Help: PitchBook’s research process

Would you like to know how we can get our world-class private financial market data? None of it would be possible without our advanced research processes. We filter, collect, and sort the most accurate data by combining human intelligence with cutting-edge technology. This article provides an overview of the processes we use to deliver the best private market data available.

PitchBook’s research process combines human intelligence with advanced technology to filter, collect, and sort the most accurate data on financial markets. At the highest level, all entities* follow a deliberate research flow. We discover a source that mentions the entity, confirm the entity meets our tracking scope, and then research it further to create a profile. Next, we continue to revise and update the profile for as long as the entity remains in our tracking scope.

We complete rigorous quality assurance at every step to ensure the data meets our standards. Our data managers meet regularly to discuss ways to expand our tracking scope, sourcing processes, and other aspects of the research process so we can provide you with the highest-quality data. Teams continuously review profiles and supporting data to ensure we provide the most accurate and timely information. Lastly, our Data Operations team and Product teams regularly collaborate to improve our data collection tools with machine learning and natural language processing.

Our research processes can be categorized into two main types: secondary and primary research.

• Secondary research – Involves gathering information from publicly available sources such as news articles and press releases.
• Primary research (also called “Survey”) – Collects additional information through direct calls and emails to people involved with entities.

In addition to the research process described here, some debt data points are gathered through the research process developed by LCD. Similar to the PitchBook research process, this process also involves a combination of information gathered from public sources and data gathered from individuals.

*Entities are the types of organizations we track. PitchBook tracks the following entity types: companies, investors, limited partners, and service providers. We also track people in management positions as entities, as well as funds.

Once the data operations team sources new data and verifies it using the process described in this article, the information is published onto the platform. The publishing cycle works continuously to ensure the most up-to-date data is always available on the platform. Once a cycle completes publishing new updates, another cycle immediately starts. The length of time each cycle takes depends on how many updates there are. There are six publishing cycles per day.

PitchBook uses a variety of sources to ensure accurate, up-to-date information reaches our platform:

• Web crawlers – We utilize over 7,000,000 web crawlers to capture information about companies, investors, funds, and transactions from regulatory filings, news articles, press releases, websites, and other public sources.
• Natural language processing and machine learning – Our technology utilizes natural language processing and machine learning to distill the massive amount of unstructured data, extract meaningful information, and integrate it back into our system.
• Secondary sources – Our specialized data teams verify additional information, such as public data, equity pricing, company valuations, revenue, IRRs, and more, from secondary sources to produce the most accurate, comprehensive picture of the company, person, fund, or investment in question. Secondary sources are those authored by someone not directly involved in the deal or fund. These sources are publicly available and found through targeted online searches.
  
  • News – PitchBook’s largest source of information is the news. We have two methods for collecting information from the news. The first is a team of researchers who evaluate articles from a curated list of news outlets. This curated list includes major news outlets focusing on the private market space like BusinessWire or PR NewsWire. The second news team evaluates all other news outlets for new information. They do this with the help of machine learning, which searches for articles and keywords that mention an entity we track and/or private market data that interests us.
  • Other online sources – Both teams pass along the articles they’ve found to other secondary teams, who will look for more sources, including press releases, filings (like earnings reports and Form Ds), and company website information.
• Quality assurance – Our team uses preventive validations, corrective validations, and mutual reviews to relentlessly vet every piece of data.
• Primary sources – Our primary research and survey team communicates directly with the companies, advisors, investors, lawyers, accountants, lenders, and others involved in deals to cross-validate the data we’ve collected and gather details that are not publicly available. Each quarter, the team reaches out to a curated list of contacts for information about their recent deals and funds. When you reach out to have your profile updated, you contact this team.
  
  • Some information is provided to our Data Ops team by participants with the goal of further completing their organization’s PitchBook profiles.
  • Other information is provided to comply with FOIA (Freedom of Information Act) requests in the US and similar laws internationally.

The following image shows a visual representation of the process.

If you believe there is an error in our data or you would like to see specific data added, there are two tools you can use to submit your feedback directly to our Data Operations team.

• Request Research – Use this tool if you would like to see a deal or a new profile (such as for a company or an investor) added to the platform.
• Data Feedback – Use this tool if there is existing information in the platform that needs to be updated or corrected. Note: If the information was found through reliable public sources, we are unlikely to remove the information. However, we will remove any information collected through primary sources that are found to be incorrect, unsupported, or have been asked to be removed. We also remove information in compliance with GDPR and similar laws.

For guided instructions on how to use these tools to your benefit, check out our article How to request research, submit data feedback, or update your profile.

Learn more about our data feedback and profile review tools.

• How to request research, submit data feedback, and update your profile
• Update your PitchBook profile

To understand how this feature fits into a broader workflow, check out PitchBook Pioneer’s course – PitchBook 101.