A sneak peek into Point Nine's investment thesis

Over the last couple of weeks and months we spent some time putting our investment thesis on paper. The purpose of this exercise was to challenge and discuss our implicit assumptions and to get everyone on our team aligned on what kind of investments we seek.

One of the things that being very clear about our investment focus helps with is getting to “no” faster. If that sounds pessimistic, remember that we see thousands of potential investments every year but can only do 10-15 of them. Just like it’s crucial for sales teams to have clear qualification and disqualification criteria, it’s important for us to focus our time on “higher probability deals”. That means we’ll have to be able to quickly pass on a large number of deals that are likely not a good fit for us. Our “filter” is of course not perfect, so we’ll inevitably pass on lots of great companies, some of which will end up in our growing anti-portfolio – but there aren’t enough hours in the day to take a close look at each company that we see.

A fast decision process is also important for founders. As we’ve learned from this survey, being left in the dark is the single most important reason why fundraising often sucks for founders. We will obviously never be able to make decisions based on a simple algorithm, if only for the fact that the founding team remains the most important of all criteria. But anything that helps us streamline our decision making process is welcome.

Once the document is in a publishable form we will post it. Bear with us for a little while as we’re polishing the document a bit to make it more self-explanatory and to remove the worst typos. ;-) In the meantime, here’s a sneak preview.

We will continue to focus on two business models: SaaS and marketplaces

SaaS

• We use a broad definition of SaaS. Usually the first “S” stands for “software”, but sometimes it stands for “something”, e.g. a combination of software and hardware or software and data.
• We’re interested in horizontal and vertical SaaS. What counts is that the startup is aiming to solve a big enough problem for a large enough number of potential customers in order to build a big business. As a rule of thumb, we’re looking for markets that consist of at least 3,000 whales ($1M ACV), 30,000 elephants ($100k ACV), 300,000 deer ($10k ACV) or 3M rabbits ($1k ACV). ¹
• We’re equally interested in companies targeting SMBs (AKA rabbit and deer hunters) and companies targeting enterprises (AKA elephant and whale hunters). What’s important is the right founder/market fit. For companies targeting very small businesses (AKA mice and rabbit hunters) we want to see the potential for viral distribution.
• We’re looking for companies that we think can build a 10x better product and/or drive a paradigm shift in the industry. ²
• We want to invest in companies that can eventually build moat e.g. by becoming a system of record or a “system of intelligence”; by building a large data set that in combination with machine learning translates into a superior product; by building a platform; or by becoming a SaaS-enabled marketplace.
• With very few exceptions in areas like accounting, we’re looking for companies that have the potential to win the US market.
• We’re looking for SaaS companies that have the potential to get to $100M in ARR within 7-8 years and to $250-300M ARR within another 2-3 years.

Marketplaces

• Like in the case of SaaS, we use a broad definition for marketplaces. For us, a marketplace is a digital platform that brings two or more parties together and enables them to “transact”. The object of the transaction can be a physical product, a digital product, a service, or in some cases a piece of information or knowledge.
• We look for startups that leverage marketplace dynamics to create unique user experiences in fragmented markets, with the potential to develop a moat through network effects.
• We believe that marketplace platforms will continue to emerge in the most unexpected of places and in the most unexpected of forms. They will continue to transform entire industries.
• We are open to all of C2C, B2C, B2BC and other types of marketplaces. We are particularly excited about B2B marketplaces and “SaaS enabled marketplaces”.
• We are trying to identify platforms able to become international leaders. Thus, we will typically look for early proof of ability to operate in more than one country or globally.
• We are looking for early signs of liquidity. ³
• We look for founding teams with strong commercial sense.
• We think that blockchain technologies have the the potential to disrupt many marketplace models as we know them today; we will be exploring them in depth.
• We look for marketplaces that can become truly significant. In monetary terms, this means the potential to ultimately generate hundreds of millions of dollars in annual net revenues and billions in GMV.

Thanks for contributing this section, Pawel. Expect a follow-up post with more details from Pawel (who’s leading most of our marketplace investments) soon.

We will continue to invest in new areas and technologies that we like to dub “Frontier Tech”

• While we’re focused on two business models – SaaS and marketplaces – we’ll continue to keep our eyes wide open with respect to new technologies.
• We’re extremely interested in new opportunities in areas such as AI/ML, blockchain and cryptocurrencies, IoT and hardware-as-a-service, drones, or AR/VR. We have already made investments in most of these areas and will continue to do so.
• In many of these cases there are complex tech problems that must be solved. We’re happy take a certain level of technology risk, but at the same time we’re looking for founders who find ways to bring a product to the market quickly and cheaply.
• While a superior technology will usually be key to entering the market and have some early wins, most technologies will eventually be commoditized. Therefore we’re looking for additional sources of long-term defensibility such as high switching costs and large data sets (see the section on SaaS above) or network effects (see the section on marketplaces above).

