In this episode
Dan Bennett, CMO and founding team member, and Alexandra Seaman, Co-Founder and SVP of Furniture.com, join Greg Kihlström at the company’s New York headquarters to explain how they’re building a “decision layer” for furniture — a confidence-building platform that sits on top of commerce rather than adding to the endless aisle. They argue that discovery is no longer the customer’s problem; confidence is. The conversation covers why furniture shoppers spend hours researching a single couch, the year of machine-learning work it took to standardize furniture data, how a 140-piece-a-month editorial operation turned into Large Language Model (LLM) visibility and higher conversion, and how Furniture.com measures a “trust score” built from real shopper behavior rather than speed to checkout.
Key takeaways
- Choice isn’t the problem — confidence is. Bennett says there are already a million things to choose from; shoppers want a tool that helps them decide confidently, not another 200 sofa options thrown at them.
- Furniture is a massive, high-stakes category. As a retail category it’s bigger than movies, music, and video games combined, and it’s dominated by expensive, tactile purchases people are anxious to get right.
- The confidence gap is why people spend ~9 hours researching a couch. A sofa is big, expensive, and often bought unseen, so shoppers worry about fit, comfort, and whether returning it will be a nightmare — Furniture.com’s job is to remove that uncertainty.
- Standardizing the data was the hard, expensive foundation. None of the retailer feeds match, so the team spent roughly 12 months of machine-learning work standardizing diverse furniture data into a usable data lake before any AI could be built on top of it.
- Editorial content became LLM visibility. A bet made two years ago on an in-house editorial group publishing ~140 pieces a month didn’t pay off early, but now drives strong presence in LLM answers — and that content-sourced traffic checks out at a higher rate than already-strong organic.
- LLMs don’t yet deliver confidence on high-consideration buys. For a house, car, or sofa, an LLM’s answer can change on re-prompt, so Furniture.com aims to be the trusted arbiter of the decision wherever checkout ultimately happens — even inside an LLM.
- The AI has to be a category expert, not a generic model. Their assistant “Dottie” is trained to reason like a designer and merchant so answers carry real furniture context, because the category is as much art as science.
- Measure confidence, not just conversion. The team builds a “trust score” from behaviors like time on site, content engagement, return visits, review use, and Dottie prompts — resisting the instinct to truncate the journey just to reach a faster sale.
- Partner quality protects the experience. With an agentic cart handing fulfillment to ~90 vetted retailers, partner curation, white-glove relationships, and a launch-partner feedback loop are core to the model, not an afterthought.
Chapters
- 00:00 — Cold open: the paralysis of endless choice
- 01:54 — Meet Alexandra Seaman and Dan Bennett
- 03:08 — What Furniture.com is: a decision layer, not a marketplace
- 04:59 — Why other categories had this and furniture didn’t
- 06:03 — The hard part: standardizing furniture data
- 07:40 — Why nine hours to buy a couch signals a confidence gap
- 10:00 — Being found in LLMs, and where content pays off
- 12:09 — Content as an interior designer that builds confidence
- 13:12 — Why the decision layer is broader than AI
- 13:41 — What LLMs still can’t do for high-consideration buys
- 15:46 — Furniture as art and science; training “Dottie”
- 18:36 — Solving for individual taste vs. the endless aisle
- 19:50 — Measuring a trust score, not just conversion
- 22:05 — How partner quality protects the experience
- 26:00 — The next five years of furniture shopping
- 28:00 — Shop by attribute: Furniture.com as the first stop
- 29:13 — Staying agile
Why choice isn’t the problem — confidence is
Bennett opens by reframing the whole premise: furniture already offers a million ways to choose and endless aisles of product, so more options aren’t what shoppers need. During roughly six months of early diligence, the team found people wanted a way to make a decision confidently — a tool that guides them, not one that surfaces another 200 sofas. Seaman adds that other big-ticket categories already have this: buyers turn to Zillow for homes and AutoTrader for cars, but nothing equivalent existed in furniture, largely because it’s expensive and difficult to build.
What a “decision layer” actually means
The “decision layer,” in Bennett’s framing, is really about confident decisions. Furniture.com starts by working only with trusted national retailers, so shoppers begin from a place of trust, then layers on machine learning that lets people describe what they want semantically and get to the right product quickly. Seaman describes the product side — advanced semantic search, the “Dottie” AI assistant, mood boards, and a “Complete the Room” feature — as tools that behave like an interior designer, helping shoppers not just find a product but assemble a confident choice around it.
Why data standardization was the hard part
The unglamorous foundation, Bennett explains, was data. Furniture feeds are extremely inconsistent — the same piece is described differently from retailer to retailer — so the team spent roughly 12 months of machine-learning work, with a large data-science group, standardizing that data into a data lake that AI could actually be built on. Without that step, he says, anything built on top would be garbage. It was time-consuming and expensive, but the size of the category and the payoff of a genuinely better consumer experience made it worth the lift.
