Pawel’s blog

I have spent some time playing with and reading about steemit. The vision of the media world it suggests we might be heading towards is fascinating.

In short, steemit is like reddit/medium on top of a blockchain with tokens integrated into the system to reward publishing and other activity on the platform. It is still a complicated system that I do not get entirely, but I think I get a sense of how the dynamics are. 

Steemit rewards content contributions, both by publishers and curators ('likes/votes'), with tokens. It also makes the tokens useful in other ways. Holders of tokens can pass additional rewards to the authors they like, but they also get to enjoy more influence on the platform as well as benefit from the platform's success by getting a share of the tokens getting mined.

The tokens can be traded on exchanges against bitcoin, i.e. they represent real value, not just an internal currency of the platform. The supply of tokens is pre-defined for the future, which makes the users not worry about token inflation.

Creating a system that gets this mini-economy to balance is likely a tough job for the platform's operators. I think at some point we will see best practices for such communities emerge, but now it still seems like a learning process and that's why the system is somewhat clunky and hard to understand. Yet, most importantly, it seems to be working and growing and some content contributors make real money of it.

Below is a chart of the traffic of steamit taken from similarweb. The dynamics is clearly positive, and with 6m visits in May it is at a respectable size. 

It is of course still small by the standards of reddit (1bn+ visits in May) or medium (close to 100m). It is also possible that the increase in interest for steemit and other token-based services in May was driven by the strong crazy price increases in crypto in that month and that it will subside in the coming months. The price of steem (steemit's token) went up by ca. 20x recently, which by no means was an exception in the market. In any case, the activity on the platform is signifcant.

What I find most fascinating about steemit is that it suggests a direction for development of the media landscape in the future. Imagine a world where journalists are truly independent. Where they can create and publish their work, get audience and make a living, without the need to be part of a larger publishing platform or a magazine.

One can take it a step further. Imagine similar rules applied to music. Something like a distributed soundcloud. Where artists can publish and make a living of it. Moreover, the tokens being tradeable, how about a music band issuing a pre-sale of tokens entitling the users to listening to their not-yet-existing album once it has been released? I think many bands would prefer this than dealing with music labels. Or independent film makers pre-selling (tradeable) tokens enabling one to view their films in advance of a public release once they have been made? Amazing!

I am still learning about steem and other platforms working on applying the blockchain paradigm to the web as we know it and reinventing it. It is still early, but it feels like strong new concepts are emerging. If you want to help me find more practical ways how blockchain technologies can impact the world, or want to add something to or comment my logic above, please do so, I will be thankful for additional insights.

P.S. Fred Wilson wrote about steem before and it is an interesting read.

Contract clauses and financial models illustrating how proceeds are distributed among different shareholders at exit are frequently referred to as waterfalls. They can be very complicated.

It seems that the VC funding market cooled off a little after reaching a peak in 2015. In colder markets, investors have a stronger tendency to get creative on terms, especially if not much competitive pressure is felt around a deal. One of the the things investors like getting creative about is liquidation preferences. In this post I will outline why I think you should try to push back hard on complicated and aggressive liquidation preferences during your financing round.

A lot has been written about liquidation preferences and you can check out this post by Brad Feld for a good intro. In short, the liquidation preference language describes how the proceeds are distributed between shareholders at exit. It describes what the holders of individual preferred share classes are entitled to receive before the rest gets served.

It typically has two components, (1) a minimum multiple or interest relating to the original investment amount and (2) a participation or lack thereof which describes whether the preference represents a minimum return threshold after which everyone gets served pro-rata or whether it comes ‘on top’.

A simple (1x) non-participating liquidation preference ensures that, at an exit event, investors get the money they invested into a company returned to them first, before proceeds are distributed to everyone else according to ownership %-tages. I do believe that a simple 1x non-participating liquidation preference is absolutely the right thing in early stage VC deals. Fred Wilson outlines nicely why in this post. Anything beyond that for the investors, however, is tricky, not always fair and full with potential for unintended consequences.

