The Increasing Cost of Consumer Startups
How to Not Fail When Predicting the Future
The Birth of the Web 3.0 Rollup
The Economics Underlying Chatbot Mania
Ideas Matter

The Increasing Cost of Consumer Startups

If you have the dubious distinction of spending as much time around consumer facing seed VCs as I do, you likely start or end every conversation with some variant of “what’s exciting you these days?” And for the first time in my nearly 5 years in venture, the answer for the past six months has inevitably echoed some variant of “it’s been slow.”

So what is going on?

The narrative for the past several years has been that the proliferation of cloud services, AWS in particular, has vastly reduced the cost of getting a startup off the ground. But I believe the low cost days have rather abruptly come to a halt – at exactly the wrong time. Not because AWS is suddenly expensive, but because the types of consumer behavioral shifts now being targeted are fundamentally more expensive.

Although I’ve been socializing this for a few months internally and amongst peers, I finally mustered the courage to write this piece after reading Sam Lessin’s Era of Lean Startups Comes To An End and Michael Eisenberg’s The Big Disruption.* Recode’s The App Boom is Over also plays a supporting role. Here’s the punchline: the low tech, low cost, low hanging fruit of the digital (and subsequently, mobile) era – digitizing or mobilizing offline to online processes – are coming to an end**. And until the next great platform or behavioral shift, the costs of innovation will be material higher. As we used to say in my card playing days, “the price of poker just went up.”

Here’s how Sam Lessin, writing at The Information puts it (pretty graph and all):


For the past 10 years or so, startups have had two defining characteristics. First, they cost almost nothing to start. The intersection of good open-source software, infrastructure as a service, inexpensive distribution, and some plug-and-play monetization options like Google’s Adwords or an app store put all the power in the hands of small technical teams.

Second, when something worked—as rare as that was—the ability to scale and create extreme value very quickly was unprecedented…

But things are changing once again. Open-source software and infrastructure as a service will remain cheap forever. But the low-hanging fruit of highly scalable software startups has mostly been eaten. The winners in areas like media, messaging, advertising, and games have been established.

The next big opportunities seem to be shaping up around things like self-driving cars, on-demand services, VR, bots, bio, drones. But such opportunities lack turnkey generic infrastructure which enables development costs to drop close to zero. They are all expensive games in which to participate.

And Michael Eisenberg, founder of Israel-based Aleph takes it one step further:

Much of the next level of innovation will play out in the physical world, the financial markets and the world of experiences. Those innovative companies will take more money up front and they will become ever larger parts of VC portfolios.

I actually think Eisenberg gets even closer to nailing the emerging reality than does Lessin. For example, on demand services are expensive to scale, but cheap to launch. Bots are inexpensive to code. A lot of drone infrastructure is commoditized and modular at this point. Etc. But there are certain assets that are immutably capital intensive: real estate, logistics, banking, just to name three.

Let’s take the future of transportation for example (old timers might call this “logistics”). Boom is building the next generation of supersonic jets in a hanger outside of Denver. An actual real life jetplane. That is inherently capital intensive. The TSA/airport disrupters, as related in the WSJ’s All You Can Fly Experience, either purchase planes outright, lease planes, or pre-book large amounts of latent inventory – all demanding large upfront capital commitments. Turns out, putting a physical plane into the air is pretty darn expensive. Other next gen transport cos such as the Hyperloop raised $11M out of the gate to prove a prototype and another $80M soon after with a working test run. Good luck as a MicroVC.

Or, let’s take the reimagination of living. Whether its co-living operators such as WeLive and Common or subscription based flexible living models such as Roam, these companies also demand large upfront leasing obligations and build out costs. The paradox for seed investors is that a traditional seed round of a couple million dollars really only provides for a single (or at most, a couple) leases – all of which remain below full utilization for many months, constraining growth and economics. Want to bet on tiny homes or e-commerce driven pre-fab modular homes? That comes at a price too: Bluhomes raised $11M to launch its prototypes, and tiny home networks such as Kasita and Montainer also demand heavy prototyping and manufacturing costs to scale up their networks. ***

Even the millennial-first ecommerce brands (Bonobos, Trunk Club, Warby Parker, Baublebar) have, over the past several years, shown an increasing disposition towards brick and mortar exposure to combat clogged online acquisition channels. Not to mention a significant push from investors into private label inventory in order to boost otherwise anemic contribution margins across the sector. Stores. Inventory. All material overhead costs. And early in a company’s lifecycle too.