How editorial content became LLM visibility
About two years ago, Furniture.com invested heavily in an in-house editorial team that operates almost like a magazine, publishing around 140 pieces of content a month. Early on it didn’t drive volume — it mostly made the site look polished, like high-end rug guides. That investment now pays off in LLM visibility: because the content is expert, trusted, and used, Furniture.com shows up well in AI answers, and shoppers arriving from that content convert at a notably higher rate than the site’s already-strong organic traffic. Bennett’s takeaway is that authenticity can’t be underestimated in these new channels.
Measuring confidence, not just conversion
Bennett calls measurement the multimillion-dollar question. Beyond standard site and marketing metrics, the team tracks a deeper layer of behavior — whether shoppers spend time on the site, engage with content, view multiple pages, return, use reviews, and how many prompts they put into Dottie — and matrixes those into an overall trust or confidence score. He deliberately resists the common instinct to truncate the journey for a faster sale; the goal is for shoppers to build genuine confidence, which he believes comes from spending the right kind of time. That’s backed by ongoing NPS and brand-lift studies, questionnaires, and in-person consumer work.
Where furniture shopping goes in five years
Seaman expects that even five years out, most furniture shoppers — more than three quarters — will still visit a store at some point, because the product is so tactile and visual, but that AI will be ingrained in how nearly everyone searches rather than a subset of shoppers. Bennett’s prediction is sharper: because most people shop by attribute rather than by brand, he expects consumers who don’t have a specific brand in mind to go to Furniture.com first — the way they turn to KAYAK for flights or AutoTrader for cars — whether that interaction happens on the site or inside an LLM environment.
FAQ
What is a “decision layer” in e-commerce? It’s a confidence-building layer that sits on top of commerce rather than competing with it. Instead of adding more options, it helps shoppers make a confident choice — through trusted retailers, semantic search, AI assistance, and expert content that guides the decision.
Why doesn’t more choice help furniture shoppers? Because the real problem is a lack of confidence, not a lack of options. Furniture is big, expensive, and often bought unseen, so endless choice across countless channels and brands becomes overwhelming rather than helpful.
How does Furniture.com use AI? For simplification and decision support, not just personalization. Advanced semantic search and the “Dottie” assistant — trained as a design and merchandising expert rather than a generic model — help shoppers reach a relevant product quickly and with context.
Why did Furniture.com invest so heavily in content? An editorial team publishing roughly 140 pieces a month initially made the site look polished without driving volume, but it now generates strong visibility in LLM answers and delivers traffic that converts at a higher rate than organic.
How do you measure the success of a decision layer? By tracking behaviors that signal building confidence — time on site, content engagement, return visits, review use, and Dottie prompts — combined into a trust score, alongside NPS, brand-lift studies, and in-person consumer research.
About Dan Bennett
Dan Bennett is Chief Marketing Officer and founding team member at Furniture.com, where he’s reshaping furniture retail through AI, data, and an open marketplace model. Since 2022, Dan has helped scale the company to 80+ employees across two offices, connecting millions of shoppers with hundreds of brands. Before Furniture.com, he was CMO at Packable, a Carlyle-backed e-commerce platform that grew into Amazon’s largest third-party marketplace partner and reached a $2 billion valuation. Earlier, Dan spent over a decade leading digital and brand strategy at McCann and Grey, driving growth for global names like P&G, Facebook, Microsoft, and Nike. Dan brings a focus on clear priorities, fast execution, and building teams that fuel sustainable growth.
Dan Bennett on LinkedIn: https://www.linkedin.com/in/bennettdaniel/
About Alexandra Seaman
Alex Seaman is Co-Founder and SVP at Furniture.com, where she leads strategy and partnerships for a fast-growing furniture shopping platform redefining how people discover and buy furniture. Since launch, she has expanded the platform to more than 70 participating retailers and earned recognition as one of Furniture Today’s “40 Under 40.”
Resources
Furniture.com: https://www.furniture.com
The Agile Brand podcast is brought to you by TEKsystems. Learn more here: https://aglbrnd.co/r/2868abd8085a9703
We’re proud to be a media partner for #MAICON26 – Oct. 13-15! Learn how AI can power your marketing and business and help you grow smarter. Use code AGILE150 to save! https://aglbrnd.co/r/7fe458ced0f04658
Reach your customers with Reddit. Spend $500 in ad spend, get $500 back in ad credit! Learn more: https://advertalize.com/r/491818c79fb1873f
Chaser is the only Slack-native project management platform that helps teams turn messages into tracked tasks, automate follow-ups, and maintain team-wide visibility, without adopting another tool. Now integrated with Claude and other generative AI tools, Chaser is the only platform that brings AI-powered project management into Slack, where teams already work. Chaser is based in Toronto, Canada. Learn more at trychaser.com.