We believe, and our experience and industry statistics prove it, that in early stage venture investments money is not made with contracts. You typically either make a lot with an investment, lose all or most of it, or make very little. In a bigger picture, even if liquidation preferences at exit come into effect in a way that benefits the investor, this does not increase his or her overall returns materially.

Yet, many investors will ask for liquidation preferences, because, in theory, a liquidation preference sounds great. After all, it reads like a guarantee to lock in returns. An investor might think: let's ask for a participating liquidation preference and I’ll make quite some return even if the valuation does not increase. Or, even better, let’s ask for a participating liquidation preference with an interest. How about 8% per year? After all, this is my hurdle rate (minimum fund return below which no profit participation is paid to fund managers), so I need to make at least that much in every deal.

Sometimes entrepreneurs may be happy to agree to a participating, growing or multiple liquidation preferences, typically in exchange for a higher headline valuation. Here are five reasons why I think that such structures are a bad thing, especially in early stage VC deals (Seed, A, B):

• By design, every liquidation preference skews the distribution of exit proceeds away from ownership %-tages. I think this works OK in case of 1x non-participating liquidation preferences, which are irrelevant if the exit price is above the entry price of a negotiated funding round. Multiple and / or participating liquidation preferences can skew distributions much more heavily, even at higher outcomes. This leads to sometimes very different financial incentives for investors vs founders or ESOP holders and in consequence is likely to result in conflicts. An example could be a discussion around a  proposed sale of a company at a price that will allow investors to make a nice return, but due to the liquidation preference structure not much would be left on the table for common shareholders. Since we believe that the VC game is all about aligning interests, everything that misaligns them should be avoided.

• Early deal structures frequently create a precedent for later stage financings. Later stage investors often ask for (at least) the terms that the early investors got...and add some things on top. So if you create a multiple / growing liquidation preference as part of your early financing you can expect that all future money will come-in at these terms, or worse. Thus, if considered long term, aggressive terms are not only disadvantageous to the founders, but can also be disadvantageous to the very early stage investors that ask for them in the first place.

• The terms around a liquidation preference are frequently a heavily negotiated topic. Entrepreneurs may struggle with understanding the idea when confronted with it for the first time. It just makes negotiations and contracts longer, more complicated and more emotional.

• Modelling liquidation preference waterfalls :-) in the contracts and later in Excel can become extremely complex, time consuming and very prone to misunderstandings.

• If too much money goes into a company with aggressive liquidation preferences attached to them, it can become very demotivating to the founders and team members owning stock, thus hurting the company overall or triggering a discussion around increasing or restructuring the ESOP, e.g. through carving it out from the liquidation preference stack (have fun modeling that!).

The above are just some of the issues that come to mind when you think about the impact of aggressive liquidation preferences. In the past we made a few investments where we got participating liquidation preferences ourselves, but because of all these issues we stopped asking for them a few years ago.

If you ended up with an aggressive liquidation preference stack at your company, you might still be able to negotiate it away in your next financing round, if your position is strong enough. Or if you cannot get rid of it, you might manage to introduce a cap or minimum return threshold after which it disappears. We have seen and supported such restructurings in the past - they can sometimes be achieved, frequently to the benefit of everyone involved.

I would recommend to any venture funded entrepreneur to try to keep the liquidation preference stack as simple (ideally 1x non-participating) as long as possible. Even if it should be at the cost of a lower valuation. It will pay off in the long term.

If you are not sure you understood the concept of a liquidation preference well enough you might wanna check out the cheatsheet below prepared by Clement. He reviewed a version of this post and thought a cheatsheet will do a better job at explaining the basic concept of the liquidation preference than my text :-) Thanks Clement, and Christoph, for reviewing this post.

I recently met an entrepreneur for dinner. He is currently working on his fourth company and thus has a lot of experience in various areas of entrepreneurship and (business) life in general. Among many things, we discussed fundraising for startups. He mentioned to me a metaphor that I would now like to share with you.

