”.

Katrina brings a practical, “leveling the playing field” lens to this conversation. She argues that we have reached a turning point where generative AI tools are finally making enterprise-level capabilities, like personalization and demand forecasting, affordable for everyone. Her chapter traces the history of retail AI from Amazon’s early recommendation engines to the modern “Target lesson” on data privacy, revealing that the real advantage for small businesses isn’t just speed or price, but the ability to use AI to enhance human-centered service.

In this conversation, Katrina Watson explores the fine line between being helpful and being “creepy” with customer data; why “naming the rule” is better than letting an algorithm run silently in pricing; and how AI can be a powerful tool for sustainability by reducing the massive environmental cost of returns. Whether you are a small business owner looking to compete or a consumer concerned about data surveillance, Katrina’s insights offer a roadmap for a retail future built on transparency and purpose.

buy the book: shewritesai.org/buy

---

Below is Katrina Watson’s full interview response to Farida Khalaf’s questions on AI in Retail and e-Commerce from AI Everywhere, Volume 1.

Topic: AI in Retail and e-Commerce (Chapter 18)

Intended Audience: For small business owners, e-commerce entrepreneurs, and retail leaders seeking a strategic framework to integrate AI tools affordably, and for consumers interested in understanding the privacy and ethical boundaries of the modern shopping experience.

Part 1: The Professional Path & The Underdog Mission

The Professional “Why”: Katrina Watson, you launched your first company at 22 with the goal of “making the world better one website at a time.” What was the spark that shifted your focus from individual websites to helping the “underdogs”, small businesses, compete against corporate giants using AI?

Katrina: My desire to help small businesses started before I started my first company.

A small business owner walked into the company I was working for. She had a good business, and didn’t need anything complicated; she just wanted a better website so she could improve her online presence. But it was a niche boutique agency that specialized in websites for research organizations and everything we offered was far too complicated and expensive for what she needed.

That night, after a quick google, I realized there was a huge gap in the market. Good, affordable websites for small businesses simply didn’t exist locally. Everything was either really expensive or poorly designed.

So a few weeks later, I launched my own agency, and spent the next nine years trying to help as many small businesses as I could. But it never felt like enough.

The gap between what big companies had and what small businesses could afford wasn’t closing, it was widening.

So I pivoted into tech startups helping small businesses, where I could try and help more of the people who needed it most.

The Global Tribe: You mention that in Australia, you love helping the underdogs. As an author in this global collective, why do you believe that change is finally coming for small businesses who felt “crushed” by the corporate giants’ early dominance of AI?

Katrina: Historically powerful data-driven technologies were so expensive that they were really only accessible to well-resourced corporations.

There was a big difference between what a company like Amazon could build and what a small retailer could afford.

But now a new category of AI, generative AI, has emerged, and that barrier to entry is much lower. Capabilities that would have cost millions to build just a few years ago are now available as off-the-shelf tools.

Technology has always been a great leveler when it becomes accessible. Small businesses can now access the same quality of personalization, pricing intelligence, and demand forecasting that the giants spent years building. Which fundamentally changes the game.

Part 2: The History & Paradox of Personalization

The Amazon Legacy: You trace AI in retail back 25 years to Amazon’s "Customers who bought this also bought..." feature. Why do you argue that this feature succeeded not just because of data, but because it created trust by solving a real customer problem?

Katrina: The reason Amazon’s recommendation engine worked was because it was solving a genuinely frustrating customer problem. Before that feature existed, if you finished a book you loved, figuring out what to read next meant hours of research, asking friends, or browsing aimlessly.

Amazon’s “Customers who bought this also bought…” solved that problem, drawing on real purchasing behavior of people. The suggestions didn’t feel random or generic, they felt relevant, because they actually were. Put simply, Amazon found a way to digitize word of mouth. And people trust word of mouth more than technology and sales. Customers didn’t need to understand the algorithm to experience the result. If the recommendations were good, trust followed.

We’re seeing the same thing right now with AI. AI can earn (or erode) trust, depending on how well it solves a problem, not by being technically impressive.