Thanks to Mr. Frontier Tech Rodrigo for your help with this section, and looking forward to your follow-up post as well.

We will continue to focus on early-stage investments

• We’ll continue to focus on seed investments, investing anything from a few hundred thousand dollars up to around $2M in “seed” and “late seed” rounds, typically in companies that have strong indications of Product/Market Fit and promising early traction.
• We will continue to make what we call „founder bets“: Idea-stage investments into proven entrepreneurs from our close network. In these cases most of our „rules“ don’t apply. When people like Doreen Huber, Fabian Siegel, Iñigo Juantegui, Pan Katsukis, Sebastian Diemer or Stefan Smalla start something new, we want to be part of it. ⁴

We will continue to invest internationally

• Europe is our home market – we’ve made investments in most European countries and we’ll continue to invest all over Europe.
• Especially in SaaS we will continue to invest outside of Europe as well – e.g. in the US, Canada, Australia, New Zealand and other countries.
• In SaaS, our assumption is that you can start almost anywhere but you have to win globally (which requires winning the US). In marketplaces we want to find companies that can win several large markets.

We continue to aspire to be a “Good VC”

• We don’t pretend to be the right investor for every startup. But our aspiration is that if we do invest in a company, we’re the absolute best partner the founders can dream of and that we’ll play a significant role in helping the company get to the next stages.
• We’re optimizing for the long run in everything we do. You “always meet twice in life”, as the German saying goes.

_________________________

¹ Check out this post if you have no idea what I’m talking about. Then, get your poster.
² See Sarah Tavel’s post about “10x better and cheaper products” for a similar concept from the consumer Internet world.
³ Defining liquidity is tricky – a topic for another post!
⁴ True story – these are all guys who we backed or worked with closely before and who subsequently founded Lemoncat, Marley Spoon, OnTruck, Remerge, Finiata and Westwing, respectively.

What animals are WE hunting?

[This article first appeared as a guest post on VentureBeat. Thank you for publishing it, VentureBeat. I'm re-posting it here with a few small edits.]

Of all posts that I’ve written so far, the one in which I asked what kind of animals you’re hunting was one of the most popular ones. That begs the question: What kind of animals are we hunting?

Paul Graham wants to farm black swans. Dave McClure likes ugly ducklings, little ponies and centaurs. Almost all large VC funds are looking for unicorns, while some people argue that investors should hunt dragons and others talk about decacorns.

If you have no idea WTF I’m talking about, here’s a quick refresher. The term unicorn was coined by Aileen Lee about two years ago to describe those rare and magical tech startups that have reached a valuation of $1 billion or more. Since then, $1B valuations have become somewhat less rare and there are now several private tech companies valued at $10 billion or more, for which the industry has come up with another name: decacorns. Before unicorns were called unicorns, people used to call these rare outlier companies, which create massive returns for their early investors, black swans (or homeruns – back then, it wasn’t mandatory to borrow terms from the animal kingdom). Duckling is Dave McClure’s name for companies that don’t become quite as as huge, and ponies and centaurs is what he calls the ones that have reached valuations of $10 million and $100 million, respectively. Finally, a dragon is a company that returns an entire VC fund.

So – what I mean by the question in the title of this post is what kind of exits we are aiming for. It’s a question which every VC needs to think about: If you have, say, a $250M fund and your goal is to return $1B before costs, should you aim for one huge outlier, e.g. a $10B exit in which you own 10%? Or are you better off shooting for 20% stakes in 50 companies which exit at $100M each? Or something in between?

For large funds the answer is pretty clear. Although the number of smaller exits is of course much bigger than the number of large exits, the exit value is highly concentrated on a small number of huge winners. This power law distribution of venture returns, which Peter Thiel has spoken about extensively, is what makes it almost impossible to return a large fund without hitting one or more outliers. Or as Jason M. Lemkin put it: VCs need multiple unicorns just to survive.

But what about a small (~$60M) early-stage fund like ours? We spent a lot of time thinking about this question in the last years, and our conclusion – or, let’s say working assumption, because it’s still early days for us – is that (sticking to the terminology described above) we’re hunting for dragons, hoping for unicorns.

