Services
Articles by Purva
-
Lily AI at Series B: Transforming the Retail Stack by Injecting the Language of the Customer
Lily AI at Series B: Transforming the Retail Stack by Injecting the Language of the Customer
What a journey it’s been. I’m thrilled to announce today that Lily AI has closed our $25M Series B financing round with…
76 Comments -
Customer Intent: the Universal Language that Supercharges the Ecommerce Stack
Customer Intent: the Universal Language that Supercharges the Ecommerce Stack
For years, fragmented and disjointed ecommerce stacks have been missing a universal language that can help present…
4 Comments -
Lily AI raises $12.5M to help retail understand the emotional context of individual customers
Lily AI raises $12.5M to help retail understand the emotional context of individual customers
We just raised $12.5 million in our A round led by Canaan with support from existing investors NEA, Fernbrook…
83 Comments
Activity
7K followers
-
Purva Gupta shared thisThis just made my day. We received an endorsement from Kristin Morseman, SVP of Americas Digital Transformation at Shiseido, for what our teams have built together over the last two years. Shiseido is one of the rare beauty companies whose name carries 154 years of brand equity. Today, its portfolio includes more than 30 brands, with roughly $6.5 billion in annual sales and over $7 billion in market value. Companies at this scale do not hand out endorsements to vendors casually. They care about one thing: results. That is why Kristin’s words mean so much to us: “Lily AI helps us turn product content into a performance and AI discovery advantage, improving how our brands perform across Google, Meta, and emerging AI recommendation surfaces, from stronger paid media efficiency to category-leading visibility in LLM-driven discovery.” What makes this endorsement even more meaningful is that it was not based on one successful experiment. Over the last two years, the Lily and Shiseido teams have run multiple tests across five brands, covering Google, Meta, and SEO. That progress came from the Shiseido team’s willingness to keep testing. Their internal teams and agencies worked alongside us, asked difficult questions, and used what we learned from every experiment to make the next one better. Kristin’s endorsement captures the bet we made years ago. We did not want to build for just one channel. We wanted to build the product understanding that sits underneath all of them, because every system deciding what to show a shopper depends on the same thing: how well your products are described in the language customers actually use. This endorsement was hard-earned, and I could not be prouder of what our teams have built together. Thank you, Kristin, and the entire Shiseido team.
-
Purva Gupta shared thisActivant Capital just named Lily AI in one of the most structurally defensible categories of agentic commerce. That category is catalog enrichment. Their new market map places catalog enrichment underneath the largest shopping agents and commerce platforms for a simple reason: ChatGPT, Gemini, Amazon and every future shopping agent can only recommend the products they can understand. That is the product intelligence layer Lily has been building. As I shared in the report, “The retailer describes it as midnight French terry athleisure. The customer searches for a navy hoodie.” That gap seems small when you’re looking at a single product. Now multiply it across millions of products, thousands of attributes and every query where an agent is deciding what deserves to be recommended. - An agent cannot infer that a dress is appropriate for a summer wedding when the occasion is missing. - It cannot recommend a serum for dry, sensitive skin when the ingredient and skin-type data is incomplete. - It cannot surface the right running shoe when cushioning, arch support and pronation are buried inside vague marketing copy. The product may be exactly what the customer wants. But to the agent, it effectively does not exist. That’s what makes catalog enrichment so defensible. Every new shopping interface increases the need for clean, structured, consumer-centered product intelligence. The agents will evolve. The commerce surfaces will evolve. But the underlying product truth still has to be created, enriched and continuously optimized. I am proud to see Lily AI included in this map and to have our perspective represented in the research. We help retailers translate merchant language into the language customers actually use. And agentic commerce is now turning that translation layer into core infrastructure.
-
Purva Gupta shared thisThe Trade Desk has lost over $20 billion in market value this year, and its stock is now down more than 45%. There are some obvious explanations: its growth has slowed, Amazon's ad platform is competing for the same ad budgets, the company has faced executive turnover, and Publicis has publicly challenged its fee practices. But underneath all that, there is a much larger reason. And if you are in advertising, you need to pay attention to it. The Trade Desk built an extraordinary business around execution intelligence: bidding more precisely, targeting more sharply, allocating budgets faster, and finding better inventory across the open internet. For years, that was the advantage. The companies that could execute better won. But AI is steadily absorbing that work. Bidding, audience selection, campaign structure, budget allocation, creative production, and query interpretation are all becoming automated. As every advertiser gains access to similar optimization machinery, execution becomes less differentiated. So where does the advantage go? It moves upstream, into the inputs the machinery cannot create for itself. That is why product intelligence is becoming so important. An advertising system can optimize the distribution of a product, but it first has to understand what that product actually is. It needs more than a title, a category, and a generic description. It needs to understand material, fit, occasion, aesthetic, function, compatibility, and the language a customer would naturally use to describe what she wants. A bidding engine cannot invent that understanding. Neither can an AI shopping agent. The quality of the product intelligence layer increasingly determines which searches a product matches, which audiences it reaches, which recommendations it enters, and whether the media spend behind it has any chance of performing. Advertising is not becoming less valuable. But the advantage is moving from buying attention to making the product intelligible to the systems deciding where that attention goes.