If, as Eisenberg suggests the “next level of innovation will play out in the physical world,” there is likely to be a bifurcation of the seed stage funding stack. Strategic angels still have a place as founders may need $500k-1M to show an MVP, prove one element of the technology, and surround themselves with credible advisors. But the emerging world of capital intensive or asset heavy companies puts a material strain on the $25-50M MicroVC funds who are most comfortable writing $250-750k checks into $1-2M funding rounds.

Sure, these rounds may still occur, but they won’t be as a frequent. Future funding requirements for those businesses – often in breakneck succession – might well crush the model. And, at a minimum, only the MicroVCs with the absolute strongest upstream brands will survive. These VCs will have two options: ship checks into far larger rounds ($7-15M) with a similar risk profile or target more incremental, derivative companies with lesser upside and more balanced financing needs. The shift is occurring “at exactly the wrong time” – right when hundreds of seed funds have recently been raised.

Of course, of course there are still amazing, cash efficient, asset-light opportunities across consumer sectors. But, anecdotally, they are harder to discern and appear less frequent. Irrespective, these more physical/experiential companies represent bold bets on the future – and investors had better come to terms with it. For better or worse, the price of startups has just gone up.

* Tomasz Tonguz at Redpoint released “The Decline of New SaaS Company Formation” this morning with data suggesting that new SaaS companies are being started with reduced frequency in 2015 & 2016. It’s likely the same “low hanging fruit” argument could be applied to traditional SaaS (non AI/ML, chatbots, etc) as well, as Tomasz notes: “The key systems of record in SaaS are already in place. Salesforce, Netsuite, Marketo/Eloqua/Pardot/Hubspot, Zendesk. Subverting those incumbents is going to require a meaningfully better product or substantially more effective customer acquisition channel.”  

** Grubhub and Opentable are two quintessential examples of this transition. Whereas in an offline world, one might pick up a phone and call a restaurant to book a reservation or order food delivery, Opentable enabled that process to be fulfilled via frictionless point and click. Many of these digitization transitions, including our investment in Spothero, have the additional benefit (and value creation) of opening up highly opaque inventory, thereby increasing frequency. 

*** One could make the case that startups can offset these overhead costs via pre-order/pre-sale/Kickstarter revenue.

How to Not Fail When Predicting the Future

“In the strict formulation of the causal law – if we know the present, we can calculate the future – it not the conclusion that is wrong, but the premise.” – An overview of the “Heisenberg Uncertainty Principle” from Uncertainty: The Life and Science of Werner Heisenberg

It’s common parlance to suggest that venture capitalists are in the business of “predicting the future.” Investments are referred to “bets,” and there’s a material uncertainty about the future. Case in point – a VC primer from The New Yorker’s 2015 profile of Marc Andresseen, “Tomorrow’s Advance Man”:

In venture, it’s not batting average that matters; it’s slugging average. Boldness is all. When Google Glass appeared, a16z joined a collective to seek out investments, and Andreessen declared that, without the face shield, “people are going to find they feel, basically, naked and lonely.” Google withdrew the product in January. But, he would argue, so what? His thesis is that such a16z failures as Fab and Rockmelt and Digg and Kno are not merely a tolerable by-product of the risk algorithm but a vital indicator of it. It’s fine to have a lousy record of predicting the future, most of the time, as long as when you’re right you’re really right. Between 2004 and 2013, a mere 0.4 per cent of all venture investments returned at least 50x. The real mistakes aren’t the errors of commission, the companies that crash—all you can lose is your investment—but those of omission. There were good reasons that a16z passed on buying twelve per cent of Uber in 2011, including a deadline of just hours to make a decision. But the firm missed a profit, on paper, of more than three billion dollars.

At Chicago Ventures, we are often discussing and brainstorming which factors or precedents we can identify that are predictive of future success. But over the years I’ve come to believe that venture investing isn’t the act of betting on the future; on the contrary, investments are a test of how well we truly understand the present.


Over the weekend I was re-reading William Duggan’s Strategic Intuition, a book I first read in business school, and which goes to lengths to demystify the unplanned creative sparks that underpin legendary strategic insights. Early on, Duggan takes a deep dive into understanding the scientific method, using Thomas Kuhn’s The Structure of Scientific Revolutions as a guide; and while I expected to mostly skim that section, it set off lightbulbs for me as regards our challenge as investors:

Kuhn goes on to explain that a breakthrough is part of both the past it came from and the future it starts, in the same way that a bend in the road serves as the end of one direction and the beginning of another…The common idea of how a lap of progress happens is a leap of imagination. Kuhn gives us an alternative to imagination that we can apply to realms of achievement other than science. He shows us in detail how the bend in the road happens: a selective combination of elements from the past makes something new. The elements themselves are not new.

