Thrasio's Impending Doom

Amazon FBA rollups are in trouble

I'll start with a concrete prediction: within 2 years (note: sooner now - wrote the bulk of this note on January 26!) investors' stampede into Amazon merchant rollups will turn into an absolute rout. Equity will be locked up in these illiquid investments for years to come - exits will be few and far between, and at multiples far below what's currently being underwritten (let alone vs. current multiples). The "platforms" that these companies tout as their secret sauce will in most cases turn out to be nothing more than a collection of flimsy excel models, fast and cheap debt financing, supply chain messes and a fatal overdependence on Amazon search rankings. VCs who thought they could turn a quick buck will learn quickly why private equity firms manage 10x as much money as they do: by limiting downside risk, using debt prudently and underwriting to reasonable - not homerun - outcomes.

2x + 2x = 20x

The key to these models is that idea that you can buy many Amazon FBA businesses for low multiples of profit, bundle them together under some umbrella entity, and then exit (i.e. list on the public markets or sell to an acquiror) at a much higher multiple. And the deals to be had in early days of these businesses were stunning - you could buy these Amazon FBA businesses for 2x profit, time and after time. The story to investors was simple: "Proctor & Gamble trades at 20x EBITDA on almost no revenue growth. We have the opportunity to buy the brands of the future at 2x multiples and with our 30% growth and strong EBITDA margins we should get at least a 20x multiple on exit, so why shouldn't we blitz and buy everything in sight?" Since those early days, capital has flooded into the market, and as competition for deals has increased, so have multiples. So now these companies are paying 5-8x. But if you really believe you can exit at 20x, it's still a steal. The problem is, you can't exit at 20x, or anywhere close to it. Back to the in a bit.

Lions and Pandas and Penguins, oh my!

In 2011 Google had a problem: online publishers like eHow, Livestrong and About.com were gaming the Google search algorithm in order to show up higher in search results. The result was that users were increasingly unhappy - when they clicked on the top search results for their queries they were getting sent to sites with tons of ads, slow load times and in some cases almost no content (some pages had literally 3 sentences surrounded by ads everywhere).

So Google decided to fight back by making massive updates to their algorithm in 2011 (called Panda) and 2012 (Penguin) that significantly reduced the reach of these sites with poor user experiences. These impacts demoted the search rankings for these companies’ articles way down the search results page. The impact was immediate:  after going public in 2011 and hitting a $2 billion valuation, Demand (which owned eHow and Livestrong) lost 75% of it's value by 2013. It never recovered. About.com went from being acquired for $300 million in 2012 to bleeding revenue a few years later (it ultimately came back with a vengeance as DotDash, but prior to it’s recent acquisition of Meredith Corp. was still reliant almost solely on Google for traffic). And countless other sites saw their rankings plummet.

Why do I mention this? Because these Amazon businesses run the same risk that eHow and About.com did: they are completely reliant on one partner - in their case Amazon - for their traffic and revenues. These businesses have not built brands that consumers recognize: they've built product pages that rank high in Amazon’s search rankings. Those search rankings are maintained by ratings, reviews, conversion rates and a whole host of other metrics, but ultimately Amazon wants to ensure that it's customers are receiving the best experience. If one day Amazon decides that companies are gaming the system to rank higher - like Google did with eHow and About - they will have no qualms about making a change to the algorithm - again, like Google did. And because these FBA businesses have no alternate traffic sources, they will fall precipitously.

And lest you think Google is the outlier: Facebook did it too (remember the meteoric ascent and then rapid fall of Upworthy, LittleThings, ViralNova and EliteDaily?), and developers suing Apple for anticompetitive practices were so concerned about the same happening to them that they got Apple to agree to a 3 year freeze of its app store search algorithm. Every platform is going to do what's best for their customers, and sometimes that may conflicts with what's best for the people who have built businesses on top of their platform. 

Whence the Exits?

Seems like almost every day another FBA rollup company gets funded - but where are the exits? These companies are raising money at 20-30x revenue valuations, but none of them have shown that they can actually sell their business for anything close to that (or at all). Ultimately the value of a product or company is what someone else is willing to pay to buy / acquire it; and to date we've seen no exits of these rollups. Will they exit for 30x revenue or 5x revenue? Or will it be based on EBITDA - 4x or 40x? No one knows yet. But let's assume they will be able to exit - what public market signals do we already have that can point us appropriate exit multiples?

P&G They Are Not

Consumer goods conglomerates have traded on stock markets for over 100 years, so we can look to them as a starting point - think Colgate-Palmolive, Unilever, Proctor & Gamble, Kimberly-Clark, Church & Dwight, etc. What do we find? On the plus side we find that these companies have: massive scale ($5-$80 billion in revenue), strong profit margins (20-25%), incredible brand equity (e.g. Tide, Clorox, Swiffer), extensive track records and audited financials. On the negative side, the primary knock against them is that their growth only marginally outpaces inflation (if that) - 0-10% per year.

And where does all of that net out? Well, these businesses garner of 15x on the low end up to 25x on the high end. And those are EBITDA multiples, not revenue multiples. So what can now say about the FBA rollup valuations?

Most importantly, the revenue growth that these companies are seeing is not sustainable long term. Much of it is inorganic (aka they're growing by acquiring businesses), and what is organic is almost always a result of one-time post-acquisition conversion or marketing optimizations. The end result is that just like the public comps, these businesses will end up with the same growth rates as the big boys - 0-10%. The only ways to avoid this low organic growth scenario for an extended period of time are to either acquire brands who are (a) taking significant market share from established competitors or (b) creating and growing an entirely new market. The former is virtually entirely absent from the acquisitions made to date, and the latter is very rare (and usually involves a product with a relatively small market size).

Once you accept that the revenue growth unsustainable, your problems are compounded by the fact that these FBA rollups fail to come close to matching the other factors driving the consumer goods giants' valuations: infinitesimal operating histories, negative or at best subpar margins (<10%), much smaller scale, zero brand equity and financials that usually an absolute mess. And back to perhaps that most important point in this piece: they still have a massive search algorithm problem which puts their whole model at risk.

So where does that leave us?

I predict that:

1. None of these rollups will exit at a valuation above 10x EBITDA, and that even that figure will be reserved for the "successful" ones (a far cry from the multiples they raised money at in their latest funding rounds). Before we get there, though, we'll see the disintegration of many copycats, with their components sold off in pieces.

2. VCs will pull back from their investments here, realizing that they're playing in the wrong sandbox; instead, the PE funds will emerge as the consolidators (but the blue-chip names will stay away given the search algo risk).

3. We won't see a public market exit for at least 3 years (if at all); most of these businesses will be traded from PE fund to PE fund.

4. Success will accrue to talented operators rather than talented fundraisers. Bread & butter CPG skills like logistics, marketing, retail distribution, etc. will become critical.

5. Asset acquisition prices will generally stay within the narrow band of 3-7x EBITDA, with premiums paid for products that have built strong brands (even or perhaps especially if in niche categories).

And lastly, despite everything I’ve written here, investors will still look to fund the “Thrasio of X”. Why? Because early investors in Thrasio still did well, and more importantly this kind of model is extremely lucrative for founders because of the liberal use of debt (perhaps that will be a different post). We’ll see these rollups models for Shopify merchants (and apps), mobile apps, Salesforce apps, etc. - the cycle will start all over again. Perhaps this time we’ll avoid the froth…

…probably not though!