In spite of the growing number of unicorns in the last years it’s still exceedingly rare for a startup to reach a valuation of $1B or more. According to Aileen Lee’s research, only 0.14% of venture-backed tech startups become unicorns. We can make around 30-40 investments with our fund, so statistically the chances of hitting a unicorn are very low. That doesn’t mean that we’re not trying hard to beat the odds – and if you don’t believe that you can beat the odds you should never become a founder or a VC in the first place – but it means that our business model is not dependent on having a unicorn in every fund that we raise.

We’re small enough for not being dependent on unicorns, but – and that’s the big difference to angel investing – we’re too big for generating a great performance by piling up a larger number of small exits. If we tried to get to, say, $240M in exit proceeds in chunks of $10M (corresponding with e.g. 20% of a $50M exit) we’d need 24 of these exits. It’s not realistic that 60-80% of the companies, in which we invest at a stage when there’s often just a handful of people and a few thousand dollars in revenues, will go on to become $50M exits though. That’s why we need a few of the animals which in the beginning of this post have been called dragons and which we internally just call “fund-makers”: Investments which return an entire fund, which in our case means, for example, 20% of a $300M exit or 15% of a $400M exit.

The final question is if all of this has any practical implications at all. Isn’t it impossible to look at a seed-stage startup and predict how large it can become anyway? Those are very hard prediction to make indeed, but still, knowing what kinds of exits we need informs several important decisions that we have to make – how many companies we want to invest in, what ownership stakes we’re aiming for, how much capital we reserve for follow-on financings, and so on. It also makes it clear that we shouldn’t invest in companies which for some reason we feel don’t have enough potential to move the needle for our fund.

The very last thing I want to say, just to be sure that I’m not misunderstood, is that I have absolutely nothing against unicorns. :-) In fact, we love ‘em. We’ve found two so far, Zendesk and Delivery Hero, so we’ve seen the beautiful side of the power law distribution first-hand. So: Hunting for dragons, hoping for unicorns.

Four (more) things we look for in SaaS startups

More than two years ago I wrote about what we look for in early-stage SaaS startups. Since then we've looked at hundreds of SaaS startups and have gained additional insights through the work that we've been doing with the SaaS startups that we have invested in. Therefore I thought it would be time for a follow-on post with some additional thoughts.

In the original post I focused primarily on early metrics as an indicator of product/market fit and of a favorable CAC/CLTV ratio in the future. Today I want to put more focus on factors that kick in a little later in the lifecycle of a SaaS company – aspects that have an impact on a company's ability to scale customer acquisition, increase ARPA and create lock-in. In other words, factors that can make the difference between a "good" and a "great" business.

Note that none of these factors is a must-have for building a successful SaaS company. For each one of them you'll probably find some great counterexamples. The point is that all other things being equal, these characteristics increase the odds of creating a big SaaS success:

1. High search volume combined with limited SEM/SEO competition

Search volume on Google is a good indicator of the awareness for the problem that you're solving. You may have a fantastic solution for a big problem, but if no one is looking for it, marketing it will be much harder. It means you'll have to spend more effort on educating the market and that you may not have a lot of low-hanging fruits on the customer acquisition front.

Related to search volume is of course competition for the relevant keywords, both with respect to SEM/PPC and SEO. If there's lots of competition for your keywords, PPC advertising might be prohibitively expensive and SEO will be much harder.

If, in contrast, there's high search volume and limited competition this not only indicates demand for your product, a gap in the market and potential to acquire customers via SEO/SEM. It also means that you have an opportunity to establish yourself as the thought leader in your space by doing great content marketing.

2. "Land and expand" and "bottom up" customer acquisition

Selling to big enterprises is tempting because one big enterprise deal can be worth tens or hundreds of thousands of dollars. But it's also tough: Sales cycles are long, you need to convince various different stakeholders, there are special requirements for the product and you have to do multiple meetings to get the deal. Anyone who's done or tried it knows what I mean. Conversely, selling to SMBs is much easier, but the value of each customer is obviously a lot lower as well.

A "land and expand" or "bottom up" customer acquisition strategy has the potential to give you the best of both worlds. There are different variations of this strategy, but the idea is always that a single user or a small team of people inside a company starts using your product, making the initial sale easy (if any "selling" is involved at all). Over time, more and more people inside the company use it, and eventually you can sell an enterprise account to the entire company.

Perhaps the most famous example of a successful bottom up adoption is enterprise social network Yammer. Within the first two years after launch, the company's freemium distribution model attracted users from 80% of the Fortune 500 companies and got Yammer into more than 90,000 customers. According to this Mashable article, 15% of these companies subsequently upgraded to a paid plan.