This sequence for paradigm shift – achievement, than theory – is exactly backward from common ideas on how progress happens. In the typical notion of the scientific method, first you posit a theory, and then you conduct an experiment to test it. If your experiment works, you have an achievement. But this is the experimental method, not the scientific method. The experimental method is part of the scientific method, but it’s not the first step.

Scientific advance does not come about by a leap of thought to a new theory, but rather from combining specific achievements that lead to a theory, which explains them. It’s an act of combination, not imagination. Specifically, it’s the selective recombination of previous elements into a new whole. Pieces of the past come together to make a new future.

I think as entrepreneurs and investors consider the future, they should be singularly focused on the question: what has actually been achieved?

For example, Webvan – one of the public failures of the dot com era – was predicated on the assumption that the web (html/desktop) was a sufficient enough technological achievement to enable on-demand last mile delivery. It turned out that while the internet did facilitate real-time transfer of information, the opex of managing trucks and warehouses was prohibitively expensive (communication was not yet ubiquitous enough to enable crowdsourcing). The working assumption for this decade’s last mile players (Deliv, Postmates, Doordash) is that the GPS-empowered smartphone is a strong enough achievement to enable cost-effective crowdsourced delivery.

This take on experimentation versus achievement touches on broader themes I’ve been struggling with such as bitcoin and virtual reality. Is Facebook’s acquisition of Oculus an achievement for VR or merely an incremental step in its experimentation period? I am of the belief that VR is still deeply experimental and that technological progress, especially for consumer facing technologies, are rarely “achievements” until their presence has become ubiquitous through a native or modular platform layer (web, and more recently mobile are the two ultimate examples of this).

Bitcoin, on the other hand, may actually be somewhat silently transitioning into its achievement phase. Were historical increases (or recent surges) in bitcoin’s market price “achievements?” No, of course not – bitcoin is a commodity – the price is utterly irrelevant. But I’ve begun encountering increasing number of companies that have integrated the blockchain for accounting, certification, escrow, location or identification, often without advertising themselves as “blockchain companies,” but simply because blockchain tech enables them to optimize a process they couldn’t without it. While price is largely irrelevant, broad adoption of platform is not.

As entrepreneurs and investors, it’s a heck of a lot of fun to imagine the future. But bets on the future need to be predicated on a discerned analysis of the present. This approach was a new, nuanced insight for me into the question of “why now?” and I hope its helpful to readers as well.


The Birth of the Web 3.0 Rollup

Many of the first generation of large internet properties – often termed Web 1.0 (in my mind roughly 1998-2007) – were holding companies employing roll-up strategies, acquiring and aggregating multiple brands. Everything from IAC’s (NSDQ: IAC) entry into digital media, to GSI Commerce, Liberty Media Group, even XOXO Group (NYSE: XOXO). The purpose, in many cases, was to become a category killer – not simply operating a single product, or targeting a specific demographic – but owning an entire category: fitness, makeup, weddings, holidays, etc.

The years that followed brought a host of unbundling. Entrepreneurs recognized the underleveraged value that could be unlocked by catering to specific groups of customers within platforms such as Craigslist or eBay. This trend brought us Airbnb and Etsy but also lots of single purpose or vertical/niche specific apps that failed to reach venture scale. I wrote about this recently in The Market Has Spoken: Go Horizontal not Vertical. But it was unclear to me at the time what those horizontal extensions might look like.

The Return of the Rollup

Earlier this week, Victorious, a publisher of more than 100 apps targeted to micro-communities of superfans, announced a $25M Series B round of funding. Six months prior, Massdrop, a platform that hosts dozens of transactional micro-communities around certain hobbies and products raised a $40M Series A from August Capital. Around the same time, Amino Apps, a creator of a wide range of interest specific forums raised its Series A from USV and Venrock. ReplyYes, a fast growing SMS-based curator for niche products recently expanded from targeting just vinyl enthusiasts to comic books fans, with more to come.

While this studio model of launches is commonplace in the gaming world, it’s a new development in the social category, and a reversion to an older model in commerce. It also appears to be a compromise of sorts (not a rejection) around the emergence of single purpose apps or vertical specific commerce – an affirmation that they do provide a benefit to end users, but an admission that on an individual basis their reward did not justify venture investment.

