Posts from April 2020

I got a lot of feedback on my post yesterday and a bunch of it was around the issue of privacy and whether we really care to keep our personal data private.

I think we are increasingly aware of our data and the need to have more control over it and how it is used.

When people ask me about this, I like to show them this chart:

That is the daily search traffic on our portfolio company DuckDuckGo’s private search engine.

The 65mm searches a day on DuckDuckGo is about 1% of Google’s estimated 7bn searches a day.

But it is also the case that search activity on DuckDuckGo is growing faster than Google’s search activity. If that continues to be the case, then that 1% can grow to something much more than that over the next decade.

People do care about privacy, but the sacrifices we make for privacy must come at a low enough cost that we will make them. As DuckDuckGo has improved its product, more people have used it. The combination of increasing awareness of the issue of data privacy combined with better user experiences for privacy-focused competitors will drive us all to an online experience where privacy has a lot more value to everyone.

Christina Farr at CNBC has a good post that details the back story of how Apple and Google came together to implement an interoperable system and a set of APIs and SDKs to allow third parties to build exposure alerting apps on their mobile operating systems.

Like many in tech, I have been interested in exposure alerting (which we used to call digital contact tracing) since this pandemic started spreading around the world. I have always believed that these little computers we carry around with us can help solve many challenging problems and this sure feels like one of them.

But I was concerned about the need for interoperability, privacy, and security. I told everyone who I talked to about this problem in February and March that I felt like Apple and Google had to implement something in their mobile operating systems for this to work. And I was deeply concerned that would not/could not happen.

But it did. And when I heard the news, I was so pleased. Difficult times can make for interesting bedfellows.

This tweet cited in Christina’s story is great:

Apple and Google’s APIs are coming on May 1st and I hope to install an exposure alerting app that runs on them on my phone as soon after that as I can. I hope all of you join me in doing that.

This moment we have been living through over the last two months has put pressure on many companies to figure out how to keep the lights on and stay in business. It has also been a second wind for some companies that have been struggling to survive in difficult sectors like delivery, food, and e-commerce.

At the intersection of those two things lies a truth that I have seen over the years. If you stick around long enough, you can often catch a lucky break. But that lucky break can’t come for you if you didn’t figure out how to stick around long enough.

Survival instincts are something that I have learned to appreciate in founders. There are other things that are important too, possibly more important, like the ability to get the right product to market, the ability to sell, recruit and raise, and the ability to inspire and lead. Those are all necessary for success.

But the survival instinct is related to luck, which is an underrated factor in success. You can’t be lucky unless you are at the right place at the right time. And you have to survive for that to happen.

I wrote a blog post in September of last year arguing that gross margins and operating margins really matter when valuing companies. I argued that “software companies with software margins” are better businesses than tech companies that are not really software companies but a tech-enabled version of some other business.

But gross margins, in particular, can be tricky to compare. In some cases, a software business is in the middle of the revenue flow, takes the revenue, and then passes on a lot of it, and is left with what looks like a low margin, but is in fact a high margin.

An example of that is the Dutch payment processing company Adyen. Here is a screenshot of a part of Adyen’s income statement from Yahoo Finance:

So Adyen operated in the last twelve months with an 18.7% gross margin. Many would think that was a “very low margin business.” But the truth is Adyen is simply passing through that $2.1bn of revenue to financial institutions in the form of interchange and other fees. They do very little with that money.

Let’s compare that with the big retailer Macy’s. Here is a screenshot of a part of Macy’s income statement from Yahoo Finance:

So Macy’s operated at a 40.1% gross margin over the last twelve months, more than double what Adyen operated at.

That $15bn cost of revenue on Macy’s Income Statement is the cost of purchasing everything you might find in a Macy’s store, the inventory costs associated with that, and the cost and effort of displaying all of that inventory in the stores.

So while it is the case that Macy’s has more than double the gross margin of Adyen, I believe Adyen has a much more attractive business from a margin perspective than Macy’s.

That is because Macy’s expends enormous amounts of working capital and operating expense and effort in its $15bn cost of revenue where Adyen expends very little working capital and operating expense and effort in its $2.1bn cost of revenue.

The trick, I think, is to wrap your head around the cost of revenue or cost of goods sold line item in the income statement and think about what is going on there. If it is very little to no effort, and largely just an accounting entry, then you may have a “low margin business” that is actually a high margin business. On the other hand, if it is a lot of work and capital investment to produce those margins, well then you have what you have and that is often a low margin business.

Yesterday, Gabrielle Hamilton, the chef/owner/creator of Prune, a tiny twenty-year-old restaurant in the east village that The Gotham Gal and I and our family have always loved, wrote a piece in the New York Times that really captures the sense of loss we all feel around our favorite local places, particularly for the people who made them what they are every day, day after day, year after year.

I can’t get her words and the images and emotions they convey out of my head.

So I want this Funding Friday to be for people like her, who have given a large part of who they are to serving us.