The “Creepiness” Factor: You share the famous 2012 Target case study where a father discovered his daughter was pregnant through mailed coupons. What is the “Target Lesson” for retailers today on the fine line between being helpful and being “spooky” or invasive?

Katrina: Target used personal data to surface something the customer hadn’t chosen to share, effectively crossing a line and going from helpful personalization to invasive. The father hadn’t told Target his daughter was pregnant. Target had inferred it and acted on that inference, which felt like surveillance.

That lesson is still just as true today as it was back then, and that is that businesses should only collect data they genuinely need, and only use it in ways the customer would consider fair and relevant.

Personalization should help customers find something they actually want, without feeling like a breach of privacy. The moment it starts to feel like a company knows more about you than you’ve offered them, it stops being helpful and starts being unsettling. Getting that right is less a technology question and more an ethical one, but is something that every business who collects data should consider.

Personalization-as-a-Service: In the past, companies spent millions to build recommendation engines. How has the rise of “personalization-as-a-service” platforms leveled the playing field for a retailer with a small budget?

Katrina: Building a recommendation engine used to require significant engineering investment, one that only well-resourced companies could justify. But that’s no longer the case. Platforms like Dynamic Yield have made enterprise-quality personalization available as a service, which means smaller retailers can now access the same caliber of tooling that brands such as Skims, On, and Lacoste are using, without needing to build anything from scratch.

Small business owners no longer have to manually segment their customer list and write multiple versions of the same newsletter. AI-powered software solutions can handle personalization automatically, drawing on browsing behavior, purchase history, and patterns across similar customers.

Research suggests that personalized experiences help businesses grow 2.5 times faster than those that don’t offer them. For a small retailer competing on tighter margins, that can make a huge difference, and the ROI is now much more attractive.

Being the cheapest or quickest is not sustainable.

Small businesses have a chance to win on their own terms now.

Part 3: Dynamic Pricing & Strategic Transparency

The VPN Revelation: You share a personal story about booking flights where your sister’s VPN trick revealed that dynamic pricing is often hidden from the average consumer. How can small retailers use dynamic pricing without losing customer trust?

Katrina: The core problem with dynamic pricing is that consumers find it frustrating when the rules feel hidden. My sister’s VPN trick, discovering that the same flight was priced differently depending on where the search appeared to originate, is a good example of that discomfort.

For small retailers, the lesson is that dynamic pricing can be used effectively, but transparency can turn it into a positive customer experience instead of a negative one. This can be done by communicating the benefits clearly, for example “prices drop at 4pm,” and “we’ll refund the difference if the price falls within 48 hours”. The algorithm doesn’t need to be explained in technical detail. Customers just need clear, consistent rules that they can trust.

Wendy’s vs. Rema 1000: You contrast the backlash against Wendy’s "surge pricing" with the success of Rema 1000, a Norwegian supermarket. Why is "naming the rule" (like "Markdowns from 4pm") more effective than letting an algorithm run silently?

Katrina: Wendy’s problem wasn’t their pricing strategy, it was the language. The moment the media described it as “surge pricing,” it changed the story into a negative thing. Surge pricing implies you pay more when demand is high, which customers experience as being penalized for wanting a burger at lunchtime. The outrage wasn’t irrational.

Rema 1000 understood something important. The way you explain your pricing model makes a big difference. “Prices only go down. Reset every night.” That’s a rule, not an algorithm, and it’s a rule that works in the customer’s favor. “Markdowns from 4pm” creates anticipation and a sense of opportunity, rather than anxiety. When customers understand the logic and feel it benefits them, they stop resisting and start engaging.

The Sustainability Bonus: Rema 1000 reduced food waste by 40% using dynamic pricing. How does this support your argument that AI allows retailers to play a “smarter and more sustainable game”?

Katrina: Rema 1000’s results are pretty remarkable. They were able to achieve a 40% reduction in food waste and 23.5 million discounted items sold in a single year. Which makes me think how other retailers can leverage AI to help them implement improvements that everyone benefits from.

By automatically reducing prices before goods expire, they recover revenue that would otherwise be lost, reduce waste, and strengthen customer loyalty in the process. That’s the smarter, more sustainable game.

