• CDF Technologies

Five AI use cases that your VC fund should start employing

It’s 2020 and AI has been mentioned more than ever. It holds the premise to improve almost every area in life and to impact almost every industry. Have you ever wondered what AI can do in the venture capital (VC) market? After having worked in this field for the past years, we want to share some of the use cases that are already in use.

1. Finding the best startups Without doubt, sourcing is the biggest use case for VC funds and it is already employed widely. Searching and selecting the best startups represents the most basic daily routine for VC funds and consumes much of their time. It requires a lot of networking and communication — which is highly inefficient. With the help of AI-driven web crawling, VCs can resolve these inefficiencies and screen the web for interesting startups to consider and meet with the most interesting ones. This does not only save time but moreover helps VCs to find startups earlier than other investors — through which they can gain a significant competitive advantage. Hone Capital, a Silicon Valley-based venture capital, is famous for its machine-learning model. It analysed data from AngelList, Crunchbase, Mattermark, and Pitchbook to successfully build an AI sourcing model. “One of the insights we uncovered is that startups that failed to advance to series A had an average seed investment of $0.5 million, and the average investment for start-ups that advanced to series A was $1.5 million. So if a team has received a low investment below that $1.5 million thresholds, it suggests that their idea didn’t garner enough interest from investors, and it’s probably not worth our time, or that it’s a good idea, but one that needs more funding to succeed,” says Wu, the managing partner of Hone Capital.¹ Blossom Capital, a pan-European series A investor, uses AI not only to make faster investment decisions but also to overcome geographical barriers. One of their fundamental beliefs is that great teams and companies can also be found in peripheral areas outside the traditional startup-hubs. You can not only find great teams there but it is also easier to get into these deals as you have less competition from other investors. Rather than only taking widely available metrics such as website traffic, team growth and media mentions and building models around them, Blossom Capital tries to replicate its own decision procedure and mindset. It, therefore, supervises its model and includes common metrics about the team, market and product.² 2. Improving the due diligence On average, it takes about 30 to 60 days to complete a due diligence process and the selection is often biased. Similar to sourcing new startups, the due diligence is highly inefficient and consumes a lot of time and resources. With the help of “Wendal”, a big data analytics platform, Connetic Venture is able to perform a first due diligence in only 8 minutes. Wendal deselects 92% of the companies that apply and only forwards the best fitting 8% to a second round. According to Connetic, the success rate is much higher when they include Wendal in their decision making. Taking the Series A as a success indicator, the overall rate of companies that raise a series A round increases from 23% to 38% with Wendal. At the same time, the rate of bankruptcy could be reduced from 16% to 8%. This means that Wendal is almost 2 times better at selecting winners and avoiding losers than solely relying on human judgement. “We are using data and technology where the cost of human capital is very high — Sourcing and Screening. We also believe the Gut + Data will always beat Gut alone.” states Connetic Venture.³

Overview of AI use cases

3. Facilitating a startup’s development AI in Venture Capital is not limited to finding new investments but can also be used to make existing investments more successful. Scale Venture Partners, for example, is one of the funds which use their own knowledge about data-driven technologies to help their founders. They launched the platform “Scale Studio”, which gives startups performance benchmarks in order to evaluate their own business. The benchmarks are based on 20 years of data across more than 300 companies and include information about revenue growth, sales efficiency, revenue churn, and cash burn.⁴ NFX has a different perspective than other VCs. Most of the time and except for the hot deals, VCs choose between different startups and startups have little margin to negotiate. In order to resolve this unilateral pitches and selection, NFX built a matching platform that allows startups to create a list of investors who are best suited to their companies’ needs. More importantly, with the help of “SIGNAL” — a tool which was developed by NFX and utilizes social graphs — startups can find the best intro path to their chosen investors. NFX moreover believes that the success of a startup is highly driven by the quality of their network and social resources. As a result, SIGNAL helps startups to efficiently use and expand their social network to attract investors, customers and employees.⁵

4. Building successful ecosystems More and more venture capitals and accelerators try to implement a “community-as-a-service” model, where they build a helpful ecosystem around their startups. “There are a lot of inefficiencies that happen in Europe because we aren’ a unified country. For example, the best start-ups are in London, Paris and Berlin, but the best engineers are often from Bulgaria or Estonia“ says Andre de Haes of Backed VC.⁶ He believes that the ecosystem around a startup is one of the biggest variables that influence the success of a startup and is often overlooked by VC funds. Backed VC, therefore, built a platform through which entrepreneurs can interact with each other digitally and for example, share promising job candidates who were great but did not fit the industry or the respective company culture. Moreover, Backed VC has established an extended community called “SCOUTS” where they bring together high-impact humans. Some of them are founders, others are investors, engineers, or designers. Backed VC supports their personal development and helps them expand their networks. In return, these talents refer qualified candidates for the vacancies in their portfolios and assist Backed in selecting valuable investments.⁷

5. Predicting future founders Building a strong relationship with founders as early as possible is often key to participate in their round. Entrepreneur First (EF) claims to have built the world’s largest and most comprehensive dataset of great founders before they become founders. Every year they use their AI model to spot and select 100 potential founders out of 1,000 of applicants. EF then matches fitting founders and works as an accelerator for the newly formed teams. Robert Stojnic who participated in the project reflects on his experience: “I was a software developer with a technical Ph.D., now I’m the CEO of atlas ML. Without EF, I wouldn’t have met my co-founder and our company wouldn’t exist”.⁸ We believe that more and more positive impact will be brought to the industry through new machine-learning technologies. If you are interested in knowing more about AI in Venture Capital, feel free to reach out to us.

About CDF Technologies CDF Technologies leverages the benefits of machine learning technology to help startup investors focus on the deal that matters. Their Pitchdeck Parser enables VC funds to extract data out of pitchdecks thereby allowing to build their own intelligence and data-driven investment models. You can contact them at

¹ ² ³

27 views0 comments