abc

 

 

 

1
Thank You: On Leaving Chicago Ventures
2
The Future of Government (A Primer on Masternodes)
3
Unboxing Stitch Fix’s S-1
4
On Being Early and Being Right
5
Make the Merchant the Star

Thank You: On Leaving Chicago Ventures

The last six years have been the most exciting, important and fundamental of my life. Over that period, I graduated from business school, was blessed with three beautiful children with my wife Shira, and purchased my first home, amongst other major life events. But throughout it all, there was a constant presence – my second home: Chicago Ventures.

It’s bittersweet to tell the world that I’m leaving the team that gave me my start as a Venture Capitalist. Bitter, because it’s difficult to leave a group of wonderful friends. Sweet, because I take with me six years of learnings, relationships, and growth – and get to pursue exciting new opportunities that only exist because of the platform & exposure that Chicago Ventures provided me.

I’d like to formally thank my teammates Stuart Larkins, Kevin Willer, Lindsay Knight, Peter Christmas, Jason Duboe, Rob Chesney and Jackson Jhin for six years of lasting memories, laughs, and counseling. I’d also like to thank CV’s investment committee, Pat Ryan Jr, Adam Koopersmith, Eric Reeves, and previously, Bob Fealy, who took me and my recommendations seriously, well before they had any reason to suspect I had any clue what I was doing (pro tip: I still don’t and that’s OK!).

But the absolute hardest part of it all is walking away from the founders and teams that I’ve had the privilege of working with on a daily basis for many years. Starting a business is one of the most complex, volatile, yet rewarding pursuits any human being can engage in and I’m humbled that dozens of founders were willing to open up to me, trust me, and depend on me with their babies. I can only hope that our 11:30pm freak out sessions, our shared tears and smiles, and all out brainstorming meetings helped cement lifelong friendships. I’m grateful to have had the opportunity to learn from each of you.

Thanks for all the lasting memories:

cvparty

Chicago Ventures holiday party circa 2015

Studentfellows

Opening dinner with the 2016-2017 Chicago Ventures Student Fellows

IMG_0539_1517338412743

Going away dinner circa last week 🙂

 

The Future of Government (A Primer on Masternodes)

Happy new year everyone! As many readers know, I’ve been actively writing about and investing in the cryptocurrency sector since early 2013. And 2017 was obviously been a pretty extraordinary year for the space. Over the past few months, there’s been so much noise about bitcoin in the mainstream media, that I haven’t felt I’ve had much to add – if anything, I wanted to play a small role in reducing the hype.

Though, I’m gratified that in the past 9 months, the world has woken up to my first true loves in the crypto world – bitcoin and decentralized apps – I wanted to begin sharing a theme I feel is underreported and comparatively disruptive: decentralized governance and voting.

In early 2016, I was turned on to a protocol called Dash which had pioneered a new form of self-funding and self-governance through Masternodes. Masternodes are an incentive system available to large holders of a digital currency – in Feb 2016, that was defined as approximately $6,000 of DASH (which is worth $1.25M today) – that pay a daily dividend in exchange for “locking” up the coins, effectively telegraphing to the broader network that those coins are “stored,” and can be used by the protocol itself for network functionality for an extended period of time.

In exchange for the dividend, the network reaps the benefits of added layers of security, stability and speed. Specifically, the Masternodes only function if they are connected to the blockchain 24/7 via a unique IP address, and serve the network as basically outsourced computing power – keeping additional full copies of the decentralized ledger on more computers across the globe (24/7), and processing transactions more quickly.

Masternodes perform one additional service: they provide each holder with a single vote in how a protocol utilizes its financial resources. Masternode protocols have a built-in deflation rate (similar to bitcoin) that typically ranges from 5-100% to start. Some percentage of the annual inflation rate returns to the protocol itself to fund projects that would be accretive to the overall platform economy. Think of it is this way: if these crypto communities are basically tribal digital governments, then each token holder is a citizen, and each Masternode holder is a Mayor or Congressional Representative.

masternode1

As most of these protocols are still in relative infancy, the types of projects that are funded vary substantially. Projects regularly funded include things such as 3rd party developers looking for seed funding to build an application on top of a given platform, a community manager for a protocol looking for funding to attend a conference to espouse the benefits of a given technology, or even just a budget to pay a protocol’s monthly web hosting costs. Although there are dozens of emerging Masternode networks (listed here), I am currently invested in three: Monetary Unit, Crown and Bitsend*. As a practical example, here’s a current screenshot of Crown’s voting proposals:

masternode2

To help understand the above, there are currently 9 proposals that have formally passed by the required vote threshold. There are an additional 8 proposals at varying levels of popularity. The totality of approved proposals represents 25,150 Crown tokens, which at $3.10 each is $77,965 total funding. There is a new round of voting each month.

