Why We Will See More Use of Data and Analytics in Venture Capital

The past three to five years have seen an increase in  the use of data and analytics in venture capital. The business of venture capital is based on capitalising on disruption and innovation. While venture capital firms are skilled at identifying entrepreneurs and businesses at the forefront of changing trends, the venture capital industry as a whole has seen significant innovation and change in recent years.Data and Analytics in Venture Capital

When you talk to people who work in venture capital, you may hear them mention having a good “gut feel” – the ability to make wise financial decisions based on primarily qualitative information, as well as quantitative data provided by the technology business.

Can we, on the other hand, still respond sufficiently to the inner voice approach in 2022? With the significant rise in data analysis technologies, greater access to large and small data sets, and a more demanding and competitive venture capital environment, the question is if gut feel is still sufficient in these circumstances.

By 2025, the typical pitch experience will have changed substantially, and founders will most likely have to deal with investors that are better informed since they have real-time data at their disposal.

Over time, the venture model has undergone many iterations and transformations that have influenced how businesses operate. More recently, big data, artificial intelligence, and machine learning trends, which venture firms have helped catalyse advancement in, have converged on venture investing itself. With more data on startups and entrepreneurs than ever before, sophisticated models can be built and applied to the venture investment process.

Increased use of data and analytics in venture capital

According to Gartner, venture capital can stimulate and promote innovation more than other traditional forms of investment and financing for the following reasons: First, as an effective equity financing method, venture capital focuses on long-term investment in early small and medium-sized enterprises, which can effectively reduce R&D underinvestment in enterprises with tight financing constraints, lower external financing costs, and boost innovation performance. Venture capital institutions have better screening and supervision capabilities than banks, and their valuation of businesses is more accurate.

As a result, the venture capital industry is undergoing a transformation. With data-driven processes increasingly being used in both the sourcing of investment opportunities as well as the decision making processes within VC firms . In the past, venture capital was based on carefully developed personal networks of general partners and VC brands themselves, together with relying on market pattern recognition and “gut-feel” to make investments. Through this approach VC firms would either attract investment opportunities through their reputation or from time to time be able to recognise where  opportunities existed.

Why data and analytics in venture capital

The new, data-driven model prioritises data analytics to inform the various stages of the venture capital investment process, such as sourcing, screening, selecting, and monitoring investments, as well as assisting portfolio companies post-investment.

Data Inclusion in Decision Making

A growing number of new venture capital firms rely on advanced data analytics processes together with accessing a wider variety of data sources on which to rely. The richer the volume and variety of data we have accesso to, the more useful the insight gained from analysis become.

While the intensity of data use varies, it appears that many tenured firms and the majority of new firms now have data-driven components to their strategies. The reason for this shift should come as no surprise. The competitive nature of venture capital investing, combined with the significant increase in news businesses starting up, are forcing VC firms to find new and different ways to secure the best deal flow for their investments.

An increasing amount of VC firms are now discovering the benefits of using data supported signals to help with their decision-making – essentially supercharging traditional venture capital models by leveraging the growth and availability of big and small data combined with the advancement of data analytics and machine learning techniques.

Other benefits of data and analytics in venture capital

For seed and early-stage businesses, where there aren’t usually any performance-related signals, the focus has been on proactively finding promising entrepreneurs and founding teams through looking at specific data signals. The  use  social media signals for instance, can help to identify  high-potential engineers and product managers at top technology companies who might want to leave and start their own businesses. In the future, these individuals can then be approached by venture capital firms before they launch their business.

Beyond the early and growth stages of business growth, team-related data is still important, but there are also other factors that needs to be considered. The industry, board members, and investor syndicates of a company all become factors, as do more quantitative inputs like deal history, valuation, and ownership structure. Growth metrics and other key performance indicators are also available for evaluation.

As businesses mature, factors that can be measured become more important and predictive models become more accurate and useful. The types of data that are available, as well as the amount and type of data, can have a big impact on how firms act and how useful scoring and predictive models are. Consumer-facing businesses, for example, usually have more data to work with. Venture capital firms can get specific information about how users grow, stay, and last. Furthermore, the key performance indicators for enterprise and software companies can be different based on their business model and vertical, which makes it more difficult to make comparisons between them.

Ultimately, the goal at the sharp end of the investment process is to better source and screen opportunities. This helps VC firms to target the entrepreneurs they want to approach more effectively. Data is increasingly changing how venture capital firms approach monitoring and assisting portfolio companies post-investment, in addition to sourcing and screening opportunities.

Monitoring existing portfolio companies is transitioning from superficial management level engagement between CEOs and venture capital firms to a deeper more insightful relationships. Venture capital firms increasingly have direct access to front-line data on what is happening in businesses. This data can be used to anticipate and proactively address company-level issues. Furthermore, direct access to information enables venture capital firms to share data signals across portfolio companies in a way that was previously not scalable.

Identifying the effectiveness of marketing spend in one company’s approach to a single vertical, for example, can be beneficial to all portfolio companies that address the same vertical. Data is also increasingly being used post-investment to help teams better understand growth drivers, uncover new opportunities, optimise resources, and identify potential weaknesses or risks.

Data and analytics in venture capital – a Data-Driven Processes

Today’s venture capital firms using data-driven tactics take a variety of approaches, displaying a wide range of sophistication. Despite this, there appears to be a convergence of common elements and techniques. Data remains the most important input, as in any other machine learning or artificial intelligence application. A large volume of high-quality, trustworthy, and well-organized data is ideal. Perfect data sets are often difficult to obtain for the majority of applications.