And it’s replicable beyond food retail. Anywhere that demand is variable and products have a shelf life or carrying cost, AI-driven pricing can align the retailer’s financial interests with better outcomes for both customers and the environment. Being more sustainable and being more profitable don’t have to be in tension; Rema 1000 is proof that they can reinforce each other.

Technology has always been a great leveler

when it becomes accessible. Small businesses can now access the same quality of personalization that the giants spent years building.

Part 4: Returns, Regulations, and Ultra-Fast Fashion

The Hidden Cost of “Free” Returns: You note that returns can cost 40% of an item’s original value and create massive CO2 emissions. How are AI tools like “Size Calculators” or Augmented Reality (AR) try-ons helping smaller brands solve the “fit problem” before the shipping label is even printed?

Katrina: When a retailer offers ‘free returns’, the cost doesn’t disappear, it gets absorbed into the business and passed on to consumers through higher prices elsewhere. Processing, inspecting, and restocking a returned item can cost up to 40% of its original value.

A better solution is to tackle this problem before the return needs to happen.

Zalando identified that half of its returns came down to fit, so its AI systems now analyze customer feedback, purchase history, and even the stretch characteristics of specific fabrics to guide customers to the right size before they buy. AI-powered size calculators have been shown to reduce returns by up to 45%. And when returns dropped, loyalty rose.

Warby Parker and IKEA approached the same problem, but with a different solution - augmented reality, letting customers visualize products on themselves or in their own space before purchasing.

For smaller businesses without the engineering resources of a Zalando, new products are starting to enter the market which will help any retailer do the same. But the entry point is even more accessible, and something any business can start doing now.

I used to have some clients that had hundreds of new products arrive every few months. But they were small businesses and didn’t have time to go and write product descriptions for everything. So they’d just upload the product to their website, without one, hoping that someone would be looking for that exact item, find it and buy it. But it can be as simple as better product descriptions generated with AI, giving customers accurate detail about fit and material; and returns monitoring that identifies which products have elevated return rates and why.

Retailers don’t need to jump straight to a full system, they can start making small improvements that overtime compound and improve their bottom line.

The Ultra-Fast Fashion Crisis: You describe the shift from Zara’s “fast-fashion” to Shein’s “ultra-fast-fashion,” which produces tens of thousands of products a month. Why is this model a “long-term viability” risk, and how can small businesses use AI to be the ethical alternative?

Katrina: The problem with ultra-fast fashion isn’t the speed. Their whole business model is to sell as much as possible with tiny margins. Shein’s demand-driven production reduces manufacturing waste. But the consumer behavior it incentivizes, buying more, wearing less, discarding quickly — creates waste at a different point in the chain.

And while some might argue that not everyone cares about sustainability as much as I do, the financial signals are starting to show that enough people do.

Despite dramatic revenue growth over the past five years, net profits are declining for fast-fashion retailers. Regulators are moving: new tariffs designed to raise the effective cost of ultra-cheap goods are being introduced across multiple markets, and France and Italy are actively developing legislation to restrict advertising for ultra-fast fashion items.

Millennial and Gen Z consumers, now the dominant purchasing demographic, are increasingly factoring sustainability, ethics, and circularity into their buying decisions.

For small businesses, this creates a genuine opening. A small retailer can’t compete on price with Shein. But they can stand for something. They can use AI to build products people genuinely want to own for longer, create honest and detailed product content that reduces mis-buying, and build authentic relationships with customers who are actively looking for an alternative.

Navigating New Regulations: By July 2026, the EU will ban the destruction of unsold clothing. Why do you believe AI is the only feasible way for retailers to track the environmental footprint data (water, emissions, recycling) that will soon be required by law?

Katrina: The scale of the data challenge is the key thing to understand here. From July 2026, large companies operating in the EU will be required to track and disclose environmental footprint data — water usage, emissions, recyclability — for individual products. That’s not a reporting task that can be managed with a spreadsheet, particularly for retailers with hundreds or thousands of SKUs moving through complex, multi-tier supply chains.