I can’t help but be fascinated by this emerging mode of digital governance and these miniature tribal governments that are springing up across the globe. In contradistinction to traditional capitalism where the wealthy’s interests are rarely aligned with poorer citizens, the wealthy in these networks (Masternode holders) are incentivized to vote for services that improve the economic benefit of literally every token holder (citizen). As more prospective citizens recognize the value inherent in a tribal government’s tokens, they are incentivized to purchase those tokens and accordingly reap the benefits.

Meaning that while today these budgets mostly revolve around community management, application development, etc, there’s no fundamental reason why they couldn’t expand to include health benefits, housing subsidies or other governmental aid in the future. The better the benefits, the more buyers (citizens), and the better the dividend to the voting class. This is circular – with the rich/poor capitalist alignment remaining irrespective of cycle.

Further, I believe this is a more sustainable model for new protocol funding than ICOs. ICOs enable a startup crypto protocol to raise a large amount of funding pre-launch or at launch with zero information available to the prospective investor about its future utility, usage or engagement. Masternodes enable protocols to provide dividends to large supporters but also force a higher level of inflation on the broader ecosystem. In order to survive, adoption of the protocol/token holders must increase quickly to offset the high level of inflation in this digital government (or else sellers will outweigh buyers and its value will go to zero). It’s akin to a startup that must hit multiple growth milestones year over year in order to reach its next level of funding (or else go to zero).

While I expect 2018 to be an extraordinarily volatile year for cyptocurrencies, one thing is clear: what started out looking like a toy, is now the next big thing.

*I am not a registered financial advisor. These protocols are stated for informational purposes only.

Unboxing Stitch Fix’s S-1

The following article was initially published by TechCrunch on October 23, 2017

When Zulily went public in 2013, it sparked a resurgence of interest in an otherwise lagging e-commerce market. Quickly trading up to reach an $8B market cap – valuing the company at over 10 times revenues – Zulily became the poster child of the flash sales boom. Over the ensuing months and years, concepts such as curated commerce, conversational commerce, flash sales, contextual commerce, digitally native vertical brands, subscription commerce and assisted commerce became a frequent part of the startup lexicon.

But of all those e-commerce segments, one stood head and shoulders above the rest: assisted commerce. And for one simple reason; as Amazon devoured an increasing amount of the online purchasing economy, their margin structure forced them to underinvest in human service. This left a hole around verticals that demanded expert guidance through the purchasing funnel – a funnel which had often gotten harder to serve on mobile devices. Beneficiaries of this opportunity included both Stitch Fix and Trunk Club in apparel, and Havenly and Laurel & Wolf in furniture, for example. [I referenced additional reasons 18 months ago on TechCrunch in The Middleman Strikes Back.]

For consumer tech enthusiasts, Stitch Fix’s S1 represents the most exciting IPO since Snap and a rare peek under the covers of one of the most cash efficient e-commerce unicorns in history. It’s important to remember that S1 filings actually carry less data and detail than your average Series A fundraising deck, and Stitch Fix’s filing is possibly the least detailed prospectus I have ever read – so the document leaves us with more questions than answers. But there remain some fantastic insights which we’ll unpack below:

For the Love of Organic Growth

Stitch Fix may be the most cash efficient e-commerce marketing machine since the start of the modern Facebook marketing era. Here’s the punchline:

1-Marketing-Rev

Stitch Fix’s numbers – especially for FY 2016 (and presumably prior) – are so far outside the normal distribution, that they look more like the growth of a viral social app than a commerce company. My personal guess is that Stitch Fix, with its differentiated & innovative experience, still saw much of its lift from Facebook – but built a secret sauce around Facebook’s tagging features, enabling them to leverage Facebook’s social graph for hyper growth, without being dependent on all of its costs.

