Discover the evolution of data-driven initiatives and AI in the VC industry. From early exploration to mainstream adoption, learn why VCs should embrace data for efficiency and uncover untapped opportunities.
SwissCognitive Guest Blogger: Penny Schiffer – “The Power of Data, Unleashed: Transforming the VC Industry with AI”
In the era of technological transformation, few areas of business remain untouched by the powerful ripple effects of digitization. At the heart of this wave are artificial intelligence (AI) and data-driven decision-making, which are rapidly transforming an array of industries, including venture capital (VC). To delve into this fascinating world and help us unravel its complexities, we asked one of the preeminent voices in the field — Penny Schiffer.
As the CEO and Co-Founder of Raized.ai, a Startup Coach, an Angel Investor, and a Global Thought Leader in AI in the VC industry, Penny embodies the intersection of technology and investment strategy.
Join us as we venture into this exciting and complex terrain, guided by her seasoned expertise.
How far has the VC industry come in terms of data-driven initiatives and AI?
We observe three phases in the industry:
- Up until 2019: some VCs were exploring how to source and evaluate startups with data, mainly large VC firms (like a16z or EQT) or very specialized smaller VC funds.
- 2019-2022: More VCs realize that data and information it contains are key and either start their own activities or look for helpful tools. Yet, this proves to be more difficult than initially thought and added value is sometimes hard to prove.
Most data pipelines struggle to deliver a favorable signal to noise ratio at manageable cost, allow for fast innovation without having to generate massive training datasets and complex, hard to maintain machine learning models.
Thank you for reading this post, don't forget to subscribe to our AI NAVIGATOR!
Sidenote: It was more work for us at Raized.ai than we initially thought, too. 😉
- 2023 and beyond: Using data has become mainstream, now more complex AI capabilities become relevant – and are finally feasible as large language models and autonomous agent technology become available.
Pure sourcing and enriching of startup data is not enough as we can soon facilitate use cases such as:
- Ad hoc market analysis for any given industry / startup that a VC sees
- AI enabled inbound: Dynamic application forms and ways to generate deal memos before meeting a startup in person
- AI enabled outbound: Autonomous identification of relevant startup companies, personalized outreach and assessment if a company should be taken to due diligence based on the information gathered
- Advanced company analytics taken the founders, funding and financials of a startup company into account
What are the reasons for VCs to embrace a data-driven approach?
There are mainly two reasons:
VCs can become more efficient (important in times of budgets that are harder to get) and they can increase their chances of not overlooking outlier opportunities.
Those VCs who play their narrow niche of expertise successfully and neither look for more or for better business may stay in their current state for another few years, i.e. until competition walks in or their niche transforms.
For all other VC’s and/or investors data make positive differences:
- Get a more complete overview over investment opportunities out there not to be missed
- Be more efficient, i.e. do more right investment decisions per month
- Over time improve your business and decision processes, as experiences from data driven decisions add to your personal and company knowledge.
What is your viewpoint on the buy versus build dilemma? Is it a choice between the two options or a combination of both?
Let me tell you a story: A couple months ago we had a call with a tech savvy associate from a large German VC fund. He was proud to tell me that they are now building their own startup discovery pipeline and that he will be live with it in October. Having gone through this myself I was almost envying him for his unrestricted optimism. He had no clue about the difficulty of getting hard to get web data at scale, the often boring work of creating and maintaining a data scheme and what happens if you have no proper orchestration and monitoring of your pipeline.
Last week, we had a chat again and discussed how he could leverage our data and add value to it for his fund.
We have seen that it consumes much more resources than initially thought to build and maintain a solid data pipeline that can identify and screen startup companies. And to make it into an engine that is really giving a VC firm a competitive advantage one will need to spend another hundreds of thousands of USD.
However, a standard system or generic database is not able to add value per se, so it will need customization and thus tech skills inhouse. This is probably only feasible for funds that have deeper pockets and/or specialize into an AI sourcing approach.
In short, our proposition to the VC industry simply is: Let’s do the core once and do this right. Then let everybody profit via tailored and personalized services. With Raized.ai we try to build customizable solution on a standardized platform.
“If you want to be fast, go alone. If you want to go far, go together.”
Do you know any company programming their own spreadsheet calculation program? Yet, it looks like very few VCs have a clear strategy on this, get interns and tech students working on half-hearted data-driven projects that lack critical mass and top management support to lead to data pipelines of even AI systems that can add real value to the VC firm.
When developing internal tools, which aspects of the value chain hold the highest importance for you?
We’re convinced an A to Z vertical solution is key, conditio sine qua non. It interconnects all relevant elements turning internet data into meaningful business action.
For relevant investment opportunity information this covers: Data sourcing – aggregation – enrichment – interpretation of content through algorithms including AI, personalized and configured presentation to client.
Leave one out and the vertical chain is broken.
Disregard the interplay between business and data sense and you just have another IT project.
What are the primary obstacles or limitations you face when considering data-driven initiatives?
On the technical side it’s keeping up with the new platforms and technologies, see e.g. availability of ChatGPT, relevant to our solution while simultaneously keeping up an operative, stable and running business. Continuous innovation and maintenance is key and more often than not underestimated upfront.
On the business side, adopting and embracing the additional horsepower which data-driven solutions bring to the table profits of clear commitment and leadership. Sparking enthusiasm for melting business knowledge with high tech is more than justified and key to do.
With the new way of developing models with large language models the complexity to integrate business knowledge with technical expertise rises even further: With prompt engineering, data scientists have little possibility to test the quality of what they are building as there are no large validation data sets required and they oftentimes need a subject matter expert to judge the quality of the system response.
What advice would you give to a VC firm seeking to adopt a more data-driven approach?
Be aware and enjoy that you embark more on a journey than reaching a destination. Data-driven approaches initially adapt to known business needs and processes. Simultaneously they allow for more efficiency / quality and may foster improvements or simplifications in business processes. It’s this mutual benefit which creates a cascade of improvement steps.
The more you are aware how you work today and why, the faster you can connect business and data.
About the Author:
Penny Schiffer is the CEO & Cofounder of Raized.ai, an AI engine for the VC industry. Named a global thought leader by datadrivenvc.io and a top 10 European female investor by Insider, Penny is an ex-VC turned entrepreneur with a background in data science. With experience investing in 30 startups and sourcing thousands of investment opportunities, she believes the industry is ready for disruption.