I spent the first few months of the year testing a business hypothesis I had. At the end of March, I concluded I was wrong, and the business wouldn’t work. I wrote a retrospective on the process and what I learned. Since sharing it with friends and mentors, a few people have suggested I publish it as a blog post. In the spirit of helping others avoid blundering, below is a slightly redacted version of the memo. Failure is overdetermined, and I hope there are lessons in here that improve someone else’s odds of success.
You can jump to the key lessons here or read on to understand the journey to get there.
Idea
In January, I began exploring the idea of building a business-to-business (B2B) wholesale platform for wholesalers and distributors to sell to buyers. The initial kernel of the idea came from my experience at Inventa, where we didn’t have the same long tail of independent brands to connect to retailers that Faire has in the U.S. and Europe. We began acquiring small distributors as sellers and even built an in-house distributor to support the marquee brands we wanted. When I left Inventa, I investigated competitors and saw that Qogita had raised $125M to build a clever automated brokerage platform in Europe. The market appeared large, and many new B2B platforms emerged with different models to experiment with. I had reason to believe there was something there.
Europe
Qogita enabled price improvement for buyers by aggregating supply and dispatching orders to the lowest-cost supplier. They used their significant funding to expand from health and beauty to other verticals. They had also launched in the U.S. and were testing Latin America. However, as I spoke to more ex-Qogita employees, I learned their apparent success was not what it seemed. They relied on unsustainable gross merchandise value (GMV) to achieve early growth, raise funds, and hire talent. Once this dried up, the founders stepped away, and the company laid people off. They pulled out of the U.S. and discontinued verticals outside of Health and Beauty. The company is still going, but it doesn’t provide as much validation for my idea as it first appeared.
The legitimate part of the business relied on grey market transactions and currency arbitrage between Eastern and Western Europe, something that does not replicate in the U.S. Feedback suggests that the founders were unwilling to adapt to the U.S. market and were hampered by their lack of product-building experience. Their failure in the U.S. did not indicate the idea wouldn’t work here if appropriately executed. The key would be understanding the U.S. market and finding a pain point for buyers and sellers.
Signal
One of the first wholesalers I contacted immediately replied and wanted to talk. He explained he had been a buyer and seller on Qogita and thought someone needed to bring it back to the U.S. He was trying to start a company to do it. This conversation was a solid early signal that there was something here. The recurring challenge in our calls was more clarity on the ideal customers on either side of the platform. Is it distributors selling to wholesalers, big wholesalers selling to smaller wholesalers, distributors selling to retailers, or some other combination?
I spoke to another former seller on Qogita in the U.S. who told me he had a bad experience as they did not vet buyers and sellers and needed better support. His words to me were, “It’s a good idea, poorly executed.” I thought there was a way to build a platform that would do the vetting on both sides and provide credit to buyers. That would minimize the overhead required for suppliers to maintain relationships and fulfill orders. Over time, we could get them to lower their minimums and serve downstream retailers directly, helping to cut out middlemen and lower costs.
When I pitched this idea to my friend, Eric, he got excited enough to explore it with me as a potential cofounder. We spent time together to be sure we wanted to build the same kind of company and culture. There was real momentum.
USA
I learned about the U.S. market from conversations with a helpful and knowledgeable woman Qogita had hired to lead their expansion here. Brands have much more control over the supply chain, and distributors work with them to enforce it. Grey market goods divert through Miami, and liquidation channels, but the market is generally more mature than in Europe. Online consumer demand is served through more concentrated players such as Amazon and Walmart. Brick-and-mortar is more cohesive as retailers such as Ulta and Sephora have much higher market share than European beauty chains, which tend to be country or region-specific.
Multiple interviews confirmed that the U.S. supply chain continues to mature and shorten. Brands (or their small number of authorized distributors) want to ship inventory directly to retailers and consumers without going through many layers of distributors and wholesalers. Verticalized brands do this explicitly, and large retailers have direct relationships with these brands. Quince is pioneering the manufacturer-to-consumer (M2C) retail model. Acktify allows brands to ship directly to consumers by listing inventory on third marketplaces and fulfilling orders directly from the brand or their distributor’s warehouse. These models skip the cost of a warehouse hop and allow brands to build an inventory buffer at only one point in the supply (or none at all). There are fewer small players to intermediate than in Europe, and they will keep getting cut out as distributor consolidation continues.
MVP
By this point, I had convinced Eric to join me in exploring B2B and finding a company to cofound. We built a minimum viable product (MVP) to solve the inventory problem for an independent perfume shop in Seattle. We convinced about a dozen wholesalers to give us access to their catalogs. Eric wrote a script to aggregate them into a catalog we listed on a Shopify store I had set up. Three days later, Shopify shut us down without explanation. I surmised that the brands have agreements with Shopify to prevent shops from listing products they are not authorized to distribute. This experience was a great example of the control brands have in the U.S. market.1
Eric rebuilt the site using Swell while I set up search functionality on Algolia. We relaunched and put the MVP in front of our perfume shop user. The feedback was that she would be happy to use our platform because our pricing was at least as good as what she could currently find. She was also excited about the potential for price improvement as we aggregated supply. But we could only address 10% of her buying. She purchases most products directly from brands or their single authorized distributor. She receives net terms, collateral, training, testers, and gifts by working directly with brands. And she doesn’t want to risk her relationship with them, even if we can get her a slightly better price.
We validated this with several independent perfume stores and boutiques. We even tried driving traffic to a landing page from Facebook and Reddit ads to see if the value proposition resonated with other small health and beauty businesses. The lack of engagement from these experiments told us we were probably selling vitamins and not painkillers in this market.