However – as the novelty of Stitch Fix’s product offering has waned – that virality and customer excitement appears to be shifting to a more traditional marketing mix. Marketing costs nearly tripled in 2017, yet the company grew only 34%, meaning that each dollar became materially less efficient. That change in efficiency implies that they are facing headwinds in paid marketing for their first time ever. It also loosely confirms that Stitch Fix was leveraging a non-traditional channel for its early hyper growth and must now cultivate more traditional, sustainable marketing facilities.

The Company’s S1 intentionally obfuscates every other marketing metric an analyst might care for: the number of customers acquired annually, the average number of orders per year, even their customer acquisition cost. Nevertheless, we can make some rough assumptions:

The company tells us that they had 2.2M active, purchasing customers in the prior twelve months (2017), up from 1.675M active customers in 2016. However, they don’t give us any insight into churn or customer overlap from year to year. They do provide a somewhat ambiguous, but compelling statistic: that 86% of revenues in FY 2017 were driven by existing customers (meaning a customer who had purchased a fix somewhere between 2012-2016).

If that were the case, I’d estimate the company’s marketing mix as follows:

  • 280,000 net new customers acquired in 2017 [1]
  • Blended Customer Acquisition Cost of $280
  • 2017 Cohort Payback: ~3 years [2]

The problem with the above analysis is that it treats any prior customer in the Company’s life as an existing customer, even one who had only ordered once five years prior. If we want to take a more generous approach and treat any customer – whether net new or re-activated from pre-2016 as a new customer – we can do so simply by subtracting their stated 2017 active customers less their stated 2016 active customers. It’s also reasonable to assume that some percentage of 2016 actives didn’t order in 2017 – we’ll use 30% as a placeholder. If that were the case, the company’s 2017 marketing mix would look something like this:

  • 938,000 new customers + re-activations not active in 2016
  • Blended Customer Acqusition Cost: $75
  • Average 2017 Cohort Spend: $500 [3]
  • 2017 Cohort Payback: <6 months [2]

It’s an imperfect science, but a 9 month payback for any business, especially one at Stich Fix’s scale would be extraordinary. I suspect the truth is somewhere in the middle – but anything under 18 months would still be very healthy of a business above a billion dollars in revenue.

But Slowing Growth

Stitch Fix is heading into its IPO as the slowest growing of any of its e-commerce peers:

2-annualrev

At the same time, as reflecting in the marketing section above, the Company has consciously elected profits over growth.

In a sense, growth is a bit of a vanity metric as it can be inflated unsustainably. Although Stitch Fix’s 2017 suggest it is battling rising marketing costs, the Company is making a bet that the public markets will reward deliberate, efficient growth over high burn hyper growth.

But Investors will need to decide whether these choices are due to prudent, careful operations or unsustainable, systemic customer acquisition conditions in more competitive categories such as men’s and plus size fashions. If the acquisition cost reality for a new paying customer is anywhere near our first estimation, then it would make me nervous of Stitch Fix’s economics in the short-term while they fight for market share in emerging categories against a variety of extremely well funded competitors [more below.]

Unit Economics

Assisted commerce is actually a rather complex business with multiple layers of costs, specifically:

  1. The cost of the product
  2. The cost of the experts (in this case stylists)
  3. The cost of logistics

What differs from traditional e-commerce are the costs related to: (a) the experts and (b) the return shipping. E-commerce companies typically experience a 20-25% return rate whereas assisted commerce companies are roughly around 80% – which makes sense given that they are sending you a variety of items and it only takes one unwanted piece to warrant a “return.”

That said, Stitch Fix’s filing again intentionally blends much of this data together, but we’ll do our best to peel the layers back.

Let’s start with their gross margins: Stitch Fix boasts 45% gross margins which is in line with traditional retail margins, but significantly lags mature e-commerce players, especially those that source extensively from independent merchants –[ normally between 55-65% gross margins.

3-retailermargin

But Stitch Fix’s margin accountings are tricky – not purely inventory retail margins as they take into account the net effect of all shipping/return and re-stocking costs as well. As noted, these costs are higher for assisted commerce than traditional e-commerce because the percentage of (partially) returned fixes Is roughly 80%.