Around 2.5 years ago, I met a successful entrepreneur who had built a $200M revenue “category killer” business in the mid 2000s and believed he could build the same in what was otherwise a terribly noisy category. I was skeptical about the market but passionate about the entrepreneur and had him spend time with the entirety of our team. In selling the deal to my partnership, I focused on selling his operational excellence and impressive cash efficiency in spite of the headwinds in the space. But we passed, partially because of the market, but also partially because of the multiple business lines he was growing simultaneously – all interconnected, but each with unique marketing and operational challenges – which suggested to us a lack of clarity and focus.

That company is now doing extremely well, and our reason for missing the investment is because I failed to comprehend (and subsequently sell) his founding insight: that his niche specific competitors would quickly stagnate on growth and face increasing acquisition costs. That a category killer or roll-up strategy would likely win: enabling him to keep growth consistent & cash efficient and cross-market different brands to his existing, tangentially related customers.

I won’t make that mistake again.

Now that I’ve seen the roll-up/studio/category killer model begin to play out again in certain social and commerce categories, I’m still left wondering if or how it plays out in b2b SaaS, especially given the constraints around scaling up sales teams. I’ll leave that as an open question for the future.

The Economics Underlying Chatbot Mania

Over the last several weeks, we’ve reached peak AI/Bot mania. Most of the conversation has centered around Chatbots and the potential emergence of a new platform/distribution layer. If you’ve been mostly ignoring the press, some good reads are:

Tl;dr – the major takeaways are as follows: given that consumers don’t really download apps anymore, brands & retailers have a new access point to end consumers, sitting on top of existing messaging platforms and leveraging chatbots to ensure mass scale. The truth is that the chatbot platform conversation is really just an extension of the one we had about a year ago during the emergence of Magic/Operator and SMS as the new platform, which we discussed in Are We Already Rebundling Mobile. An important extension given that such bots have been democratized and can now be spun up not just by tech companies, but by traditional retailers (on their own or within Messenger) or even by individuals such as you and me.


All @TayTweets joking aside, the more intriguing aspect this time around is that Artifical Intelligence and Machine Learning have improved by leaps and bounds. A few articles that reflect this point are: Why AlphaGo Is Really Such a Big Deal, The Current State of Machine Intelligence, Can Machine Learning Predict a Hit or Miss on Estimated Earnings, and The Humans Hiding Behind the Chatbots

But again, the vast vast majority of analysis has focused on bots living within the worlds we frequent (messaging & SMS) and the platform implications. SO – I wanted to spend a few paragraphs to quantify and explore the effects of these advancements from a unit economic or business perspective. The big ones are twofold in my mind: (1) Properly executed AI can transform certain human capital marketplaces from operating as take-rate businesses and transition them into high gross margin software businesses. (2) Chatbots in their current function as customer service agents can make a material impact on contribution margin & overall EBITDA if they can successfully remove the customer service expense line.

20% Take Rate —-> 90% Gross Margins?!?!?

First, some context. About four months ago, writing in “The Middleman Strikes Back” I suggested:

“It’s clear why a hybrid AI/machine learning model is the holy grail for several verticals – replicating [a high] level of personalized service while minimizing overhead labor costs and maintaining extraordinary software level gross margins.”

But in fact, I was wrong. Dennis Mortenson, founder of in “The Humans Behind the Chatbots” believes that the hybrid in “hybrid AI /ML” should be minimized even further…to zero:

“The two scheduling e-mail bot companies have divergent plans for expansion. Clara, which is slowly letting people off its waitlist and said it currently serves hundreds of companies, charges $199 per month per user., on the other hand, plans to move from limited beta to a public release later this year and wants to charge about $9 per month. Dennis Mortensen, its founder, wrote in an e-mail that “only a machine-powered agent can take on the 10 billion formal meetings that U.S. knowledge workers schedule every year.” Mortensen said the service will start asking e-mail senders to clarify when the computer can’t interpret an message—“Did you mean Monday, April 4?”—instead of having an employee read it and infer. “We want to give the job away for free, or for $9, which you can only do if it’s software,” he said.”

Executing on Dennis’ vision, by removing human labor in the middle, you have effectively transformed a take-rate marketplace into a high margin software business, while managing to provide a similar product. Here are examples of some of the service/agent businesses* that could see their economics transform towards 80-90% gross margins when fully leveraging AI:

The Unit Economics Underpinning the AI

Additionally, the company’s variable costs (ie the cost to provide each incremental production or engagement) will transition from significant (paying humans) to immaterial (software cloud hosting fees) enabling a much lower cost of servicing demand, should these companies choose to lower prices. Doing so could expand the audience for such products tremendously.

While I was initially a big skeptic of, the combination of a product that increasingly works with minimal human intervention and a product priced to undercut the market tremendously at $9/mo is a proof that this transition from human marketplace to pure software is already underway.

