Dressing Down the FarFetch IPO
Earlier this week, UK based Far Fetch filed to go public, with media experts expecting an ultimate public market valuation above $6B. You’d be forgiven if you actually didn’t know much about FarFetch – it’s largely been a non US focused story to date. But you’d also be remiss if you didn’t study the business: at a $1B+ GMV run-rate, with large marketplace take-rate margins, and sub six month payback periods, it’s one of the few truly globally scaled digital marketplaces operating today. FarFetch’s IPO is exciting because it’s the first digitally native fashion startup to go public since StitchFix’s late-2017 IPO.
Led by a highly charismatic founder, Jose Neves, the business zigged where the rest of the market zagged. While startups from TheRealReal to Vestiaire Collective to ThredUp went to war, each raising hundreds of millions of dollars to scale managed marketplaces for high end re-sale or consignment goods and accessories, FarFetch instead pursued direct relationships with independent boutique retailers and later large luxury brands (such as Gucci) in order to bring their inventories online and aggregate all their products into a centralized marketplace.
What’s clear is that consumers love the business. But what’s unclear is why it’s losing so much money, and what it would take to turn a profit. Below we’ll dive into the FarFetch F-1, get a better understand of the business, and assess its health with comparisons to other globally scaling marketplaces and retail players.
Not FarFetched At All…
FarFetch was founded in 2007, just months before the global recession began to take shape and is an additional data point in the consensus that recessions are an ideal breeding ground for innovation. At 11 years to IPO, FarFetch actually slightly lags many of its peers in time to IPO – likely reflecting the emerging reality of mega-rounds and increased availability of lightly dilutive capital (the Company has raised upwards of $700M, including $387M from JD.com):
As an aside, while debates around Uber and Airbnb’s time to IPO have raged since 2015, they are actually roughly on par with their peers’ time to IPO, Airbnb having been founded in 2008 and Uber in 2009.)
Before diving in to the Company’s specific metrics, it’s important to actually understand the context and background of FarFetch’s business. Whereas consumers typically only see the marketplace front-end, the business actually has a myriad of offerings which have the effect of deeply integrating the supply side of their business:
1. FarFetch Marketplace: the Company’s primary transaction portal where consumers may buy product directly from the marketplace’s suppliers. This makes up approximately 90% of FarFetch revenue.
2. Browns: two retail stores in London which sell goods indicative of the offering on FarFetch. The stores together comprise about 5% of FarFetch revenues.
3. 1st Party: under the umbrella of its Brown’s retail stores, FarFetch actually sells some amount of product, direct to consumer, through its own marketplace. In 2017, this represented a surprisingly large percentage of total marketplace sales (more below).
4. Black & White: a full suite of white label tools for both Brands and Retailers that integrate seamlessly into the FarFetch marketplace, allowing suppliers to use these tools to manage their native businesses as well as their FarFetch marketplace sales.
5. Storm & FFLink: these are seller focused tools that enable them to integrate their inventory management, data dashboards, payment APIs and more directly into FarFetch.
6. Platform Fulfillment: the Company invests aggressively in assisting their suppliers to optimize of logistics costs and friction, by providing them data on the fastest and cheapest means of shipping goods, as well as provide concierges for customs clearing needs, given that over 90% of transactions are cross-border. (This is also imperative because the business holds minimal inventory, so customer delight depends on ensuring their suppliers operate efficiently).
7. Content Creation: FarFetch owns and manages four production studios across Europe, US, Brazil and Hong Kong that develops substantial original content for brands and retailers, showcasing their products through the help of professional models, stylists and photographers. In 2017, they processed almost half a million products.
8. Customer Service: FarFetch manages the vast majority of customer service requests, needs, and returns considerations in-house via its own service department.
FarFetch aggressively curates its retail partners, limiting the platform to 980 sellers as of mid-2018 and charges absolutely nothing for each service they provide to their suppliers, with the exception of the take-rate (commission) on the goods they sell through their marketplace.
The effect of these suite of services is twofold. First, the strength of integrations and data accessability actually provides a substantial amount of multihoming costs (the friction/financial loss associated with transacting on multiple marketplaces) to the Company’s suppliers. Why? Because each of these suppliers has invested their own IT resources into these integrations, have a strong logistics partner in FarFetch, and benefit from substantial amounts of data about their customers and purchasing behaviors. (Not to mention, 99% of top suppliers actually sign some form of “exclusivity” with FarFetch for the benefit of these services).
Second, because FarFetch is giving away so many of the services listed above, they are a “value-added intermediary” - effectively a managed marketplace - even though they never take ownership of any inventory or actively process any logistics. In my 2017 piece for TechCrunch, Anatomy of a Managed Marketplace, I noted that managed marketplaces have “a take-rate (gross margin) that is a significant premium versus other buy/sell options in the market in order to offset the premium service level or risk transfer that has occurred.” That is actually what FarFetch looks like: it boasts 30%+ take-rate margins (on par with fully managed marketplaces such as TheRealReal), even though it never touches a transaction.