Now, again, Stitch Fix doesn’t report the costs behind their logistics or returns so there are two methods to estimate:

  • The first is simply looking at other comparable companies. Shipping, returns materials, and re-stocking labor costs for most commerce companies is between $15-20 on a blended basis. Stitch Fix is a mature, scaled company and presumably deserves to exist on the lower end of that range.
  • Many companies will offer a discount for keeping one’s entire shipment. It’s actually rather intuitive as a full keep actually offsets the return and re-stocking costs ($8-12) in addition to boosting average order values. A typical Fix has a $275 retail value (average of $55/item), so their 25% discount is contra revenue by ~$55 ($55 x 4 items x 25%) – likely ~$15 worth of contribution margin. Since they earn additional margin on the 5th and final item, this is a solid deal for the Company if it mitigates (or better) their return + re-stocking costs. It is also implies the return and re-stocking costs are somewhere in this range.

If we give Stitch Fix credit for the lower end of the logistics cost range – $15 – and make some simple assumptions around their ordering frequency, their inventory margins look something like:

4-Grossmargin

A 55% gross margin is really right where they should be for the category. Best in class would be 65%+, but if Stitch Fix is materially more popular than I’m guessing and the typical customer actually orders a fix 4 or even 5 times a year, then it’s margins range towards 60%+ and approach best in class.

Which leaves the final wildcard: how much does all that stylist labor cost?

The company buries and blends that information into their Selling, General & Administrative (SG&A) costs which is reasonable, though it would be nice to know how much leverage they’re getting from the gig economy. Stitch Fix Stylists – now 3,400 active – typically work part time and remotely and are paid on an hourly basis, $15/hour on average according to Glassdoor.

5-stylists

If this breakdown is correct, Stitch Fix would unequivocally be best in class and gaining a substantial amount of leverage from their stylist model, degrading net margins by only 4% in exchange for a substantial lift in customer loyalty, personalization and authenticity. To be honest, I suspect these estimates are low – but even if accurate – offer evidence of why this is a tough model for Amazon to emulate at scale, given that their net retail margins today hover around only 2%. 

What’s Left Unsaid

Too much, unfortunately. The filing is coy if not cryptic. And while I’m sensitive to companies wanting privacy around their secrets, especially when the stakes versus Amazon are as high as they are, there are certain metrics that should be demanded by analysts when assessing the company.

  1. LTV:CAC Ratio – This should be fairly obvious, and although Stitch Fix tries to give us insight into their Lifetime Value analysis, it’s mostly useless without knowing true CAC, nor contribution margin.
  2. Keep Rate – This is the metric to track in the category, and has historically been difficult to move the needle on. It is calculated as [# of Items Kept by Customer / # of Items Sent in Fix,] Stitch Fix does show us evidence they’ve been able to lift this metric 22% in 2 years – but we don’t know from what base.

In my opinion, this number is doubly important for Stitch Fix, a company that has continually defined itself by its data science capabilities, under the leadership of Eric Colson, and frequent blogging on their website of their personalization algorithms.

That they don’t publish their keep rate is dumbfounding to me, especially because it could be done without revealing any trade secrets. Further, Stitch Fix is seeking a substantial multiple on its revenue, hoping to be valued as a technology company, not as a retailer. But with 3,400 stylists – which feels awkwardly similar to “store associates” at scale – it needs to prove that its data science is moving the needle in ways that others cannot copy. Data is a moat. People, unfortunately, in this case, are not.

  1. Breakdown by Category – This may seem high level and less than imperative, but in my opinion, could answer a lot of questions. In the past 24 months, Stitch Fix has launched products for men, plus size, petite and maternity. Each of these categories is a venture scale opportunity in its own right as evidenced by the relative scale of both Trunk Club (men) and Dia and Co (plus size).

Stitch Fix’s growth slowed materially in 2017 and marketing became significantly less efficient which is a bit surprising given that they actually increased in the number of niches they were able to target. Investors need to be able to see how these emerging categories are performing to better assess whether Stitch Fix will ultimately be a predominantly women’s service or whether they have a shot at running the table across all the assisted apparel categories from men’s to plus size.

My suspicion is that as Stich Fix has entered these new categories, they are for the first time facing headwinds in customer acquisition costs against Trunk Club (bankrolled by Nordstrom) and Dia & Co (backed by Sequoia’s deep pockets), whereas they scaled with relatively little friction within the women’s category. It’s a battle I don’t think the others will give up cheaply.