Further, there’s even an even bigger opportunity in play. If Matt Turck’s suggested “data network effects” take hold many of these service marketplaces – few of which are operating in actual winner take all markets – could be replaced by a software layer, leveraging data network effects so strong that those markets actually become winner take all.

$$$ Massive Value Creation in Public Markets

Again, this is mostly a theoretical exercise, but let’s imagine for a moment the value creation consequences of fully automated chatbots successfully managing 100% of a company’s customer service interactions.

Here are a handful of examples of some companies you might be familiar with showing their current enterprise value, approximately how much they spend on human capital customer service and the effective sensitivity in their valuation if they could maintain their current level of customer service via no-cost** chatbots:

The Unit Economics Underpinning the AI2

It’s not clear to me this is the optimal way to assess the economic value of chatbots – I kind of doubt anyone knows yet – but what is clear is that the effect to profitability would be tremendous. Massive. And that’s just by automating customer support alone.

Importantly, and intentionally, this exercise ignores the real, global cost of losing so many service level jobs from the economy. I’m not qualified to assess that cost nor its effect. There are many opinions in the market for what an AI economy might look like; Roy Bahat at Bloomberg Beta recently offered an intriguing one to The Twenty Minute VC – a futuristic marketplace actually placing a premium on any human manufactured products or human assisted services because of its rarity. There are others as well. Much discussion around AI is conjecture, but its economic effects are a very serious business.

* Certain of these companies do not publicize take-rate or offer non take-rate subscription models. In these cases, gross margins have been assessed from public interviews, or estimated based on reported hourly worker wages and expected throughput.

** This assumption is implausible as there would still be some associated costs related to the software itself and managing/customizing the software on a daily basis.

Special thanks to Dan Abelon for his feedback on this piece.

Ideas Matter

The debate amongst venture capitalists over whether to prioritize markets or people in investment decision making is as old as the industry itself.

Fred Wilson, famously authored an oft quoted 2004 blog – Execution Matters, Ideas Don’t – which referenced USV’s failed incubation of FaveMail leading to the following conclusion:

“The lesson i take away from the whole thing is great ideas don’t make great investments – great entrepreneurs do.”

Three years later, Marc Andreessen, co-founder of Andreessen-Horowitz, opined differently in “On product/market fit for startups,” noting:

In a great market — a market with lots of real potential customers — the market pulls product out of the startup.

The market needs to be fulfilled and the market will be fulfilled, by the first viable product that comes along.

The product doesn’t need to be great; it just has to basically work. And, the market doesn’t care how good the team is, as long as the team can produce that viable product.

Most recently, University of Chicago Professor Steven Kaplan in his 2009 study “Should Investors Bet On the Jockey or the Horse” in the Journal of Finance concluded:

The results for both of our samples indicate that firms that go public rarely change or make a huge leap from their initial business idea or line of business. This suggests that it is extremely important that a VC picks a good business. At the same time, firms commonly replace their initial managers with new ones and see their founders depart, yet still are able to go public, suggesting that VCs are regularly able to find management replacements or improvements for good businesses.

In spite of Andreessen’s comments, I would say that the working consensus in the early/seed stage venture world for the duration of my tenure has been to bet on special people and let the rest fall into place. That approach was verbalized in my interview with David Hornik on this blog where he outlined his investment approach as “people, people, markets, people.”

But the world is changing. The vast majority of companies I’m referred to are largely derivative ideas of larger tech/startup competitors. Problematically, this is actually fundamentally different than going after large monolithic incumbent corporates in a given space. For whatever reason I think it’s because entrepreneurs are trying to improve processes/better execute on problems rather than re-imagining the reasons those habits/processes even exist.

My sense is that the market is cycling back to the importance of bold, unique, creative ideas – above and beyond the obvious focus on “big markets” or “founder/market fit.” Of course people still matter, just as a large addressable markets have always mattered, but in my estimation there is a subtle yet undeniable shift of interest away from improved processes/products towards ideas that challenge the fundamental assumptions underlying the existence of those processes or products themselves.

In that vein, I was recently asked to prep some talking points for a firm offsite on the state and challenges of investing in consumer tech. I’ve attached those slides here. The theme was very clearly that we are in the “reimagination” stage of venture and my observation is that companies solving pain points but not re-imagining or re-conceptualizing consumer behavior are out of favor with top tier investors.

[Note: It’s always nerve wracking putting detailed thoughts into the market – I’m sure a lot of people disagree with me – but I do appreciate any feedback and counter-examples of places where I’m wrong.]

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...