He said: startup fundraising is similar to the job of an imaginary train conductor stopping his train by a train station and having the task to depart without delay, but also to take a required minimum number of passengers with him. The hard part is that he cannot depart without passengers who, as is often the case, might be running late or stand on the wrong platform.

Now, imagine a train station where there are a few guys (potential passengers) hanging around and smoking cigarettes. These guys are VCs. They want to get on a train, but are unsure which one brings them toward their destination (a good exit). In fact, they do not care so much about the type of the train or where it comes from as long as they know that the destination is reachable on time. The conductor’s task is to make them want to get on his train quickly, so that he can depart on time.

The job is challenging, because even if the guys smoking cigarettes like your train, they will rarely jump on it quickly, if they believe it will not depart without them or is not departing immediately. They will hang around and smoke a few more cigarettes…until the train starts moving.

So if the conductor wants them to take his train, he has to convince them that the train is about to depart or is already slowly moving. The problem is that you cannot start moving if you don’t know how many, if any, passengers will jump on it. The train cannot depart empty.

The way to solve this paradox is to maximise the number of folks at the train station with whom the conductor chats about the upcoming journey. In that way he can quickly figure out who is convinced that the train will most likely take him to his destination and is therefore most likely to take it. Once that’s done, and some of them start making first steps toward the train, the conductor needs to quickly blow his whistle and shout (very loudly!) that the train is about to leave the station or even slowly start moving the train (this is known as "we are now closing the round"). In most circumstances, those that started walking towards the train will start running, and those still pondering upon their decision will swiftly throw away their cigarettes and try to jump on the train as well (this is known as fomo :-)).

Two months ago Christoph attempted to answer the question on what it takes to raise money in SaaS in 2016. The result was a table outlining ballpark indicators across startup development stages that need to be met to get to the next financing stage in SaaS, i.e. from Seed to A, from A to B, etc.

It seems the SaaS fundraising napkin, as Christoph called it, was very helpful to SaaS founders out there. The feedback from the community was very good and many interesting discussions evolved.

Given that marketplaces are also a strong focus area for us, we immediately thought that it would be great to have a fundraising napkin for marketplaces too. After thinking about it for a bit, we came to the conclusion that this might be a bit more tricky, due to marketplaces being a less homogenous group than SaaS startups. Here are a number of areas in which the differences are visible:

• Geography - SaaS startups tend to be global from day one, frequently go after the US market and raise funding from the same group of international SaaS investors. Marketplaces, on the other hand, are frequently multi-local, scale country by country - and have to raise funds, especially in the early stages, from local investors - who might vary in standards.

• B2B vs B2C - when we talk about SaaS we typically talk about startups selling software to businesses. In contrast, marketplaces can be B2B, B2C, B2B2C..

• Business model mechanics - marketplaces can have varying take rates, some own some do not own the transactions that happen on them, some are platforms for trading products versus others are about services and may or may not involve a SaaS component.

Last week I attended a session on M&A in tech organised by Corum and Greenberg Traurig in Berlin. I never worked with any of the firms, but thought the way they approached the topic was interesting and decided to attend. I think it was a good session - many thanks for the invite!

I do think that M&A can be a viable tool for startups. At the same time, M&A is very tricky, the way the market works is intransparent and the process is a big puzzle, especially for those with little experience. That is why I think that all good education materials about it are helpful to the early stage tech ecosystem.

Below I attach the slides that the Corum folks used during the session. It is long, some 100 pages, but there is a lot of good stuff in it. The presentation starts with some marketing stuff, but the market update starting at slide 26 is interesting. And then, at slide 61, starts the section on the M&A process. Some of the material might not be self explanatory without the story telling, but I think it is still a pretty good read.

I also attach a template for a due dilligence data request list. I have seen such things used in practice and it is not unusual for them to be very long and detailed. If you want to get a flavor of what awaits you when you go into M&A, have a look at it too.