Parting Thoughts

A Professor of mine once observed that Amazon was the world’s first mega retailer to get to scale without making a boat load of profits along the way. And Stitch Fix, in that regard, looks like the anti Amazon, historically delivering strong EBITDA results and growing quickly on its own free cash flow.

The management, the investors, the employees and the board deserve credit for building an almost unprecedented business – possibly the most cash efficient e-commerce company of the decade and taking it from zero to a billion in revenue in only six years and less than $50M of paid in capital.

While I’m hopeful that Stitch Fix’s success will usher in a new wave of e-commerce investing and interest, a better outcome would be if it forced operators and investors across the board to more thoughtfully consider business models in an Amazon world and focus on building businesses that can deliver fundamental value where Amazon cannot.

Notes:

[1] Calculated by taking 2017 revenues * (1-86%), divided by 2016 1 year cohort spend (as a baseline). It is an imperfect analysis.

[2] Contribution margin should be defined as Gross Profit less the administrative costs for merchandising, customer service, stylists, content creation, and all order fulfillment. Unfortunately, we have no true indication of this number given that most of those costs are blended and buried into a broader SG&A. For the purposes of this analysis, and based on comparable administrative costs for peer companies plus the burden of stylist overhead, I have estimated contribution margin as 25%.

[3] Estimated at $500, slightly higher than 2016’s 1 year cohort spend given that 2017 6-month spend is trending slightly above 2016’s.

On Being Early and Being Right

Over the past ten years or so, I have been relatively early to three highly nascent, wild wild west style industries: online poker, daily fantasy sports and bitcoin. In two of them – poker and bitcoin – I experienced significant professional and financial successes. In the other, daily fantasy, I lost my shirt. With bitcoin’s recent rise I figured it could be instructive to briefly explore the analogies between the three and develop a framework for spotting the emerging markets of the future.

On Being Early and Being Right

Jeremy Liew, General Partner at Lightspeed Venture Partners has repeatedly said that if one wants to spot emerging modes of communication and behavioral change, look to the products and services teen and tween girls are adopting. Similarly, I would argue that when it comes to emerging professional industries or financial markets, one should look towards how college aged kids are spending their time as well as the long-tail of the internet’s forums (including Reddit).

I believe this is the case for two primary reasons: (a) college aged kids have a substantial amount of disposable time and some variable amount of disposable income and (b) they are highly incentivized to increase their amount of disposable income – at a time in their lives when losing what little they have is not a significant loss. This combination pushes college students towards higher risk, higher return opportunities, some small percentage of which will materialize into mainstream markets and industries.

For example, although I’d played poker with my friends in high school, I found my way into online poker in a professional capacity my Sophomore year of college as the World Series of Poker was beginning to expand. It was the ideal format for a college student: success demanded thousands of hours of gameplay and continual learning and discussion via online forums. Very few adults with either part-time or full-time jobs could afford that much disposable time. My first startup, Cardrunners, where I ran marketing, had thousands of college students as customers, and many of the site’s pros were poker players who had made their millions during college.

You’re seeing a similar evolution in the world of professional gaming, or eSports; except that in eSports, the pros are often even younger, some still in high school. It is for the same reason: an adult with a full-time job, even one who loves gaming, will struggle to put in the tens of thousands of hours it requires to become great. These industries that demand obsessiveness are highly biased towards youth.

I discovered bitcoin in 2012 after the poker industry’s Black Friday as it was one of the few mechanisms to deposit and withdraw money from online poker sites. From there, I gradually became more interested in the technology, and began actively writing about bitcoin in 2014 in publications such as the WSJ and Recode as well as attending and speaking at Bitcoin conferences. I was also the first mainstream writer to publicly predict Mt. Gox’s insolvency, although, by that time, it was largely too late for most customers to withdraw their money.

During my poker years, I spent hours each day learning and communicating through the Two Plus Two Poker Forums. I still frequent them occasionally although only lurk at this point. They have a thread, over 850 pages long, all about bitcoin (there is an additional long thread about altcoins). You can watch the discussion evolve throughout the years. If you’d been paying attention to the forums back then, April 2011, and made even a small investment in bitcoin, you’d surely be a millionaire by now. [I love the first response given that it’s the same thing, verbatim, that bitcoin minimalists say now, even though it’s now trading at $3,400.]