If you want to follow in Yammer's footsteps (or just copy some pages from their playbook) here are some of the things you should keep in mind:

• Since you want to sign up users with little to no sales efforts you need a great marketing website and frictionless onboarding.
• Your product needs to provide value for a small number of users inside a company but even larger value if more people use it.
• Your pricing needs to be highly differentiated – make your product cheap or even free for a small number of users to maximize distribution and make money out of bigger accounts.
• Once you want to sell bigger team accounts or enterprise accounts you need to provide the functionalities required by bigger companies (a sophisticated role/permission system, SLAs, audit logs, etc.) while still keeping the product easy to onboard and use.

3. Virality

It's very rare for B2B SaaS applications to get really viral, i.e. have a viral coefficient of over 1. However, even though your SaaS product will never get Hotmail/Skype/Instagram/Snapchat-like growth, any level of virality is valuable because it means you're augmenting your paid user acquisitions with free users.

There are two primary ways in which a SaaS application can be viral:

a) "Sharing"
A use case which involves communication, collaboration, file sharing or the like with external parties. Examples include project management software like Basecamp (where e.g. an agency invites a client to a Basecamp project), e-signing solutions like EchoSign (where the person who is asked to sign learns about EchoSign during the process) or file sharing providers like Dropbox (you got the idea). The more affinity there is between your target group and their "collaborators", the better it is for you, since it means a higher "invite to signup" conversion rate.

b) "Publishing"
A use case where your customers use your software to create something which gets published on the Web. Examples: Shopify, SquareSpace, MailChimp or our portfolio company Typeform. Another example is Zendesk's feedback tab. The signup conversion rate is much lower in this case, but it can be offset if your product gets exposed to large numbers of people.

You can't force it if there's no sensible "sharing" or "publishing" use case for your product, but you should think about it carefully. If sharing or publishing doesn't make sense for you, you can still get some virality in other ways:

• Employee fluctuation: If you have a product that is used by lots of employees inside a company, try to make everyone an ambassador of your software who will suggest using your product in future jobs.
• Referral programs: FreeAgent's user-to-user referral scheme is a good example.
• Incentives along the lines of "get XYZ for a tweet", where users can e.g. unlock features or remove limitations by inviting people to your product.

4. Economic moat

In the first couple of years you shouldn't worry too much about your long-term competitive advantages. Oftentimes execution is everything. Working harder than your competition, innovating faster and just doing everything a little bit better goes a long way.

Having said that, the best and most profitable companies in the world are those which manage to create wide moats around them – sustainable competitive advantages that allow them to keep market share and profit margins in spite of aggressive competitors. The best examples for wide economic moat are patents (think pharma) and natural monopolies (think eBay).

These two examples aren't very relevant for SaaS companies and there is no simple silver bullet for creating sustainable competitive advantage, but there are a couple of factors which can create moat around a SaaS business:

• A platform. The best example is the Force.com platform. The large number of applications that integrate with Salesforce.com make Salesforce.com the most comprehensive CRM solution on the market and give the company a huge competitive edge. This is a classic example of a virtuous circle: More customers attract more developers which in turn attract more customers. When a platform has reached a certain size, it's very hard for competitors to attack you. 
• Distribution channels: If you have thousands of partners who have been trained to sell your software and make a lot of money doing it, this can be another very valuable asset. Admittedly the role of VARs and other distribution partners is typically lower in SaaS than it is in traditional enterprise software, and the best example of an extremely valuable VAR channel is probably SAP.
• Lock-in: A great product with a fantastic user experience alone can create significant lock-in. But different types of SaaS products have different levels of lock-in. The more people inside a company use your product, the more business partners interact with the software and the deeper the product is integrated into a company's core businesses processes, the higher are the switching costs.
• Network effects: Great examples include Freshbook's billing network and MailChimp's eMail Genome Project. What these two examples have in common is that (at least in theory) every users makes the product more valuable for all other users.
• Big data: If you have tens of thousands of customers, the massive amounts of data created by your customer base might allow you to draw insights which you can then give back to your customers. Zendesk's benchmarking reports come to mind as an example.

Failure IS an option

Failure may not have been an option for the Apollo 13 mission, but it certainly is an option for startups. In fact, since statistically the majority of startups fail, you could argue that it's the default option.

Most successful entrepreneurs have a few failures under their belt, and most "overnight" successes are the result of years' of hard work – and in many cases years' of trial and error. Before writing history with Angry Birds, Rovio had already launched 51 games that you've probably never heard. Brian Chesky described AirBnB as a an "'overnight' success that took 1,000 days".

I've had my fair shares of failures as well. It took me a text adventure for the C64 (never completed), a mail-order business for Amiga shareware (a decent success for a student business, but discontinued when the Amiga died), a PC real-time simulation game (nice game, but didn't manage to get it properly distributed) and several other attempts before I had a decent success with DealPilot.com, which I co-founded in 1997. Likewise, as an investor I've made several investments that didn't work out before landing my first big hit with Zendesk.

Considering how normal and necessary failure is in the startup world, it's surprising how many VCs are afraid of admitting failure when it happens: Logos are quietly removed from portfolio pages^*, asset deals are arranged to make it look like a successful exit, PR stories are written. What's even worse is if the death of a dying startup is delayed by putting more money into it, all out of fear of admitting failure.

If you invest in early-stage startups, you know that a large part of them, maybe more than half, won't make it. The rest of the world knows it too. So why not be open about it?



* We're guilty of this too, but we're thinking that we should keep all logos on the page and add a "R.I.P." badge when a startup died. What do you think?

KPIs for VCs

Example for a Geckoboard KPI dashboard

Last week I spent a day in Stockholm to attend a metrics seminar organized by our friends at Creandum. It was a great event with talks from people of some of the best Internet companies from the Nordic region such as Spotify or Wrapp. Thanks Johan, Joel, Daniel, Frederic and everyone at Creandum for setting it up and inviting me!

I did a talk about SaaS metrics (I'll post the slides shortly), and in the Q&A session Andreas Ehn asked a really good question:

"As a VC, what are the most important KPIs for yourself?" 

Ultimately our #1 KPI is the return that we deliver to our LPs. If you're new to the world of venture capital, LP is short for "Limited Partner" and means the people and funds which have invested in our fund. That return is expressed as a return multiple or as the internal rate of return (IRR). As it obviously takes a lot of time to build (and eventually sell or IPO) great companies it will of course take many years until we know our final performance. Like most VCs our fund is set up for a lifetime of ten years.

In the meantime we (and other VCs) track our performance by:

1) Adjusting the value of our portfolio whenever a portfolio company raises a new round of financing from a new investor at a new (hopefully higher) valuation. While there's no guarantee that we will ever sell our shares at these "Fair Market Valuations" (FMVs), the assessment of the portfolio based on current FMVs is usually the best way to measure success. Valuations are usually marked up on an ad hoc basis internally (i.e. when a new round closes) and reported to LPs on a quarterly basis.

2) Monitoring our portfolio companies' key financial data, KPIs and operational performance. This is the best near-time proxy to long-term success, and so we're constantly looking at these things. We usually get either access to live dashboards or monthly reports and I'm hoping that we'll soon find the time to create a beautiful Geckoboard dashboard with the top KPIs across the entire portfolio (requires some work because we get data from portfolio companies in a variety of different forms and shapes).

Besides these pretty obvious ones there are a few other KPIs that we're looking at:

Number of deals that we're evaluating

It's not a KPI in the sense that there's a direct "the higher the number, the better it is" correlation, as quality of deal-flow is of course more important than quantity. But there is a connection between quantity and quality, and since we're using Zendesk to track each potential investment it's easy to monitor this number (for what it's worth, we're currently at deal #3,773 since we started using Zendesk about two years ago, and in the last 30 days 148 new ones have been added). 

Response time for investment inquiries

For founders it's important to get fast responses, even if the answer is "no". Depending on our workload sometimes we're fast and sometimes we're slow. There's still a lot of room for improvement, so this is a KPI that we're going to keep a closer eye on in the future.

"Rating" of our responses

Zendesk allows you to let your end users rate the customer support experience for every support ticket. We're not using this feature yet, but I'm wondering if we should do it in order to keep track of how successful we are in leaving positive impressions with the entrepreneurs that are pitching to us.

How well are we at picking the right investments?

Of all the potential investments that we look at, how well are we at picking the winners and avoiding the losers? And how well are we doing when it comes to allocating follow-on investments among our portfolio companies? We're not yet using a simple set of KPIs to track this, but we're regularly reviewing our past deal flow, trying to understand when we were right and when we were wrong and what we can learn from it.

Finally, there's one other KPI, and while it's again not something you can quantify on a short-term basis, it's just as important or even more important than our fund performance in the long run: It's the concept of Net Promoter Score applied to us. What it means is that when we ask our portfolio founders two simple questions – "Would you raise money from Point Nine in your next startup?" and "Would you recommend Point Nine to other founders?" – we want to hear two wholehearted YESes.

PS: Just like Web startups have their vanity metrics, you can also hear VC talk about vanity metrics – i.e. metrics which sound good but don't mean much. I'll leave that for another post.