The most interesting stat of the presentation for me can be found on page 28 - average age of an acquired company: 14-15 years. While being aware of how this stat can be distorted, it is still a good indication that building companies is a long term play!

I asked the Corum guys whether the material can be shared, and they agreed.

Real estate is a massive market. PWC estimates the global stock of institutional real estate at tens of trillions USD and growing. Add to that the lower quality stock and the market size probably is well over 100 trillions. Thus, real estate certainly is one of the biggest asset classes out there, if not the biggest one. As software is eating the world, it is also impacting, in many ways, the world of real estate. And if an industry that is trillions dollars heavy gets impacted by the Internet, things ought to be interesting.

Tech in real estate is not exactly a new thing. Bringing real estate listings from newspapers to online is a process that started in late 90’s. The software that is meant to make the work of real estate agents and managers more efficient has been around for some years now and is continually evolving. These general trends have been receiving large amounts of attention and investment.

What we have been witnessing in the real estate space so far was mostly about tech and Internet improving the efficiency of old established processes, or as CB Insights put it, reshaping how real estate is bought, sold and managed. In our portfolio, a good example of a startup addressing this trend is Propertybase, a cloud based CRM solution for the real estate industry, in which Christoph made an investment as a business angel. 

But there are ways in which the Internet impacts the real estate market far beyond pure efficiency improvements. I find the areas in which Internet solutions change the patterns of how real estate is being used most interesting. I think that they have the potential to fundamentally change parts of the real estate market and thus open up exciting opportunities for value creation for startups. Below, I will dive into a few examples.

So far, the most pronounced, although indirect, impact of the Internet on the ‘usage patterns’ in the real estate market was imposed by the rise of e-commerce. As sales of goods and services shift to online, the demand for offline space to sell these goods goes down. The structure of the remaining demand changes. Large discount outlets located just outside of cities suffer and prices for such property go down. The remaining demand tends to focus on the more accessible locations and new formats for commercial real estate usage are popping up.

The impact of e-commerce on commercial real estate seems so pronounced that serious financial institutions run studies about this phenomenon (here is one by Aviva).

Probably the most prominent direct example of how the Internet changed how we use real estate is Airbnb, the well known company that enables everyone with a spare room to compete with hotels. It represents, already now, the biggest ‘hotel chain’ in the world, without owning any real estate. And it is truly global and democratic, as it offers something for ‘everyone everywhere’, from the cheap rooms way into the luxury category and business travel. It has totally changed the way we think about room rentals.

Another one are co-working spaces, like wework or the Berlin based Factory, which enable companies to have a spot in a fully equipped office, available right away, without the long-term commitments needed for a typical lease. Co-working spaces use the Internet to acquire customers when spots become available and to use their spaces more efficiently. The co-working space category is growing very quickly - driven by the trend towards more flexible work, but also by the availability and maturing of technologies needed to put together such offerings.

Interestingly, data shows a significant decline in the average length of lease for commercial real estate across all categories. It is hard to say whether the Internet has much to do with this, but I would think so.

No wonder that startups jump on this trend for short(er) term commercial real estate demand and try to fill it. Startups like Deskbookers (a Point Nine portfolio company), PopupImmo or The Store Front are all very good examples.

If one tries to categorize the above into trends, it seems that there are four main ones:

• Buying and selling real estate has shifted a lot to online platforms and will continue to do so. This makes the market more liquid and efficient.

• Utilisation - technology helps make the usage and management of buildings easier and more efficient - and thus cheaper.

• We now do online more and more of the things that we were doing offline before. Be it discount shopping or dealing with a bank. This reduces the demand for commercial real estate needed for these activities and changes the demand structure for real estate in the respective areas.

• Booking (as in the case of Airbnb or booking.com) - the Internet enables real estate inventory to be offered to users in real time. This allows for better utilization of capacity and yield management, and thus, counter-intuitively, might drive more (and better) capacity into the market, as real estate projects that might not have been profitable before can be made profitable now.