The most influential minds in software, AI, and engineering leadership will be at WeAreDevelopers World Congress North America, September 23-25 in San Jose. Learn more: https://aglbrnd.co/r/60a7299222a7bcf1
Enjoyed the show? Tell us more at and give us a rating so others can find the show at: https://aglbrnd.co/r/faaed112fc9887f3
Connect with Greg on LinkedIn: https://www.linkedin.com/in/gregkihlstrom
Don’t miss a thing: get the latest episodes, sign up for our newsletter and more: https://aglbrnd.co/r/35ded3ccfb6716ba
Check out The Agile Brand Guide website with articles, insights, and Martechipedia, the wiki for marketing technology: https://www.agilebrandguide.com
The Agile Brand is produced by Missing Link—a Latina-owned strategy-driven, creatively fueled production co-op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. https://www.missinglink.company
Transcript
[00:00:00] Greg Kihlström: Hi. I’m Greg Kihlstrom, host of The Agile Brand, and here’s a question for you. What if the infinite shelf of e-commerce hasn’t made customers more confident, but has instead created a new kind of friction, the paralysis of endless choice? Agility requires brands to shift their focus from simply providing more options to actively helping customers navigate them. Today, we’re going to talk about understanding the modern customer journey when discovery is no longer the primary problem, building value as a decision layer that sits on top of commerce rather than competing with it, and using AI not just for personalization, but for simplification and decision r- support. Welcome to Season 8 of The Agile Brand podcast. This season, we’re going all in on expert mode Martech, AI, and customer experience, talking with the people and platforms behind the brands you know and love. Now let’s dive in. To help me discuss this topic, I visited the Furniture.com headquarters here in New York, and I’d like to welcome Dan Bennett, CMO, and Alexandra Seaman, co-founder and SVP of Furniture.com. Dan and Alexandra, welcome to the show.
[00:01:35] Alexandra Seaman: Thank you.
[00:01:35] Dan Bennett: Thanks for having us.
[00:01:36] Alexandra Seaman: Thanks for having us.
[00:01:37] Greg Kihlström: Yeah, and great to be here in your offices.
[00:01:38] Dan Bennett: Welcome.
[00:01:39] Greg Kihlström: Yeah, yeah. I mean, I- I was expecting good furniture, and I was, you know, pleasantly, (laughs)-
[00:01:45] Alexandra Seaman: Not surprised?
[00:01:46] Greg Kihlström: … not surprised. Yeah, yeah, exactly.
[00:01:47] Dan Bennett: it’s a priority for us. (laughs)
[00:01:49] Greg Kihlström: Yeah, exact- exactly. Great. so before we dive in, why don’t you both introduce yourselves? Madam?
[00:01:54] Alexandra Seaman: I’m Alex. I’m the co-founder of Furniture.com, and I have been here since the beginning, employee number one, employee number two.
[00:02:05] Greg Kihlström: Nice.
[00:02:06] Alexandra Seaman: my family actually, kind of grew up in the furniture retail space, but much more the traditional legacy furniture retail space, actually fourth generation (laughs) in the furniture business.
[00:02:17] Greg Kihlström: Wow.
[00:02:17] Alexandra Seaman: And so this is, you know, sort of a kind of continuation, but very fresh approach to furniture, really, for the shopper in 2026.
[00:02:31] Dan Bennett: I’m Dan Bennett. I am the co-founder and CMO, as Alex mentioned. go back to slightly after, I had m- the most beautiful thick hair when we started. You wouldn’t believe it.
[00:02:43] Greg Kihlström: (laughs)
[00:02:43] Alexandra Seaman: It was longer than mine.
[00:02:44] Dan Bennett: It was, wow, it was sensational. And then, so I’m first generation. This is my first foray into, furniture as a category. but I’ve spent, my career building, technology brands. So the other, the, very much the other side of our business, which Alex also has experience in is, is really that we’re a technology business, in service of the furniture shopper.
[00:03:08] Greg Kihlström: But a few things today, and, and what I teed up in the intro, but why don’t we start with just giving the audience a little background on Furniture.com itself.
[00:03:15] Dan Bennett: Sure.
[00:03:15] Greg Kihlström: You know, it’s described as a, a decision layer rather than a traditional marketplace. So, you know, what, what does that mean? Who’s the primary customer you’re building for?
[00:03:22] Dan Bennett: Y- you know, a little stat first. Furniture as a category, a retail category, is giant. It’s, it’s actually bigger than, movies, music, and video games combined. It’s a large category. so it has a large shopper base. What we realized, and I think part of your, sort of supposition around this business, o- or this conversation, is that, choice is not the problem. There is a million ways to choose and a million things to choose from. and there are lots of platforms that give you an endless aisle, that give you endless choice. what, what we realized in the, in the genesis of this business in speaking with, consumers, we did probably six months’ worth of diligence on this, industry from a digital point of view, is that people want a way to make a decision confidently. They want a tool that helps them make a decision, not that throws them another 200 sofa options.