“If the customer isn’t desperate for your product you don’t have product-market fit. Nothing else matters. You can screw up everything else and still win.” —Andy Rachleff
In hindsight, we could have learned this without writing any code. But the lesson posed an important question: where was the segment of buyers desperate to find inventory and receive credit to purchase it? Large brick-and-mortar chains didn’t. Boutiques that stocked independent brands were already well-served by Faire. Who were these wholesalers selling to?
Desperation
Small wholesalers who buy low-priced inventory to resell on Amazon were the perfect candidates. These businesses source from closeouts, the grey market, and consumer sites. They use SaaS tools to price products to undercut Amazon and each other. Getting credit terms from sellers is a constant struggle for them. Conversely, they have to vet sellers to ensure they aren’t getting counterfeit or expired products, which risks them losing the ability to sell on third-party marketplaces. A platform that connected them to vetted supply and gave them credit to buy would solve an existential problem.
I was very excited about this idea and began digging deeper into the world of Amazon resale. Unfortunately, joining the Telegram groups and following the legitimate resale influencers on LinkedIn showed that these resellers are continually getting their products delisted by Amazon, and some are even having their stores shut down completely. They are playing a price arbitrage game that the brands want to stop. Pressure from brands is why many distributors and wholesalers explicitly do not sell to Amazon resellers, while others will but won’t ship to Amazon warehouses due to repackaging requirements.
Eric and I concluded that these resellers will be squeezed out over the next decade as more brands take control of their third-party marketplace channels directly. They will do this in-house (possibly using Acktify to help) or through their small number of authorized distributors. The resellers are filling a temporary gap left by the brands trying to hold out from listing on Amazon. They are not a market in which to build a generational business.
Buyers
We didn’t yet know who would be on the demand side of the platform. We learned that distributors weren’t excited about acquiring more small online retailers. The first of the two remaining hypotheses was that we could help them access entirely new categories of offline demand outside of their core categories. For example, helping hardware stores outside big cities expand their product lineups to provide general and convenience store products so customers don’t have to drive out of town to a Walmart or Target supercenter. There is evidence that hardware stores are trying to do this, but we determined that their existing distributors are already addressing the issue. Essentially, these local stores provide convenience due to their proximity, and product expansion will allow them to remain relevant in an increasingly online world.
The second hypothesis was that distributors don’t provide the full selection of a brand’s products as they focus on the top 30% that turnover fastest. Small retailers without direct relationships cannot access a brand’s entire catalog. By aggregating the long tail of brick-and-mortar demand and driving enough volume to distributors, we could make carrying the full assortment profitable. It may even be possible to go directly to the brands for this final 70%. While there may be something to this idea, as it is consistent with the supply chain shortening trend, our challenges in speaking with distributors and wholesalers make it unlikely that we would develop conviction on the idea quickly enough.
Liquidation
There is still a market for wholesalers to sell to retailers and vice versa. Supply is driven primarily by closeouts from retailers liquidating inventory. Wholesalers specializing in this inventory buy pallets at liquidation auctions and resell them to their networks of Amazon resellers, other small businesses, or consumers. Ghost is a B2B marketplace for retailers to liquidate excess inventory to wholesalers and discount retailers such as T.J. Maxx. Incumbent B-Stock is an online auction site for quick liquidation by mostly large sellers. Max Retail is trying to help boutiques liquidate through consumer channels. With e-commerce expected to grow 39% by 2027 and returns approaching $1T per year, the need for quick liquidation is only increasing.
Liquidation is another opaque and relationship-driven market. We heard that Ghost is doing well but has struggled with this since the beginning. Some of their GMV is likely still off-platform. This market also suffers from the same kind of concentration that plagued Qogita. Max Retail’s approach focuses on the fragmented boutique market but presents a different customer acquisition problem. That said, liquidation is a big problem, and we have spoken to others looking to build there.
Lessons
Our biggest takeaway is how hard it is to iterate and progress when you can’t easily talk to potential customers. Being an outsider to an industry can help bring a new perspective on problems and possible solutions, but it often comes with a cost. It was too hard to talk to distributors and wholesalers. We reached out to hundreds and managed to speak with only a handful. We did get some helpful interviews through Tegus, but these essentially confirmed the market trends of consolidation and shortening mentioned earlier. That said, a lot of people will help you. It never hurts to ask, and strangers will be surprisingly generous with their time, even if nothing is in it for them.
Faire’s market and business have many underappreciated and unique aspects that make the company powerful as a B2B platform. The market is highly fragmented on both sides, buyers need dozens of suppliers, novelty is critical to the value proposition, and decision-makers are relatively easy to find at independent businesses. The business model solves inventory risk, an existential problem for retailers. It incentivizes suppliers to refer retailers and has global network effects. I have yet to see another B2B platform with all of these characteristics.
On the personal side, wanting to found a company isn’t the same as identifying a pain point that a customer is desperate for you to solve. A market that seems to work inefficiently doesn’t necessarily have problems with clear solutions. It could be the case that a big business needs to be built somewhere in this ecosystem, but we have been unable to identify it. We followed Chesterton’s fence principle to understand why things worked the way they did before building solutions. What we learned led us to avoid building for Amazon resellers because they were desperate. Conversely, it could have made us too cautious to commit to an idea.
My goal was not to become a founder at any cost. I had a specific thread that I felt compelled to pull on. I am moving on now that I’ve convinced myself there isn’t an opportunity. I will start a company someday when I identify a problem that a customer is desperate for me to solve. In the meantime, I’m looking for another exceptional team to join to help build theirs. I will do so with more appreciation for just how rare great ideas are and a more profound respect for the founders who bring them to life.
Thanks to Taylor Murphy for reviewing a draft of this post.
Intellectual Property in the fragrance industry is a fascinating topic.