Paper: arXiv 2309.15552

Authors: Mark Potanin, Andrey Chertok, Konstantin Zorin, Cyril Shtabtsovsky

Abstract

Predicting startup success presents a formidable challenge due to the inherently volatile landscape of the entrepreneurial ecosystem. The advent of extensive databases like Crunchbase jointly with available open data enables the application of machine learning and artificial intelligence for more accurate predictive analytics. This paper focuses on startups at their Series B and Series C investment stages, aiming to predict key success milestones such as achieving an Initial Public Offering (IPO), attaining unicorn status, or executing a successful Merger and Acquisition (M&A). We introduce novel deep learning model for predicting startup success, integrating a variety of factors such as funding metrics, founder features, industry category. A distinctive feature of our research is the use of a comprehensive backtesting algorithm designed to simulate the venture capital investment process. This simulation allows for a robust evaluation of our model’s performance against historical data, providing actionable insights into its practical utility in real-world investment contexts. Evaluating our model on Crunchbase’s, we achieved a 14 times capital growth and successfully identified on B round high-potential startups including Revolut, DigitalOcean, Klarna, Github and others. Our empirical findings illuminate the importance of incorporating diverse feature sets in enhancing the model’s predictive accuracy. In summary, our work demonstrates the considerable promise of deep learning models and alternative unstructured data in predicting startup success and sets the stage for future advancements in this research area.

Complexity vs Empirical Score

  • Math Complexity: 5.0/10
  • Empirical Rigor: 6.0/10 (capped from 8.0: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 8.0)

Why this score: The paper employs deep learning models and backtesting against real Crunchbase data with specific capital growth metrics, demonstrating strong empirical rigor, while the mathematical complexity is moderate, focusing on applied machine learning without advanced theoretical derivations.

Research Flowchart

  flowchart TD
  A["Research Goal: Predict Startup Success<br>Series B/C"] --> B["Data Acquisition & Preprocessing<br>CrunchBase Data"]
  B --> C["Feature Engineering<br>Funding, Founder, Industry"]
  C --> D["Model Development<br>Novel Deep Learning"]
  D --> E["Backtesting Simulation<br>VC Portfolio Process"]
  E --> F{"Evaluation"}
  F --> G["Key Findings & Outcomes"]
  
  G --> G1["14x Capital Growth"]
  G --> G2["Identified High-Potential Startups<br>Revolut, GitHub, Klarna, etc."]
  G --> G3["Validated Importance of<br>Diverse Feature Sets"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a novel deep learning model for predicting startup success and a comprehensive backtesting algorithm to simulate venture capital investment.

Source sentence
We introduce novel deep learning model for predicting startup success, integrating a variety of factors such as funding metrics, founder features, industry category.

Assumes 3

  • A company is considered successful if it achieves IPO, acquisition, or unicorn status (valuation > $1B). data
    Source sentence
    In this research, a company is deemed successful if it achieves one of three outcomes: Initial Public Offering (IPO), Acquisition (ACQ), or Unicorn status (UNIC), the latter being defined as a valuation exceeding $1 billion.
  • Only investment rounds occurring after round B are included in the analysis. execution
    Source sentence
    This research is focused on investment rounds occurring after round B.
  • Minimizing false negatives is more important than minimizing false positives to avoid missing successful companies. statistical
    Source sentence
    Recall is of particular interest to us. The minimization of False Negatives (FN) holds greater significance than that of False Positives (FP) in order to circumvent the omission of successful companies.

Data

  • daily Crunchbase database export · Crunchbase alternative public
    Startups at Series B and Series C stages
    Source sentence
    We used daily Crunchbase database export (Daily CSV Export) as the primary data source, which is also supported by a well-documented API.

Recognised sources: Twitter / X

Costs

Unclear - sources disagree The platform classifier and the paper-text extraction give different answers, so no value is asserted.

Platform classifier: Not modeled

Paper text extraction: Not applicable

Finds 1

  • The backtested model achieved a 14× capital growth over the simulation period. performance
    Source sentence
    Evaluating our model on Crunchbase’s, we achieved a 14 times capital growth and successfully identified on B round high-potential startups including Revolut, DigitalOcean, Klarna, Github and others.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 4 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.