-
Purva Gupta shared thisWe’re thrilled to announce our partnership with Marks and Spencer, one of the UK’s most iconic and trusted retail brands, and the results are already speaking for themselves. M&S is using Lily AI to automate and enrich the creation of structured product data at scale. They dramatically reduced the manual effort and multiple partners required to bring new items to market while ensuring every product is accurately attributed, categorized, and ready for discovery from day one. They deployed Lily AI’s Product Intelligence Platform to supercharge their Google shopping and organic search performance. By enriching their product feed with high-quality, AI-powered attributes, M&S is driving stronger visibility, higher click-through rates, and a meaningful lift in revenue from paid and organic channels. "Lily AI has become a core part of how we manage and scale our product content at M&S. From setting up new items to optimising for Google, the platform delivers across the board. It’s already making a measurable difference to our efficiency and our commercial performance." Stephen Orford, Head of Online Experience, Marks & Spencer In a world where your product feed is your storefront - on Google, in AI-powered search, and across every channel the quality of your product content is a competitive advantage. Marks & Spencer gets that. And we’re proud to be their partner. If you are looking to unlock more from your catalog, let's talk. #RetailTech #EcommerceAI #ProductContent #GoogleOptimization #MarksAndSpencer #LilyAI #AISearch #UKRetail
-
Purva Gupta shared thisThat was a good one. Just wrapped “Search, Answers and Skepticism” at CommerceNext with Ken Pilot and Noam Paransky, and the three of us went straight at the real question: what in AI search is actually moving a dollar today, and what is just the loudest conversation in the room. As usual, Ken and Noam kept the room entertained 😁
-
Purva Gupta shared thisThe most expensive thing in retail right now is the attention brands are pouring into a future that isn't generating revenue yet. Every brand I talk to is being pulled into the same script. Optimize for AI answers, get ready for agents, prepare for a web where machines do the reading. Most of the discovery shift is real, but very little of it is moving a dollar this quarter, and almost none of it comes with a way to prove that it did. Meanwhile the surfaces actually producing revenue today get treated like settled infrastructure that doesn't need attention anymore. Teams fund the speculative thing and starve the working thing, because the speculative thing is where the conversation is loudest. This is the conversation I want to have in the open. At CommerceNext, I’m joining Ken Pilot and Noam Paransky for a session called “Search, Answers and Skepticism.” Ken has built some of the most recognizable brands in retail. Noam has driven over a billion dollars of online growth. Different seats, same low tolerance for a story that cannot be measured. We’ll get into what is genuinely shifting across paid search, AI Overviews, and answer engines, where brand still wins when a machine is choosing, and the question most people skip: how would you even know any of it is working?
-
Purva Gupta posted thisA large retail brand I know fired three paid marketing agencies in 2 years. Their CMO told me, with real frustration, that you just can’t find a good paid agency anymore. But when I looked closer, it didn’t look like three agencies had failed in three different ways. It looked like all three had inherited the same auction and were being asked to create an edge from levers they no longer controlled. Over the last five years, a lot of what good agencies used to win on has moved inside the platform. Bidding is mostly automated. Channel mix has been compressed into products like Performance Max. Search queries, product matching, placements, and budget allocation are increasingly decided by the system, not by someone manually shaping the account. That does not make agencies useless. It means the old agency edge has been commoditized. The setup a good agency used to build by hand is now much closer to the market default. Competitors can switch on the same machinery, use the same bidding models, and reach the same inventory. So the real question is no longer just, “Who is managing the account?” It is, “What are we giving the system to work with?” That is where the leverage has moved: product feed, merchandising, pricing, margins, hero SKUs, promo calendar, landing pages, creative, measurement, and the ability to define which customers and outcomes actually matter. The auction was not failing because three agencies in a row forgot how to do paid media. It was failing because the real edge had moved closer to the product, the catalog, and the economics of the business. So the CMO was asking the wrong question. “Who do we hire next?” might help at the margins, but it will not change the ceiling if the inputs stay the same. The thing capping the account no longer sits only in the agency’s hands. It sits upstream with the brand itself.