These margins are possible because of the nature of the luxury industry, where brands typically command 65+% gross margins when selling direct to consumer:
Given that FarFetch manages all demand generation, content creation, logistics optimization and customer service, these brands are willing to eat in to their extremely healthy margins as those are costs related to goods sold they would otherwise bear on their balance sheet.
But providing so many value added services to their brand and retail suppliers isn’t cheap and it’s ultimately one of the massive question marks around FarFetch’s businesses.
World Class Metrics & Strategy
Before analyzing all of FarFetch’s costs, let’s first take the time to understand its customers and high level narrative. FarFetch has continued to grow quickly, with top line GMV in 2017 accelerating about 55% over 2016 from $586M to $910M, and its Adjusted Revenue following suit at 61% growth; overall, growth compares well to its peers who have gone public over the past several years:
*Grubhub merged with Seamless midway through the year prior to its IPO and its growth rate is slightly accelerated due to that change.
FarFetch also appears to compare fairly well on one of my favorite metrics, marketing costs as a percentage of revenue. The business has shown increasing efficiency in its marketing spend, even as it’s spending more, implying that its network effects are beginning to kick in:
(Direct to Consumer retail businesses such as StitchFix and Zulily will typically have a better Marketing:Revenue ratio because they record the entirety of a sale as revenue whereas marketplaces such as Etsy or FarFetch earn only a commission on each sale.)
The business provides an impressive amount of transparency about its operations, economics and metrics, arguably the most information I’ve seen to date from a digitally native marketplace. Given the marketplace’s scale and 50%+ annual growth rate, those metrics look good too. Although the company breaks out only a high level CAC, I’ve taken the liberty of consolidating all key metrics intro a concise table:
The take-away from this table is twofold: (a) FarFetch has maintained an enormous amount of marketing efficiency, even as they’ve nearly doubled marketing spend – having maintained a payback period of approximately six months based on contribution margin and nearly 200% contribution margin against marketing within year one. (This is largely in line with what the Company self-reported, but did not break out itemized.) and (b) Existing Customers are growing their share of GMV, suggesting that the company’s depth of SKUs and customer service is meeting customer expectation.
Importantly, FarFetch also has some intriguing marketplace characteristics. With only 980 suppliers in their marketplace, fulfilling nearly 1.9 million orders in 2017, it’s clear that FarFetch has cultivated a captive set of power sellers, a pre-requisite for any healthy marketplace. But unlike many other platforms, FarFetch also has an impressive cohort of power buyers.
The Company notes that 1% of its active customers account for approximately 20% of its GMV. Which breaks down as follows:
It’s an exciting metric, because it reflects the continued opportunity for non-linear marketplace growth, and is a benchmark more commonly seen in B2B marketplaces than consumer marketplaces. In fact, many of these buyers are likely small businesses themselves, from smaller boutiques purchasing additional inventory to fill their shelves, to personal stylists purchasing on behalf of clients. And if we assume that these buying habits have roughly held true over the past three years, these power buyers are increasing their presence in the marketplace:
The danger to power buyers, especially those that represent smaller businesses, is that they are likely to be price conscious shoppers. FarFetch currently maintains exclusivity arrangements with many of its sellers, somewhat insulating itself from comparison shopping – but if similar product can ultimately be procured elsewhere at a lesser cost, they risk losing a material portion of their revenue.
Private Label? Or Just Private…
Although we tend to think of traditional retailers such as Macy’s and Nordstrom in the context of brick and mortar, they are actually aggregators and marketplaces at their core. Much like FarFetch, they offer a wide variety of goods from a variety of retailers and earn an effective commission on each item sold.
Long ago, they figured out of the value of private labeling their own brands – that if they could actually own the inventory, they could earn margins that more than doubled their traditional retail margins.
It appears FarFetch may be no different. Except for their efforts to downplay it.
Although the Company attempts to obfuscate its history and details, FarFetch has begun subtly selling their own inventory in their marketplace, under the umbrella of their brick and mortar retailer, Browns. As I’ve noted, this a prospectus that is definitely not lacking in detail, yet the first mention of these First-Party sales appears on page 69, with the following: “To a lesser extent, we generate revenue from the sale of inventory on the platform that is directly purchased by our Browns boutiques and sold online where revenue is equal to the GMV of such sales.”
Except that the “extent” to which it matters may be rather material. FarFetch provides no indication of how long they’ve been selling in their own marketplace (it would be 2015 at the earliest, the year they purchased their brick and mortar presence, Browns), but if we hypothesize that the vast majority of this program scaled up in 2017, these First Party sales would represent almost 15% of all of FarFetch’s 2017 GMV growth. (Lending credibility to this hypothesis is that the only “Buyers” at FarFetch I could find on LinkedIn were hired in 2017 or later.)