And the cause I’ve selected is ROAR, Relief Opportunities for All Restaurants. You can follow them here:

ROAR’s Twitter

ROAR’s Instagram

ROAR and Robin Hood, one of NYC’s leading charitable organizations, are partnering with the National Restaurant Association Educational Foundation (NRAEF) to provide direct cash assistance to restaurant workers in New York City facing unprecedented economic hardship as a result of the COVID-19 pandemic.

We made a donation today and if you’d like to join us, you can do that here.

I realize that most businesses are suffering greatly in this pandemic. Many have been shut completely.

But there are some that are experiencing the opposite situation. They have a growth spurt as a result of this moment. Businesses in food delivery, e-commerce, online education, telehealth, remote work, and cloud infrastructure are examples of such situations.

I’ve seen event driven growth spurts over the years. A plane lands in the Hudson and everyone heads to Twitter to see it. A competitor is shut down and everyone shows up on your door. Crypto gets hot and everyone wants in on the action. That sort of thing.

And I’ve been talking to leaders who are experiencing this and wondering how to model out what happens when and if things return to normal.

Each situation is different but a framework I like is to take your pre-event baseline, your event driven peak, and assume you will give up half of the delta when things return to normal and that will be your new baseline.

That won’t be right of course. It’s a model. You can revise as real data comes in.

But what it suggests is that not all of your new customers will stick around. But some will. And you will have a new and higher baseline. That has been true of almost every event driven growth spurt I have seen in my career.

I wrote about our portfolio company Duolingo’s English Proficiency Test back in August of last year. I have always loved the idea that a company that helps people learn a language can also help people prove their fluency in a language. It is two sides of the same coin.

But the road to success with the English Proficiency Test has been hard. The “incumbent provider” of English proficiency tests, Test Of English As A Foreign Language (aka TOEFL), has had all of the companies and universities who accept it locked up for many years. And if you are required to certify with TOEFL, well then you take TOEFL.

The Duolingo English Test is and has always been a way better product than TOEFL. But in some markets, incumbency matters more than better. One of the primary benefits of the Duolingo English Test is you take it at home on your computer versus having to go to a proctored location. It costs less ($49 vs $205). And the test takes one hour vs three hours. And yet, it has been hard to crack into this market.

And then the pandemic hit. No more in-person testing. As the international higher education publication PIE News puts it:

With the suspension of traditional English proficiency tests in countries most affected by the coronavirus, a wave of US institutions are now accepting the results of the Duolingo English Test, either as stand-alone proof or as a supplement to other measures of English-language proficiency.

https://thepienews.com/news/us-more-heis-accepting-the-duolingo-english-test-amid-uncertainty/

Duolingo’s co-founder and CEO, Luis von Ahn, told me this in an email yesterday:

1. The number of tests we administer per day has gone up 10x!

2. 500 new university programs have begun accepting the Duolingo English Test in the last 8 weeks (we had 1,000 before this). 

3. Both TOEFL and IELTS, after spending a lot of time saying that online tests were no good, now have online options.

So now the market is open to competition and the best product can win. I’m betting on Duolingo (and have been since we made our seed investment in the company in 2012).

I woke up to this tweet this morning:

I went to that blog post and clicked on the link and sure enough someone had swapped out my cap table template from 2011 with their own cap table. I am not entirely sure how that happened and for how long that has been the case, but I was not going to let that stand.

So I went to my google drive and searched and found the cap table that I had built for that post back in 2011, made a copy, made it public on the web but view only, and fixed the link.

Phew.

If you want to see a cap table and waterfall template in the style that I have become accustomed to over the years, here it is. Hopefully, nobody will hack it again. I will be super careful not to permission anyone to have edit capabilities (which is what I think I may have done accidentally).

I wrote a blog post on March 12th called Open For Business and thought I would return to the topic.

If you search for “vc open for business” on Twitter, you will see almost universal scorn for the idea that VCs are open for business right now.

But our experience doesn’t match that scorn. Since writing that post, we have watched a bunch of our portfolio companies close financings, some on the same terms as provided before the pandemic and some on slightly adjusted terms. We have mostly seen VC firms live up to the commitments they made pre-pandemic and in the cases where terms changed, it has not been not gratuitous.

On our side, we have signed three or four term sheets since I wrote that post, closed on a few of them already, are currently engaged in several processes.

We have had to step up to the plate in a few situations and provide interim financing, and we are certainly working very closely with our existing portfolio companies during this challenging time for many of them. But our focus has not moved dramatically away from looking at and investing in new companies and we don’t see a dramatic change in that regard among many of our peers in the venture community.

David Kelnar sent me a blog post he wrote addressing some of these trends this morning and I saw this chart in it.

I don’t think new investment activity has shrunk from 40% of the industry’s time to just above 20%. From what I see, it has shrunk a bit, but not cut in half.

There certainly are funds who are focused on sectors that have taken a bath in the pandemic and in those cases it is natural and appropriate for those funds to be more focused on their portfolios. But I don’t see that across the entire VC landscape.

It is always easy to talk yourself into the idea that the door is currently shut for you. But before you do that, I suggest you knock on it. It may in fact be open.