[00:04:16] Dan Bennett: So Furniture.com as a business has always been set up to do that. first and foremost, we work with, you know, the most trusted retailers in the country. All right? So there’s a, there’s a little decision made to start with. “Okay, I know I can trust this brand. It isn’t just a, a, a product from overseas I’m not familiar with.” and then we have a, and then we always built from the very beginning, even from before it was de rigueur, machine learning is the foundation of the business. So when you come in and you want to talk to the platform semantically, we can get you very quickly to a product that, answers what you’re looking for. And so you start to combine those types of things, and that’s what we mean when we say decision engine. We- we genuinely help people get to a decision. We truncate that journey.
[00:04:59] Alexandra Seaman: I think even, if you think about other industries, you know, real estate, homes, eh, cars, et cetera, like, all these other big, giant ticket purchases that somebody will make, and, you know, the research and or buying things online, there’s a platform to help you do that, to help you compare, to help you sort of, to help guide you along your decision through lots of different options, lots of different companies, lots of different brands. You think of Zillow, you think of AutoTrader. And so when we initially came in, four or five years ago, we sort of had that kind of idea. Like, “There’s … That doesn’t exist in furniture, why? You know, with all these amazing, trusted retailers?” the answer is it’s, it’s expensive and difficult to build.
[00:05:43] Dan Bennett: Yeah, it’s, it’s really hard.
[00:05:44] Greg Kihlström: (laughs)
[00:05:44] Alexandra Seaman: Right? but it has, and then of course, in the last couple years especially, given the technological evolution in AI, the platform has evolved a lot. We actually relaunched six months ago. It’s evolved a lot, to kind of really, um- … be a better version of that now, I think.
[00:06:03] Greg Kihlström: Yeah, yeah. Well, and I mean, aren’t those always the best ideas, are the things that it, it seems so, like, why didn’t someone do this first?
[00:06:03] Alexandra Seaman: It seems.
[00:06:10] Greg Kihlström: But, I mean, obviously you’ve realized that it’s not easy to do. (laughs) But, you know, t- the fact that it was, it was kind of a natural progression makes it, you know, it makes it an int- interesting business, right? Yeah, you know, it’s… There are many reasons why it’s difficult. first and foremost, you know, we have been able to develop great relationships with our partners, and that takes time. from a technical point of view, I’ve said this in a few other articles, so, so I, I, I’ll repeat it again, the, the, the standard of data in furniture is extremely diverse. Um- Hmm.
[00:06:48] Dan Bennett: None of the feeds are the same. Sometimes what furniture is described as is not the same. So we had to spend probably 12 months of machine learning time with our data… We have a big data science team. With that team to literally standardize how that data can live in a data lake that allows us to build AI on top of it, ’cause without that, it would be garbage. So that was a heavy lift. And it, it’s time-consuming and it’s expensive, but again, the category, the payoff in the category, we believe, is worth it, and, and perhaps more importantly, developing a consumer experience that’s better, especially in furniture, which Alex obviously is very close to, I am now very close to, is, you know, it’s why we do it.
[00:07:30] Greg Kihlström: Yeah, yeah. Well, and I mean, that, the fact that the data is so, inconsistent from the manufacturers, that also trickles down to the customers, right?
[00:07:40] Greg Kihlström: I mean, you know, there’s a stat here, you know, it ta- consumers spend an average of nine hours researching a couch.
[00:07:47] Alexandra Seaman: That’s right.
[00:07:47] Greg Kihlström: It’s, some of that, I would, I would imagine is the apples and oranges and, you know, things like that, as far as, you know, people calling their products different, and even, you know, what we talked about earlier, just the, the vast amount of choice. in this era of infinite choice, you know, what does it tell you about the, the core problem that consumers are facing and why has e-commerce generally made this even more complex?
[00:08:15] Alexandra Seaman: I think, you know, people… The, the core problem is probably a lack of confidence. When you’re buying a piece of furniture, it’s a, it’s first of all big. (laughs)
[00:08:26] Alexandra Seaman: Like you, you know, it’s expensive, it’s big. buying it online, so your question about e-commerce, is even more difficult, because now I’m buying without seeing it, I’m buying without sitting on it. Is it comfortable? Is it gonna fit in my apartment? Is it gonna fit through my door when it gets delivered? Like, what is the state of it gonna be when it gets delivered? And, you know, anyone who’s ever bought a sofa, you know you really don’t wanna have to return it. It’s-
[00:08:50] Greg Kihlström: Right. (laughs) Yeah.
[00:08:50] Alexandra Seaman: It’s, it’s really quite a hassle.
[00:08:52] Alexandra Seaman: And so I think that people certainly have a lot of choice. And you have, I mean, you think about all the channels serving you choice, right? You have everything coming through your phone, your laptop, your TV, whatever. And then you think about all of the different brands on all of those channels. It’s, it’s really, like, it’s infinite and it’s overwhelming, but how do you know I can trust this? I’m not gonna get this and it’s not gonna fit, or it’s not gonna be comfortable, or my kid’s gonna spill pasta sauce on it, and then I’m sp- you know, and I’m-
[00:09:20] Greg Kihlström: Right, right. (laughs)
[00:09:21] Alexandra Seaman: … I’m out of luck. So, like, I think really helping give them confidence, and a lot of that is the data piece, right? Like, you know, giving all of the relevant information about size, dimensions, fitting, about, you know, is it performance fabric? Does it work for my, my family with three kids and a dog? Is it, you know, all those things is really imp- Can I see a bunch of photos? Is there a video?