AI isn’t just the most feasible way to do this at scale. For most retailers, it may be the only feasible way. Tracking environmental impact across a product’s lifecycle requires processing large volumes of data from multiple sources that no manual process can handle at speed.

But compliance is just the starting point. A recent study of 102 textile enterprises in China found a meaningful association between AI adoption and actual reductions in pollution and carbon emissions. And so I hope more businesses will see the benefits of this reporting, that it’s not just a way to meet legislation, but to also make improvements to their business.

Part 5: Actionable Awareness

The Final Message: You conclude that “being the cheapest or quickest is not sustainable.” If a small business owner uses AI with purpose this week, what is the one human-centered outcome you hope they prioritize to stand out from the crowd?

Katrina: You don’t need to go all in or try to do everything with AI. Find one problem that AI could help solve. It also doesn’t need to be something ground-breaking or innovative; it can be as simple as improving product descriptions.

Most retailers haven’t updated their content to be optimized for AI agents, but consumers are shifting to using AI to make buying decisions. By improving your product descriptions, you can help your products be recommended by AI agents (and search engines), as well as helping customers when they land on your website to understand if this is the right product for them.

Just make sure any AI generated descriptions are accurate, as incorrect information can do more harm than good.

The moment it starts to feel like a company knows more about you than you've offered them, it stops being helpful and starts being unsettling.

---

Why This Matters

Katrina Watson’s perspective is a vital reminder that technology is a “great leveler” only when it is accessible and used with clear ethical guardrails.

• Democratizing Personalization: Powerful recommendation engines that once cost millions to build are now available as “personalization-as-a-service” for a fraction of the cost, allowing SMBs to grow up to 2.5 times faster.

• The Trust Requirement: Katrina highlights that AI succeeds when it solves a genuine customer problem, like Amazon’s early “word of mouth” digitizing, but it erodes trust when it feels like surveillance, as seen in the 2012 Target pregnancy case.

• Strategic Transparency in Pricing: She contrasts “surge pricing” (which penalizes high demand) with transparent rules (like Rema 1000’s “Markdowns from 4pm”), proving that when customers understand the logic, they engage rather than resist.

• Sustainability and the Bottom Line: Returns cost retailers up to 40% of an item’s value and create massive CO2 emissions; Katrina shows how AI size calculators and AR try-ons can reduce returns by 45% while building customer loyalty.

• Navigating New Regulations: With the EU banning the destruction of unsold textiles by July 2026, AI is becoming the only feasible way for retailers to track the complex environmental footprint data required by law

Thank you, Katrina Watson for your clear-eyed and empowering contribution to the Business section of our book. For smaller retailers feeling “crushed” by corporate giants, her chapter in AI Everywhere, Volume 1 is the definitive guide on how to reclaim your place in the market through purposeful automation.

Don’t miss her newsletter, Wake Up Excited, where she provides ongoing insights into combining business needs with technology and design. Her advice is a powerful call to action: you don’t need to do everything at once — just “find one problem that AI could help solve” and build your business on your own terms. I personally highly recommend following her guidance to ensure you are not just reacting to the retail transformation, but leading it with transparency and a human touch.

And if you’re interested in more conversations like this, consider subscribing to She Writes AI. I’ll continue sharing insights from the voices shaping the future of AI, including contributors from AI Everywhere, Volume 1, where 26 women from across five continents explore how artificial intelligence is changing the world we live in, and how we can change it for the better.

---

References

Book purchasing information

• Buy the book (now on Amazon, Gumroad, Lulu, Ingram Spark, and many ebook platforms). Look for our upcoming abridged audiobook edition in July!

• Tips from our publisher on how to buy an Amazon ebook from a non-KDP-marketplace country, and how to read it without a Kindle device (free)

• Review links (we’re on Amazon, Goodreads, and Storygraph)

For more insights on the book, see aiEverywhereBooks.com - our beautiful book series website created by Blessing Okpala, PhD:

Thanks for reading She Writes AI Community! Subscribe (free) to receive new interviews, weekly digests, & book news and to support our work: