Speedinvest Blog

How Do VCs Measure Traction in Deep Tech?

by 

Dominik Lambersy

August 19, 2020

A deep dive into how VC’s really measure commercial validation, which was one of the most asked questions from our very first Deep Tech AMA.

Image for post
(Photo by Chris Hearn on Unsplash)


How abstract is traction in Deep Tech?

By demystifying Deep Tech traction and commercial validation, we hope to give you an overview and an idea of the benchmarks to investigate BEFORE making your seed pitch to investors. We will individually break down how we at Speedinvest evaluate traction for enterprise, open-source and frontier tech during our investment decision process.

FYI: If you want to attend future events organized by the Speedinvest Deep Tech team or have any questions you would like us to answer in future blog articles or AMAs, please sign up and share your ideas with us here.

Image for post
Overview of traction markers in Deep Tech
How many MAUs does your seed stage quantum-inspired machine learning protein folding startup have?

— any Deep Tech inexperienced investor, always

Thinking about such a question has an unknown, yet, intriguing feeling to it, doesn’t it? Often venture capital conversations are about MRR, MAU/DAU, CAC, ARPA, Retention, avg. order value, depth of engagement and many other metrics. However, I hate to break it to you, in many cases Deep Tech ventures prove themselves differently compared to a consumer or other SaaS products.

Validation is a tough topic to wrap one’s head around, especially when one considers that Deep Tech companies often rest on a foundation of intense research and thought work. The question is made yet more complex by the fact that such startups not only take a longer time to find product-market fit than consumer software products, but also differ significantly in the strength of the signals provided. Whereas a consumer software application scores 1000s of detailed usage patterns, Deep Techs find themselves with the completion of a handful of proof of concepts (PoC) after six months.

Enter the Enterprise

Plenty of the companies we look at sell to enterprise customers with annual contract values exceeding 100K. This means 6–12 month sales cycles until an agreement on a multi-year contract is signed.

Revenues play a crucial part but are not everything.

Ever heard of “put your money where your mouth is”? There’s a good reason this is such a well known saying. While monetary commitments are often the strongest signals you can show us (we usually like to see at least 10K of monthly revenue), early engagements between your startup and other enterprises are also valid indicators of traction.

Here are some questions to help guide you in leveraging your early engagements:

  • Are senior enterprise employees interacting with you on a regular basis?
  • Are they potentially opening their organization or their customer base to you?

… and think about how you can articulate this to a potential investor!

The art of early-stage investing is to engage with winning teams who are building superior products with measurable benefits for their customer audience. Hence, the first conversation usually starts when startups are mostly in the PoC stage.

What is a Proof of Concept, you ask?

Proofs of Concept (PoCs) are the most common type of market validation we encounter in early Deep Tech startups. Usually, they are highly tailored projects with B2B / Enterprise customers in which startups and their prospective customers validate that their solution solves an actual, sufficiently valuable and immediate problem. There is nothing unusual about a PoC not being paid. Often the reason for unpaid PoCs is to improve the product or to develop a better pricing thesis. Nonetheless, never sell yourself below value. If you know there is significant value creation, then it will not be an issue to negotiate a reimbursement for the PoC.

In case you do have some advanced and paid proof of concepts, think about how likely they are to convert into full-blown customers. Even better, consider which criteria would lead your potential customer to deem your PoC successful. This exercise also helps you to identify the truly crucial outcomes that matter most to those who will pay for what you offer. In an ideal case negotiate an automatic conversion to a full commercial contract when those success criteria have been satisfied.

We usually validate the likelihood of conversion by conducting reference calls and evaluate your future revenue potential by answering the following:

  • How much pain is the problem causing, and how is that solved today?
  • How engaged are customers with the solution provided today?
  • How much value is created by using the new solution?
  • How difficult is it to substitute legacy systems?

No matter whether the PoC was paid or not, tell your potential investor what you have learned from a PoC and how you improved.

Our Investment Sweet Spot

  • At least a handful of successful PoCs which are about to convert to full-blown and referenceable customers
  • Positive user references
  • At least 10k in monthly recurring revenue

Open4all - How Open-Source Software Popularity is Understood

Aside from the common “top-down” sales approach of startups that are trying to win enterprises, another proven way to go to market is through the concept of open source. We observe an expanding number of startups that enter the markets by first building critical masses in developer communities. They then leverage those open-source assets later to create a stronger position from which to commercialize and monetize.

109 out of 202 [ MLOps ] tools I looked at are OSS. Even tools that aren’t open-source are usually accompanied by open-source tools.

As a data scientist, I am a huge fan of open-source software (OSS) projects. Open source slims down the pain of building a prediction engine to just a few lines of codes. It doesn’t stop there. The history of OSS started earlier than my beloved python libraries. One older player, Automattic, the company behind Wordpress.com (not org) — which powers 35% of websites online (that makes 455 million!) — has drawn around $700 million in funding since its inception in 2005. It allowed everyday internet users to build and publish their own websites without any background in HTML or file transfer protocols. And people were hooked! Monetization followed with exclusive advertisement and additional premium add-ons to enhance the online experience. There are a handful of other monetization strategies for OSS projects depending on the product that is being built.

Back to the topic, present investments in OSS had a CAGR of 10% over the last 5 years, amounting to $2bn invested just in 2019.

The question is how to convince an investor to take action with your early-stage OSS project when it will take three to five years until the first revenues will show their shy faces?

What they do, is what you got — User Activity

Early-stage OSS companies have the advantage of being able to gain visibility through early user activity. Do not underestimate this information. It carries tons of interesting insights about the potential success of a venture. When collecting metrics as an OSS project, you have to strike a careful balance between the granularity of the “telematics” you collect and what your community is prepared to share.

As a founder, your major goal should be to build a sticky product that users cannot keep their fingers off. Therefore, try to think of metrics that demonstrate the engagement of your users.

As an example exercise, imagine a DevTool startup that supports the user with code auto-completion:

  • Are they downloading your tool and using it after installation?
  • Are demos converting at a high rate?
  • Does monthly activity show that the tool is used 20 days a month (aka.. it is an essential tool for the users, demonstrating product-market fit)?

Remember that there is no clear rule of thumb about how much user activity is required at any given point of your fundraising process. In particular, as the whole game depends on how you define certain activities and the type of OSS tool you are building.

Let’s get back to our DevTool example. What could be other ways to define activity?

  • Lines of code auto-completed?
  • The average usage time per day or per session?
  • The MoM growth of GitHub stars you receive?

Can you think of more? Comment below I am curious to hear your ideas!

There are plenty of ways you, as the founder, have full control over writing your validation story.

Take a look around OSS

Let’s take a dive into some recent open-source funding rounds and look for common denominators in their press releases.

Image for post
Examples of traction at different stages.

OSS lives and dies with user activity. The above press releases tell a whole story about how user activity moved the needle to serious investment, no matter which investment stage or geolocation. A similar platform we see on the rise as a sign of early community validation is Product Hunt. Often, I find myself skimming through the technical categories looking for projects with high upvotes, the number of comments and their general sentiment.

It is hard to define threshold multipliers in the early stages, but when plotting the graph there should be indications of sustainable growth. Coincidentally, there is a great open-source tool to do the plotting.


Image for post
Source: Star History

The curve from Iterative.ai has a beautiful, steadily increasing growth curve. However, this is more of an exception. Usually, repos have more of a linear behavior with certain hype jumps when there is increased attention from news features or when articles being heavily shared in developer communities. We are well aware of the possibility to buy stars, hence, validating these growth jumps is always an investigation worth taking.

Our investment sweet spot

  • 800 to 1,000 GitHub stars
  • Observable growth in stars
  • Activity metrics such as contributors, forks, stickiness and depth of engagement analysis

Being on edge — Is frontier tech measurable?

Looking at early Frontier Tech and predicting what’s to come is even more complicated than exploring the dark depths of our oceans.

There is not a lot of visibility as the diver swims ahead and the hope relies on the person’s:

a) skill and expertise.

b) persistence to maneuver around deep-sea mountains and into open waters.

Let’s apply the analogy. As a seed-stage investor, we evaluate frontier tech by:

a) the intellectual property and the expertise the leadership team holds.

b) the technology’s potential impact on the market.

With investing in (a) a bright team with a deep, unique understanding of the domain and some early feeling for the commercial strategy being the by far superior variable.

There is no real way to measure frontier tech traction in the early seed stages. However, we can interpret external trends and market movements as either working in favor of or against a particular frontier tech company. Think about the amount of academic research dedicated to a topic. This could be an early justification for a certain technology. Take the quantum computing space as an example. A search in the world-renowned scientific journal “Nature” shows that since 2010 there have been 7891 publications that mention the phrase “quantum computing.” In the roughest estimation, one could say that there are around 2000 to 4000 true experts in the field on our entire planet. Hence, investing in frontier tech is — again — more about investing in the right team. Nonetheless, this is content for another blog article.

Our investment sweet spot

  • Strong team with substantial knowledge edge
  • Defensible IP
  • Large impact on a sizable market

Traction isn’t everything!

To summarize, when we evaluate the traction of Deep Tech startups, we try to use a lens that is most appropriate for that specific startup:

  • Enterprise Tech: Initial EUR 10k of MRR + a handful of meaningful PoCs/Pilots that have a high likelihood to convert.
  • Open Source Software 800 to 1000 GitHub stars + early developer/community adoption and activity
  • Frontier Tech: A compelling and very large opportunity + a stellar founder team

Keep in mind that many early-stage startups we look at, as well as invest in, only fulfill some of a full range of traits. Traction is only one of the traits we look at. You should check out our blog article on Why Deep Tech needs to be looked at differently, and our recent feature in sifted on How not to pitch your deep tech startup. ;) In case a detailed guide about startup metrics in any industry and situation could be of interest to you, get yourself a free copy of the ultimate guide to startup metrics by our growth partners from Speedinvest Pirates.


Authored by Dominik Lambersy, a member of the Speedinvest Deep Tech team, with contributions from Marcel van der Heijden, Namratha Kothapalli, Alex Zhigarev, Dominik Tobschall and Taylor Anderson. The team backs pre-seed and seed-stage European tech startups with global aspirations. With a focus on AI, cybersecurity, cloud-native infrastructure, developer tools and open-source projects, they use their extensive, industry-specific network to support founders along every step of their journey.

Sign-up to get the latest Speedinvest news, insights and resources right in your inbox.

From The Blog

Discover more articles like this on the Speedinvest Blog