[00:09:42] Greg Kihlström: Yeah. And so, you know, given, again, given infinite choice and, the, it, this becomes a decision problem, right? It’s, it’s, it, we can probably find lots of products that seem right, but making the right choice, having, having the confidence is really what it comes down to, right? So, you know, so many, and, and in so many other industries, the, the goal seems to still be to just be found, but, you know, what you’re talking about here, I’m sure it translates to other industries as well. are we moving… Obviously you need to be found, but, you know, are we also moving past that towards providing just better clarity, better confidence? Yeah, I, I think, from a marketing point of view, being found is one of our, you know, ambitions. Sure, sure. And, and there, and there are levels to that, for sure. We’re not immune to the fact that we know we need to be found in, the LLMs. And, in fact, we are showing up well in the LLMs. Th- that actually, the, the anecdote there, candidly, is, about two years ago, we decided to invest very heavily in, in content creation internally: so we have a content group here, if you think about it, like sort of almost a magazine editorial group. and what that team does month-to-month is they put out somewhere around 140 pieces of content. that, and that can vary from, social to long-form how-tos, you name it, you know. that, in the early days, d- doesn’t pay, right? It doesn’t pay dividends ’cause it doesn’t really drive volume. It makes the site look beautiful. It means that when you come to the site and you’re looking for a rug and there’s a couple of rug guides that look like they were designed by, you know, a high-end, publisher, that was the value to start with. That value now is starting to show up in the LLMs. So because we have this… Like I said, it goes back to confidence and trust. Because we have this content developed, we’re able to appear in those places. And certainly there are places we buy our way into it, you know, through Google and th- but, but in that environment, we sh- we’re starting to show up because the content is right, and it’s expert, and it’s trusted, and people use it. And then most importantly, people that come from that content check out at a much higher rate- …than, than even our organic traffic, which has very good conversion rate to start with. So I think the authenticity that has to go along with these new channels- … can’t be underestimated.
[00:12:09] Alexandra Seaman: And if you think about, you know, sort of that content, the different types of content and editorial type content in helping someone make a decision, you can almost think about it similar to, like, if I have an interior designer, right, who’s helping me, merchandise a room when I know I like this green sofa. Okay, how do I put the rest of the room together? And you think about content that helps you do that, it’s, like, a designer helps you decide. It helps, it gives you confidence-
[00:12:34] Greg Kihlström: Oh, yeah, right. (laughs)
[00:12:34] Alexandra Seaman: … to make a decision. And so, you know, in the marketing realm and editorial realm, it’s content, and then on the product side, we of course have lots of different features that, that lend themselves to this, like ha- help people decide. The obvious ones, like very advanced kind of semantic-based searching, or, you know, our AI, like, AI assistant chat, Dottie, which in herself is a mini LLM, but then, you know, things like mood boards, things like Complete the Room, which is kind of what I just described, that really do help people guide people in decision, help them get to a relevant product quickly, but then help them kind to go a step beyond.
[00:13:12] Greg Kihlström: Yeah. So, I mean, trad- I, I use the word traditionally, lightly I guess, but, you know, traditionally we would say a decision, the decision layer would be something solely like AI, and, and certainly you’re using plenty of, of those technologies, but it sounds like it’s a broader play than that, right? I mean, it’s, there’s a lot of everything from the, the guides, which are more content, but still helping people make decisions. So is that, is all of that kind of that, that decision layer?
[00:13:41] Dan Bennett: Yeah, I think the, the, the descriptor for decision would be about confidence.
[00:13:47] Dan Bennett: So it’s about confident decisions. And, you know, in parallel worlds, for example, if you use, the LLMs right now to help you find a car, for example, they’ll do a great job if it’s a new car, they do, they might do a good job of giving you pricing, maybe they can collate reviews, but, you know, a house and a car and furniture are the three most expensive things anyone ever buys. What I don’t think they do well yet, and maybe you’re talking about more than just the top three here in terms of expensive items, or important items, I don’t think those LLMs give you confidence yet, because the answer might be different if you ask it again immediately after. It might be different if you prompt it slightly differently. What we want to do is be a partner with the LLMs, you know, we certainly want to be in the conversation with them. I do think down the road it’s gonna be less frequent that people are even going to sites, candidly. I think the experience may well be in a, a sort of super app, if you will. But again, you’re still gonna rely on expertise in a domain-
[00:14:50] Dan Bennett: … to ensure that you have confidence in your decisions. And we want furniture.com, we expect furniture.com to, to be that brand, and it’s a lot of what we’re doing right now in marketing is to ensure that that’s understood. So yeah, I want them to come to the site, but ultimately if they’re gonna check out in the future on an LLM or wherever it might be, it doesn’t matter so long as we are, the arbiter or the supporter of that decision, I think.