These native sales in fact appear to be increasing, with a 400% increase in Company cash outflows in the first half of 2018, compared to 2017 – “Net cash outflow from operating activities increased to $106.0 million in the six months ended June 30, 2018 from $26.0 million in the six months ended June 30, 2017, an increase of $80.0 million, driven by an increase in the loss after tax from $29.3 million to $68.4 million and anincrease in net working capital due to an increase in first-party inventory prepayments or deposits ahead of the autumn/winter season.” (emphasis added). Though of course the prospectus wants to only focus on the after tax loss (an increase of $39M), simple math suggests an increase of $41M to purchase inventory to sell in its own marketplace. Assuming 50% margins on these goods, the Company would expect to generate $80M in revenue of this purchase order, with likely much more to come in the second half of the year with the Holiday season, a 200-300% growth over 2017.
It wouldn’t be the first time that FarFetch has made strong decisions that changed the nature of their marketplace. The Company, which started as a marketplace for boutique retailers to compete with larger brands and gain access to a more global customer, came under fire earlier this year by its early adopters for a string of partnerships with big name brands such as Burberry and Chanel that threatened to cannibalize sales away from its independent sellers.
But on a macro level, an aggressive move into First Party sales makes complete strategic sense for FarFetch. As marketplaces scale globally, they transition from the demand side constraints that early stage startups typically struggle with, to substantial supply side constraints that are harder to mitigate when you’ve sustaining millions of customers and have already penetrated a large percentage of your likely supply. It’s one of the reasons that Uber launched their own leasing program back in 2015 (it has since shut down) and why Airbnb is experimenting with launching its own hotels – a desire to begin regaining control over supply side dynamics and fully own the consumer experience. It’s a great hack for liquidity – but it’s unclear how third parties will react.
At $40M in 2017 gross sales, or 4.5% of its entire marketplace GMV, it’s safe to say that the single largest seller of goods on FarFetch isn’t any particular third party…it’s FarFetch itself.
I got 99 Problems But…No, Literally
In spite of the strength of the Company’s transactional marketplace, FarFetch’s losses are widening. Which is surprising given their increasing contribution margins, improving marketing efficiency, six month payback, and more.
So where is all the spending coming from?
Those eight value additive services listed above that FarFetch provides its suppliers, entirely free of charge, aren’t cheap. And they are a heavy burden on its operating profit. The Company provides us some indication of where the costs are coming from, but not the entire picture; it tells us that they have opened a dozen new global offices this year, presumably to be closer to suppliers and logistics partners across the globe. But it doesn’t tell us how much all the content creation for half a million new products a year costs, nor the costs of managing the entire logistics stack of their retail/brand partners. These are staffing heavy endeavors and the human capital costs are substantial. At the end of the day, we simply don’t know if the business has any operating leverage.
In theory, these costs shouldn’t be insurmountable. Zulily went public in 2013 as a five year old business on a de minimus operating loss and was profitable soon after, in spite of producing content for upwards of 1,000,000 new products per year (not to mention on tight, daily deadlines for its flash shales model).
Below is a rough table of the Company’s cost structure and where it’s fighting uphill battles:
Given our analysis above, we are mostly happy with many of FarFetch’s expenses. We like their marketing expenses because customers are contribution margin profitable on a six month payback. We like their technology expense because the Company’s suite of software tools creates stickiness on the supply side and helps keep the marketplace sustainable.
Where we’re very unsure is in two areas: Cost of Goods Sold and G&A. Again, these are two areas where outsiders have minimal visibility. We know the Company can generate customers who spend substantially more than they cost to acquire. But what we don’t know is: can they service those customers profitably while also providing concierge level service to each of their retail/brand partners?
I’ve modeled out what it would take to get the Company to profitability (my best guess at their rough plan), but to do so the following assumptions have to hold true:
Gross Margin must continue increasing, which it will if they continue to sell more of their own inventory through the marketplace.
Cost of Goods will consequently increase to reflect that they are purchasing more inventory upfront.
Marketing efficiency must remain similar with paid growth continuing to increase but scaling down to 10-12% of revenues within 3 years.
They can slow overall hiring to 15-30%/year in their service departments while maintaining 40% top line growth across their gross volume of orders, number of SKUs, GMV, etc. For context, they have grown their service department from 1,145 to 2,077 workers in the past twelve months. *This is, in my opinion, where the entire leverage of the business hinges.
And then there’s the question of what the company is actually worth. Media estimates put the target valuation at $6-8B. Here’s how that stacks up versus current market comps:
FarFetch actually stacks up fairly well on a multiples basis. It would be the fastest growing publicly traded digital marketplace I follow, and growth generally carries a premium in these markets. Comparing it to luxury retailers such as YNAP and LVMH is also favorable because of its growth rate. That said, each of the other businesses listed are now extremely profitable – and the Company’s operations suggest it will be three years, at best, until they show an operating profit.
At the end of the day, FarFetch has built an extremely impressive, strong and liquid marketplace over the past decade. Their supply side is locked in, with a buyer cohort that is increasing spend year over year. But as with many businesses in the unicorn era, they appear to have thrown dollars at certain operational efficiency challenges and the business will have to prove over the coming years that it can rein in these costs while maintaining the concierge level service they are known for with their partners.