We already made some investments in these areas and here are some recent examples from our portfolio.

• Booking:

• Deskbookers enables flexible booking of work and meeting spaces.

• Eversports lets you book sports venues.

• Utilisation:

• HappyCo digitizes inspection processes which traditionally have been paper based - real estate companies love them.

• KISI lets companies control their office doors with smartphones.

I believe that the trends outlined above will result in changes of real estate prices and definitely are something for people buying and selling real estate to watch. More importantly for me, they will continue to offer tremendous opportunities for startups. We will keep on watching this space closely and hope to make more investments in real estate related startups, preferably in SaaS and marketplace businesses trying to address these big shifts that are happening.

I am sure the above list of four trends is not perfect and the trends might be overlapping or I might be missing something - please feel free to comment or add to it.

Many thanks to the Point Nine team for reviewing the early versions of this post.

The start of a new year encourages us to look back at what happened and think about what is coming. Having come across this post by Fred Wilson of USV I thought a similar analysis of the development of Point Nine’s portfolio could be an interesting exercise to provide some long term perspective and maybe deliver some interesting insights.

A big part of our investment thesis at Point Nine is to invest in SaaS and marketplaces startups. The chart below illustrates this showing the development of our portfolio, now across three funds, since 2010. It shows, for every year, the number of active companies in our portfolio, split in three categories: SaaS, marketplaces and other. Admittedly, in some cases it was hard to clearly categorise the companies. I tried nevertheless and the result is that currently we have investments in just over 60 companies in our three funds, roughly half of them are SaaS companies, ca. ⅓ are marketplaces and some 15% are outside of the two main categories.

I originally posted the below text regarding internationalisation of startups on my old blog, pawel.ch, in March of 2014. Having just come across this story about how Uber is losing vs. local competitors in some countries reminded me about the characteristics of multi-local marketplaces and how tricky their internationalisation can be. If you are interested in this topic, I hope you will enjoy this blog post or maybe even find it useful, especially if you did not read it back when I posted it.

---

At Point Nine, we spend a lot of time thinking about how startups internationalise. As we are mostly active in Europe and given it is hard to build a really large company only addressing one of the European countries, most of our portfolio companies face the challenges of internationalisation at some point in their lives.

My current thinking is that as far as internationalisation is concerned, there are two types of startups: global and local/multi-local startups. It is very important for founders to understand which category their company is in, as it will have significant consequences for how the internationalisation process will evolve and how it will impact the development of the company.

Global startups

Global startups address an international audience from day one. They will typically launch their product in English and might add additional localised versions later. Many of our SaaS startups, even those based in Europe, are global startups. For example, Contentful, infogr.am or Algolia all fit this definition of a global startup. Being a global startup has following consequences:

- It is “easy” to go international. You sell to international audience from day one.

- Sooner or later you will have strong direct competitors, so better get funded and move fast. If the opportunity is significant and it is global, someone else out there will notice and try to exploit it.

- Especially in software, if you are global, you have to be strong in the US market, the biggest software market in the world. If you lose the US, it can be hard for you outside of it.

- First you need to master the English / US version of your product and sales and marketing. Local language versions or localised sales might be necessary at later stages of the company to scale it really big, but not to get to the first significant scale.

- The financial outcome of your startup journey will probably be binary - either you will create a very significant company or you will be crushed by competition or die of other problems.

Local and multi-local startups

(Multi-)local startups are different. They typically start with a product offering tailored to one country and after gaining some experience in the initial market go to another market and another, in a country by country fashion. Examples from our portfolio would include Delivery Hero, Kreditech or Docplanner.

- (Multi-)local startups are “hard” to internationalise. Every new market is a new logistical challenge. This can be especially hard in ecommerce, but marketplaces are not easy either. Figuring out a fast and efficient way to rollout new markets is key.

- Competition in the local markets will vary between zero and moderate, rarely will it be very sophisticated, aggressive or well funded.

- As you gain experience in a multitude of markets, it will be hard to compete with you.