[00:15:13] Greg Kihlström: Yeah. Yeah. Well, and, and the, the LLMs are, they’re always very agreeable, right?
[00:15:17] Dan Bennett: Very agreeable.
[00:15:18] Greg Kihlström: Which is, which is nice, and-
[00:15:20] Alexandra Seaman: Depends which one you talk to. You get different levels. (laughs)
[00:15:23] Greg Kihlström: (laughs)
[00:15:23] Dan Bennett: (laughs) Different, different levels of sycophancy.
[00:15:25] Greg Kihlström: Fair enough. Fair enough. But, but yeah, to your, to your point, I mean, the context becomes important then too. You know, they’re, they’re doing their best, the LLMs are, they’re predictive models, so they’re doing their best to predict, but they don’t know everything about, they don’t know, unless you feed your floor plan of your house to them or something, they don’t know all of those kinds of, or, or taste, right?
[00:15:46] Dan Bennett: Furniture’s a very, it’s a very creative business.
[00:15:51] Dan Bennett: I mean, Alex knows this extremely well, but creativity, and design, and color, and texture, and character of furniture, it’s almost, it’s almost immune to technology. It’s such a human characteristic that you really have to trust what you’re being told, I think. And, and so that’s beyond just the price point. If a sofa’s $2,000, that’s a big ticket item by anyone’s standards.
[00:16:17] Dan Bennett: So we have to, our deep, deep, deep furniture expertise really is what ladders up to all of this.
[00:16:21] Alexandra Seaman: And I think, you know, like if you think about furniture.com, how does it go beyond, like, technology, AI, predictive to help you make a decision, and by the way, there is a lot of, AI sort of, driven components that, that do-
[00:16:35] Greg Kihlström: Of course. Yeah.
[00:16:35] Alexandra Seaman: … that do help you do that. But I think it’s like, it’s sort of in this category, it’s an art and a science. Science would be the technology side, but then the art side is like the, you know, furniture is visual, furniture is design-y, it’s merchandise-y, you know what I mean?
[00:16:54] Alexandra Seaman: And so you think about just the foundation of furniture.com, the fact that we’re working with 90 trusted top retailers, you know, nationally, and the fact that we are focused in this one core category, and our team comprises lots of technologists and, and marketers, but also a lot of people who do come from, like, the furniture space, from lots of different brands, both online and brick and mortar, and come with those expertise and thinking about, okay, how do we train the AI? How do we train our Dottie, our, you know, our chat, our LLM, so that she’s a design expert, and a merchant, and, you know, your stylish best friend? That, that’s what I sort of say about her-
[00:17:35] Alexandra Seaman: … beyond just, like, an LLM who sort of has good answers to everything. She actually understands the category, which gives much richer context to the answers, and I think more confidence and, and also just a better outcome.
[00:17:48] Greg Kihlström: Yeah. Well, yeah, ’cause I, I think, you know, another big challenge with, you know, just that endless aisle or just the, the unlimited choice is, there’s the, you know, customers also bought this, or there’s all those recommendations that, I mean, you could just keep scrolling all day, right? And, and not all of them are necessarily rel- you know, if you go to a large retailer or whatever, you know, not all of those, those… … elements are relevant and also, you know, my definition of… If I say I like modern furniture, w- what does, you know-what does modern mean to Greg versus-you and you, you know? So, I think we get into a challenge of, you know, how- how do you solve for that individual taste in- in this kind of search?
[00:18:36] Dan Bennett: We have a pretty good solution to it. We have a, a… as Alex has alluded to, we have a AI interface we call Dottie.
[00:18:45] Dan Bennett: and that product has been built specifically to understand you as a person. And so you might say, you know, “Greg wants a 92-inch modern leather sofa.” now, you’re a man of taste. You probably know what a modern-
[00:19:02] Greg Kihlström: (laughs)
[00:19:02] Dan Bennett: … sofa is. But there are some people that think they want modern, and that sofa pops up in the Dottie search, and they realize, “Do you know what? I want a tufted arm, and I want a-“
[00:19:11] Dan Bennett: “… a, I want a walnut leg.” And- and so suddenly, they, w- we help take them to a place where actually, this is your style. And so it doesn’t almost need a name at that point because it’s just Greg’s taste.
[00:19:21] Dan Bennett: And, ultimately, that’s what we’re there to do. Again, confident decisions. How do we help you do that? And whoever thought up the endless aisle, I mean, that- that’s an insane idea. I think it’s horrible way of thinking about shopping for anything. It’s like a nightmare. the my favorite stores are the ones where someone’s taken the time to curate or think about who I am as a shopper, and so I don’t have to s- scroll through a thousand things. I can get straight to the heart of it. (instrumental music)
[00:19:50] Greg Kihlström: So, let’s talk a little bit about measurement here, and, you know, certainly there’s gonna be some traditional… I mean, you’re trying-
[00:19:56] Dan Bennett: We d- we do all the boring-
[00:19:56] Greg Kihlström: Yeah.