On Being Early and Being Right2

There are other gems of niche industries and markets hidden throughout the Two Plus Two forums, although none have become as mainstream as bitcoin. You’ll likely find similar insights if you scour Reddit. We all know that one day another technology market will emerge that rivals or leapfrogs the attention and excitement given to cryptocurrency. I have a strong suspicion that early adopters will discover it through forums such as Two Plus Two or Reddit. I do worry that if that discussion moves to Slack or Telegram that it will be harder for the average person to find. I wrote about that concern here.

Daily fantasy sports, where I struck out both professionally and financially, was an industry where I simply wasn’t as personally passionate. In retrospect, I feel that my interest was a bit opportunistic. I still remember discussing at length with my friend Chris, who at the time was building the first ever Daily Fantasy site, Instant Fantasy Sports (in which I invested), about how we could port the functionality from the online poker sit-n-go format into fantasy sports and make an equivalent sit-n-go draft. I thought there were a lot of parallels, but I ultimately wasn’t excited by watching sports games all day nor obsessing over player statistics. This made it difficult to be credible either as a player (gamer) or an operator. I suspect that’s the reason I failed.

When it comes to being early and being right, I believe that passion is the ultimate insight. We humans have a lot in common. If you are passionate about an emerging industry, odds are that many other people are equally as passionate, or have the potential to become passionate if awareness is raised. Not all passions will materialize globally like bitcoin, but niche industries matter – even cosplay has developed into a $5-10 billion market. These emerging interests, industries and professions will continue to accelerate as the internet increasingly democratizes communication and awareness. It’s a fun time to be alive, and if you bet on your passions, odds are that you will discover others who feel similarly.

Make the Merchant the Star

Last week, while speaking with a founding team building a new marketplace connecting buyers to suppliers, I mentioned offhand that it can actually be quite detrimental to actively advertise one’s platform as the cheapest. The insight took them by surprise and after some discussion, felt it was worthy of unpacking a bit in a post.

Here’s some context: a couple of years ago Bill Gurley flew down to spend the day with a company we’d invested in. They are a dual sided marketplace that connects buyers with independent merchants. I asked them what their biggest takeaway was from the day of dialoguing with a veritable marketplace expert. They responded: “he kept emphasizing to make the merchant the star.”

That advice has stuck with me for a long time. The truth is that many marketplaces – especially if they remove layers of middlemen or if they better optimize underutilized assets – are in fact materially less expensive than the competition. It’s therefore tempting to advertise one’s platform as the “cheapest” or that buyers can “save 25%,” both of which are extremely compelling value propositions.

But the problem is that merchants, who are typically comfortable with the status quo, and have conducted their business in a consistent fashion for a long time, are consequently hesitant of new selling channels. And they’re even more cautious if they feel that channel is focused on affecting the pricing and margins they’ve always needed to maintain their independence. No one wants to become a commodity.

That framework has always brought me back to Gurley’s comment: “make the merchant the star.” Specifically: what tools, products, data or reach could you provide to a merchant – whether a seller on Etsy, a restaurant on Grubhub, or a business on Yelp – that allows them to demonstrate themselves or their products on a level previously inaccessible to them.

Consumers are smart. If they’re used to spending $50 on a product that is now available for $35, they’ll quickly recognize they’re getting a bargain. Within our portfolio, we’ve seen companies outmaneuver their competitors – and win the loyalty of the supply side – by communicating to prospective buyers benefits such as convenience, breadth of product, or quality of suppliers. All of those messages are still compelling value propositions for the consumer that do not directly threaten the merchant.

All that said, marketplaces that sell a commodity experience such as Uber, or a managed marketplace that is its own seller should certainly focus on discounting as a consumer value proposition and aim to reduce take-rates to the point they can’t be undercut. But marketplaces that connect buyers to any form of unique merchant or product would be well served to make them the star.

Copyright © 2014. Created by Meks. Powered by WordPress.

Get BreakingVC To Your Inbox

Join hundreds of other operators & investors who get BreakingVC updates directly to their inbox.
Email address
Secure and Spam free...