- US is not a must. More importantly, US competitors will typically not be a threat. US competitors who start with a local business model in the US, frequently do not internationalize at all or go only into a few countries, do it late or are not good at it (see GrubHub Seamless, Zocdoc or Amazon).

- It is hard to get to a really big scale as it will typically require winning a large number of countries or winning in the biggest, most competitive countries (like Germany, UK or France).

- The outcome does not have to be binary. You can make it in one or a few small and mid-sized countries, fail in bigger markets and you can still have a decent exit.

Of course, the world is not as black-and-white as painted above. One can imagine that a company launches a global product offering following some success locally. Or that a company is not easy to categorise, like in the case of Spotify which goes country by country, but is building a global brand.

While certainly not perfect, I like this way of looking at the internationalisation of startups. If you have any thoughts or experiences that support, contradict or simply add to the above, I would be thankful to read about them in the comments below.

A while ago Christoph wrote a post titled Good VCs, Bad VCs which illustrates how we perceive our job as VCs and what we aspire to act like. After a discussion we had around this a few weeks ago I tweeted out the following question: “What actions by VCs do you consider most 'founder friendly' or most 'founder hostile'?’”. Quite a few folks replied (many thanks!) and the results are very interesting. You can review them here (twitter) and here (facebook) and below you can find a summary of what was most frequently mentioned in both camps.

Top unfriendly VC moves

Pre-investment behaviour / terms

1. Lack of transparency in the decision making process

2. “Going dark”, i.e. no response

3. Multiple and/or participating liquidation preferences

After investment

1. Ousting founders

2. Forcing founders to do things/deals they would not want to do

3. Not picking up pro-rata (bad signaling)

Top friendly VC moves

1. Being fast, responsive and transparent, on all fronts, pre and post investment

2. Helping founders grow

3. Offering pro-rata/money in difficult situations

While nothing on the list was totally unexpected, I was surprised by the frequency with which the quality, transparency and speed of the fundraising process and communication with VCs got mentioned versus other aspects of the founder-VC relationship. This tells quite a bit about how founders perceive the responsiveness and quality of VCs in general and also points to the area where VCs can improve the most. It also indicates what being ‘founder friendly’ really means - and it is not necessarily paying crazy prices or being extremely light on terms, but goes more into the direction of being a reliable, transparent and quick partner.

If you want to help us understand the fundraising process from the point of view of an entrepreneur a bit better and hopefully become a better VC as a result, please consider answering a few questions about this in a survey on fundraising that Christoph is currently running.

Every now and then an entrepreneur will ask us for an NDA. We do not like signing them and we very rarely do (and never in the beginning of a discussion). We have signed a few in the past and it always had only one effect: it slowed down the discussion between us and the startup. 

Below you will find a very quick summary of why I do not think NDAs between startups and VCs are a good idea. Instead of going into the details of each of the points, I will post a list of good blog posts around these topics, since so much has been written about NDAs already. I will aim to expand this list and add new interesting posts as I come across them. Then, I will just send a link to this post to the next startup founder asking for an NDA.

Why NDAs are not a good idea (summary):

• impossible to manage for the VC
• slow down the discussion between startup and VC
• signal lack of trust in the beginning of a relationship
• in practice do not provide protection anyway
• execution of an idea and the team around it is so much more important and defensible than the idea itself

Selected blog posts on the topic of NDAs:

• http://www.bothsidesofthetable.com/2009/07/26/on-ndas-and-confidentiality/
• http://startuplawyer.com/venture-capital/why-a-vc-will-take-a-lighter-to-your-nda
  
• http://www.feld.com/archives/2006/02/why-most-vcs-dont-sign-ndas.html
  
• http://avc.com/2006/02/why_vcs_dont_si/
• http://www.entrepreneur.com/article/245023
  
• http://www.forbes.com/sites/ericwagner/2013/07/09/in-search-of-funding-7-core-principles-to-raise-capital/
  

Please let me know if you have any comments or additions to the above lists.