The majority of businesses use a combination of third-party and proprietary data. Data social media platforms such as Twitter, LinkedIn and Clubhouse, together with sources like Crunch-base, preqin.com, CB Insights and AngelList are often consulted. Third-party data organisation and manipulation can be time-consuming and labour-intensive.

Pooled third-party data that has been improved and arranged in a customised manner may eventually become proprietary. Many firms also combine third-party data with their own internal data, which is primarily derived from existing portfolio companies. Some businesses hire data collection specialists or outsource data collection to third parties.

After obtaining and organising a high-quality data set, machine learning techniques such as logical regression and deep learning models are used to gain insights from which we can benefit. In general, scoring functions are created to evaluate a model’s ability to successfully evaluate relevant metrics such as team, growth, momentum, funding, capital efficiency, and other important characteristics.

Many businesses create prediction models that can forecast variables such as future growth, valuation, and the likelihood of attracting additional funding. Predictable signals are used to score and prioritise companies prior to investment, and to flag potentially valuable insights post-investment. Before we can fully rely on analytics models, we do however need to ensure that we see the whole picture.

Back testing in venture capital is a method for testing investment strategies based on past data. If such a back test shows that an investment or investment approach worked in the past,  it gives the investment manager the confidence to do more of the same in the future.

Both data collection and analytics models will differ depending on a company’s investment strategy, stage, and sector. They are typically modified and updated on a continuous or as-needed basis, based on training data and testing results. Some firms take a strategy-driven approach, in which data and algorithms are compiled and built to support a specific investment strategy.

Developing Powerful Data-Driven Investment Capabilities

Building effective data-driven investment capabilities at a venture capital firm entails several steps. The first step is to recognise and prioritise the importance of data quality. The most difficult challenges revolve around creating good structured and unstructured data sets, as well as integrating proprietary data with publicly available data. Second, in order to effectively incorporate data-driven processes, venture capital firms must have the right talent and resources in place. This has already resulted in a noticeable shift in venture capital firms’ hiring strategies.Data and Analytics in Venture Capital

Importantly, it may necessitate a different talent sourcing model and organisational structure, with ramifications for compensation and incentive structure. Indeed, if data and algorithms account for a greater proportion of a venture capital firm’s competitive advantage, the traditional partnership model may not be the best organisational structure. Third, it is critical to effectively interpret and incorporate the outputs of a data-driven investment process. Because the approach is still in its early stages, venture capital firms understand that they cannot rely too heavily on early results. As a result, almost all companies that use data analytics combine it with traditional approaches to company and deal evaluation.

Data analytics is a tool in the investment process, not a replacement for it. Fourth, even venture capital firms that are farthest along in their adoption of data analytics recognise the importance of continuous improvement. Algorithms must be constantly tested against outputs to determine their efficacy, and data sets must be improved on a regular basis. As more venture capital firms join in on this trend, first-mover advantages are likely to dwindle over time. Firms that use these tools effectively must continue to invest time, talent, and resources in order to constantly improve them.

We will see a reduction in individual and cultural bias

Data analytics can help reduce the prevalence of unconscious bias in venture capital decision-making by screening and identifying promising new entrepreneurs based on quantifiable factors. Decision-making based on gut instinct, pattern recognition, and established networks has contributed to venture capital’s diversity issues over time. Data-driven decision models, which are designed to eliminate these biases, can be gender and race neutral.

There appears to be a clear ability for data-driven decision-making to reduce unconscious bias when approached conscientiously. Firms that use data analytics in their processes must be careful not to introduce biases in their models. If successfully implemented, these models may not only help increase diversity in venture firms’ portfolios, but they may also lead to better returns associated with diversity.

The venture capital industry has always been known for its dynamism, not only in terms of the innovation driven by portfolio companies, but also in terms of how the industry itself evolves. The recent emergence of data-driven decision-making reflects the industry’s ongoing evolution. If used correctly, these new tools have the potential to be extremely powerful. They will undoubtedly become more influential as algorithms improve, training improves through better feedback loops and more experience, and data, the key input, continues to improve and grow in volume over time. However, as venture capitalists and investors, we should be cautious of becoming too reliant on technology.

We need to remember that data is never perfect. Not all factors can be quantified, and not all externalities and their impact can be predicted. You only need to look at the knowledge management approaches still used by many organsiations today. Explicit knowledge, often recorded through quantitative means (numbers and figures) are far more straight forward to measure and collect as opposed to tacit knowledge, or experience.

How do you measure experience and someone’s ability to understand a market deeply or execute an effective plan? How do we measure the strength of someone’s network and influence amongst stakeholders? Yes we could create an algorithm to help us understand it better, but the accuracy of such measurements will seldom be accurate. Furthermore, the human element of venture capital investing and company formation will always be critical for long-term success.

As the venture capital model evolves, investors should not be surprised by what appear to be radical changes, such as allowing artificial intelligence based algorithms to vote on investment committees or the emergence of purely data-driven platforms. VC firms will need to adjust to the upcoming changes. The ability to conduct proper due diligence on new approaches and data-driven models, in particular, will become increasingly important.

We suspect that sophisticated VC firms will develop these skill sets proactively, rather than reactively, and will likely benefit in the long run. Caban Investments and our UK based Corporate finance consulting firm, Caban Capital continues to monitor these changes and selectively incorporate data-driven processes in its own investment processes as part of its progressive approach to building venture capital portfolios.

Leave a Reply