-
Purva Gupta shared thisEvery time Google introduces something as optional, it eventually becomes infrastructure for a business that generates roughly $300 billion a year in advertising revenue. * Mobile-friendly was optional. * Page speed was optional. * Structured data was optional. Each one stopped being a choice. At Google Marketing Live last month, Google launched conversational attributes, a new set of product-feed fields built for AI shopping. They're optional too, for now. Look at how this one arrives. The fields won't affect whether your existing products stay approved. You can add them through a supplemental file without touching your main feed. And if the information already lives in your descriptions, Google tells you not to bother duplicating it. Every reason a busy team might say no has been removed in advance. Platforms don't sand down the friction like that for things they're indifferent about. They do it when they need the data more than you currently need to give it. And Google does need it. Its AI shopping answers fall apart on thin listings. When someone asks a long, specific question in full sentences, the system has to pull real detail from somewhere, and most feeds can't supply it today. So Google built the easiest possible on-ramp and made the ask sound like a favor to the brand. That's the pattern. Optional is how Google introduces what later becomes the price of entry. It's already visible one layer over, where Google now calls rich product attributes the entry requirement for its agentic retail surfaces, not a nice-to-have. For now, conversational attributes are optional. Google's history suggests optional rarely stays that way.
-
Purva Gupta posted thisThe fastest way to understand a company's real business is to watch what it builds after the exciting idea dies. OpenAI spent a year talking about conversational shopping - buy things inside chat, skip the website, check out instantly without ever leaving the conversation. It was framed as a clean break from Google and traditional e-commerce. That version didn’t last. Instant checkout was shut down in March. What’s showing up instead is more familiar. Product data is now being pulled into an ads system. There’s a structure where each item carries its own listing, performance metrics, and advertising signals. Feeds provide the raw input, and that input gets turned into ads. If that feels familiar, it’s because it is. It looks a lot like Google Merchant Center. So, the company that was supposed to reinvent search is, in practice, rebuilding the infrastructure that powers shopping ads on Google. Not the ranking algorithms or the interface layer, but the less visible plumbing that actually makes commerce work: structured product feeds that map items to queries. That’s not necessarily a contradiction. It’s a pattern. Every platform that tries to monetize shopping ends up in the same place. Interfaces can change dramatically, but the underlying requirement doesn’t: you still need clean, structured product data before anything else can happen. The interface battles get the attention, ChatGPT versus Gemini versus Rufus, but under all that, the same system keeps reappearing. Once you introduce ads or transactions, you need something to sell, something structured, something measurable. That “something” is always a feed. So the real signal isn’t in the announcement about how people will shop next. It’s in what gets built quietly afterward. And more often than not, it turns out to be a feed.
Purva Gupta commented on a post
2d
terrific, excited for this partnership! 👏
-
Purva Gupta reacted on thisPurva Gupta reacted on thisRecognition is powerful, but it is not the same as relevance. That was one of my favorite threads from my conversation with Elizabeth Preis, CMO of Victoria’s Secret, on my latest segment on Retail Refined. Victoria’s Secret & Co. is one of those brands that has shaped both retail and culture. But iconic brands do not get to rely on awareness alone. They have to keep earning emotional connection, consistency, and trust across every touchpoint year after year. Elizabeth shared a clear way to think about it: - Awareness is: do they know us? - Relevance is: are we meeting their needs? - Resonance is: are we helping them feel what they want to feel? That distinction matters. In this episode, we talk about leading with emotion over promotion, the evolution of “modern sexy,” the power of fandom, and what it takes to keep an iconic brand moving forward without losing its core. Here is a teaser clip from our conversation. You can see the full episode at: https://lnkd.in/epTZSWmi Or listen to it on Spotify here: https://lnkd.in/eNKzdWq2 Or Apple Podcasts here: https://lnkd.in/eHHvC5gT #RetailRefined, a MarketScale #podcast
-
Purva Gupta reacted on thisPurva Gupta reacted on thisThrilled to be joining the Board of Steve Madden. This was announced today on the heels of a stellar quarter and earnings call. Thank you Ed and Steve. Steve Madden Board of Directors will expand from ten to eleven directors, and Ken Pilot will join the Board as the newly appointed director. Mr. Pilot is the Founder and Chief Executive Officer of Ken Pilot Ventures, an advisory and investment firm focused on retail, consumer and commerce technology companies. He brings more than 30 years of leadership experience across retail and consumer businesses, having served in senior executive roles at leading retailers including J.Crew, Gap Inc. Ralph Lauren, American Eagle Outfitters Inc. and abc carpet & home. Mr. Pilot currently advises and invests in a number of companies focused on artificial intelligence, e-commerce infrastructure and retail technology platforms. Mr. Rosenfeld commented, “We are pleased to welcome Ken to our Board of Directors. His decades of experience building brands and driving growth, together with his deep understanding of digital innovation and emerging technologies, will be invaluable as we continue to execute our long-term growth strategy. We look forward to benefiting from his insights and perspective.” https://lnkd.in/gPkSUhrC
Experience & Education
-
Lily AI
View Purva’s full experience
See their title, tenure and more.
or
Honors & Awards
-
Harley Lippman Entrpreneurial Fellowship
Yale School Of Management
Awarded a $4000 grant from the Harley Lippman Yale SOM Summer Entrepreneurial Fellowship Fund which is designed to promote entrepreneurial activity within the Yale SOM community.
-
Young Leader Award 2013
Indian School of Business
Recommendations received
1 person has recommended Purva
Join now to viewOther similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content