[00:19:59] Greg Kihlström: But what, what else, you know, what else is a measure that, that maybe, you know, either non-traditional, or you know, how do you measure that, that decision layer a- and know that you’re giving people the right recommendations?
[00:20:11] Dan Bennett: It’s a, it’s the multi-multi-million-dollar question to some extent. so yes. We do all the, standard stuff you would imagine, especially from a site and, and a marketing point of view in terms of understanding the quality and health of our audience, our audience health. the level below that for us, which it gets a little bit more into the art piece that Alex was referencing, is, we wanna know from a journey point of view, yes, I want my performance marketing to drive a shopper to a PDP and get him checked out. I wanna be able to service that, of course. But at a deeper layer, I wanna know, if we’re not doing that, are people spending time on the site? Are they interacting with content? How many pages are they interacting with? Are they returning to the site? Are they u- utilizing reviews? How many prompts are they putting into Dottie? and we, we matrix all those things so we have an overall, I guess you could call sort of trust score or confidence score on the site-
[00:21:06] Dan Bennett: … that is triggered by, those, human behaviors on the platform. There are businesses I’ve been familiar with in the past where they’d be looking to shorten all of those things because sh- truncating that gets them to a sale more quickly. it isn’t that we want people to spend more time than they need to. It’s that we want them to ensure that they’re building that confidence. And honestly, spending time with us is how we think they do that.
[00:21:29] Greg Kihlström: Yeah, yeah. ‘Cause that, I mean, if, if they’re able to do that, then I assume they come back for more to-
[00:21:35] Dan Bennett: They come back.
[00:21:35] Greg Kihlström: … complete the… yeah.
[00:21:35] Dan Bennett: That’s right, for the next piece. We also do, you know, we, we, we do ongoing MPS studies. We do, awareness brand lift studies. We do a lot of questionnaires, and we do a lot of in-person consumer work, so we do speak to individuals. But we can see it in the data. If you’re, if you come through a piece of content, you viewed the content, you clicked to another object, you have a few prompts, by the time you get to a checkout, that point, I think you’ve pretty bought into the, the platform, and that, I think shows trust. And we have to protect that. We’re very, we’re very careful with that.
[00:22:05] Greg Kihlström: Yeah, yeah. Well, an- and part of that, I, I know is the partners that you work with as well, right? I mean, you’re, you’re careful about the partners you work with. Can you talk a little bit about, you know, how do you maintain that relationship and, you know, are, you know, what kind of, what kind of things do you share that help build that relationship?
[00:22:22] Alexandra Seaman: I think, you know, I help, I help lead our partnerships and sales side, and I think it… There, there is a part of it… Furniture is a interesting industry in that, like, out of any other big category, there’s the most, like the highest magnitude of family-multi-generational- family businesses. And so in a way, it like kind of all feels very familial, very friendly. And so, when we started furniture.com, we, we… Some of the other retailers, we sort of knew just like through the industry, or I- I- or like, you know, my, my brothers knew or whatever, and we met them and we said, “This is the platform.” By the way, it wasn’t live. “So this is the platform we are going to, basically we’re building and we’re going to launch. And would you guys be interested in something like this? Do you think it would add value to you? Do you think it would add value to the shopper? Who you guys know better than anybody,” right? Like, and, and, “How could we add more value?” And so we established with sort of what we call our launch partners, about 10, 15, retailers, like a true feedback loop, because they really know this industry-
[00:23:27] Alexandra Seaman: … decades and so millions of shoppers. And, and so then we, we really, I think, launched and have continued to iterate on a better platform. And then we have a very kind of white glove relationship with the retailers that we work with. I mean, of course, as we’ve gotten bigger, it’s expanded a lot. But, like, we have a partnerships team in house. They’re amazing. They speak to these retailers all the time. Like, they’re always on calls with them.
[00:23:52] Alexandra Seaman: and I think we really also, you know, measure by the quality of the furniture. Like, they need to have good quality furniture. They need to have great customer service, good logistics in delivery, because of course on furniture.com, we facilitate this. Like, a shopper checks out on furniture.com, but because it’s an agentic cart, it’s then, the rest is up to the retailer. And so we wanna make sure that that shopper experience that we deliver, which is a great one, is gonna continue. And so we, you know, we evaluate partnerships based on those things.
[00:24:25] Greg Kihlström: Okay. Yeah. Yeah, I mean, that, that partnership, it, it goes both ways, right? So, you know, they need to deliver a great product, but also… And I would imagine, you know, given your description of the industry- There might be a s- a little skepticism on, you know, some internet startup that, you know, comes along. But I think it sounds like you’ve been able to mitigate that with, you know, not only your ties to the industry, but also just building a product that… I mean, it sounds like it kinda… it gets the industry, right? I mean, it was built for a purpose, right?
[00:24:56] Dan Bennett: I think it identif- we identified early that there was a real, there was real white space for it. We’ve said before, the reason the white space exists is ’cause it’s very difficult and it’s expensive. Probably the first thing we did right, outside of, as Alex said, having, you know, great conversations with amazing retailers, we’re, we’re about 85 people now and they are completely world-class. You know, we have a data science team, an engineering team, and a PM team, and a partnerships team, and a brand team that have come from the biggest, some of the biggest brands in technology and furniture, and the broader brand world. And so when we get on a call, especially even in the early days, we’d get on a call with our head of data science, who’s a PhD, and you can’t fail if you’re one of these retailers to sit and speak with the asshole, in this case-
[00:25:45] Dan Bennett: … and think, “Holy mackerel, they’re building something impressive here.” So I think we, we were lucky enough to have a killer team, and it’s led to an amazing product, and I think what w- will now be the def- the defining brand in furniture.
[00:26:00] Greg Kihlström: Love it, love it. So let’s talk a little bit about the future here. You know, take this out five years or so. You know, what’s the, what’s the typical furniture shopping experience like for, you know, for a consumer and, and maybe the… what’s… how does Furniture.com differ from that?
[00:26:19] Alexandra Seaman: I mean, I think the typical… I mean, even still today, even with, you know, technology and e-commerce and AI, the typical shopper, largely still goes into a store at some point.
[00:26:32] Alexandra Seaman: M- at least, you know, more than three quarters of people will go into a store at some point. And it makes sense. It’s very tactile, very visual product. And I think that stores, unlike maybe other retail categories, I think stores will always be a crucial part, as they should be probably. and so that’s something that I… I mean, some people say, “It’ll get totally disrupted.” I actually don’t think it will at all, but I think that helping people, around the store experience, right? So of course there’s a large chunk of people who just want to buy online, and that’s growing and growing and growing with, you know, the-group of consumers that’s aging up to, to that buyer right now. But I think it’s like, okay, I need to discover, then I need g- help guide, being, like, guided towards my decision, and then maybe I pop into a store and see it or sh- or… Then I get back on, and then I want to, like, look at different price ranges and maybe I need help merchandising the room, and so I think the shopper in five years, the j- overall, like, mass shopper, will be much more comfortable with it. Like, I think AI will be ingrained in how they search. Not just a subset. Everybody. And I think we’re sort of on the cutting edge of that in the category. it’ll probably become more widespread and, I think my ambition, our ambition is that we continue to kind of lead that… For that furniture shopper, it’s not every furniture shopper, but for, for our segment of shoppers, like, we wanna remain on the cutting edge, remain kind of the best.
[00:28:00] Dan Bennett: I, I think if you don’t know which brand you’re looking for, and we… this is what we say to our partners all the time, if you’re looking for a Lulu & Georgia or Bloomington Hills or Rooms To Go, if you’re looking for those brands, we want you to go to those brands.
[00:28:16] Greg Kihlström: Yeah.
[00:28:17] Dan Bennett: most shoppers don’t search by branded furniture. They shop by attribute. If you’re not shopping by brand, I think in five years time, people will be going to Furniture.com first.
[00:28:29] Dan Bennett: Because if I’m looking for a flight right now, I go to KAYAK. If I’m looking for a car, I go to AutoTrader. If I want to stay in a hotel, probably go to Hotels.com. I think in furniture, people will go to Furniture.com first. I think the caveat on that is that it might be that they’re interacting with us in an LLM environment.
[00:28:50] Dan Bennett: But I think the confidence in our brand and expertise gives them… will be present, whether it’s on the site, in social, in LLM or whatever. But I think we’ll be the first stop.
[00:28:59] Greg Kihlström: Love it. Well, thanks again both for, for joining today. One last question for you before we wrap up here. I’ll ask it to each of you. what do you do to stay agile in your role? And how do you find a way to do it consistently?
[00:29:13] Alexandra Seaman: Besides running? (laughs) I’m just kidding. Um- … I think for me, it’s sort of the, the contact switching. Like, depending what the most important thing is that month or week or day or hour, that’s what I really try to, like, zone into and, and work on. And so sometimes it’s product, sometimes it’s, like, technology, way beyond, you know, (laughs) way beyond what I can really do as well as my colleagues, but… And then sometimes it’s marketing and sometimes it’s partnership, so I think it keeps me hopefully pretty agile.
[00:29:46] Dan Bennett: I was thinking about this question. What I realized, what I think we all do as a business, but I think I do particularly, or in answer to this question, is we spend a lot of time looking at what other industries do and what other, how other consumers behave in those industries. We spend less time looking at what furniture, the furniture industry does, from a behavioral point of view. Because we’re, we’re building something we think that, reflects consumer behavior as it, as it is evolving. So I think we pay a lot of attention to those other industries and we have folks that come from technology industries adjacent to furniture, and that, when you bring that back into furniture, it helps us stay massively agile. I mean, it’s kind of intimidating sometimes, but it does help us stay, I think, in a, in a mode that means we’re building something that’s compelling.
