Revolutionizing M&A deal sourcing with AI

4 min read

Revolutionizing M&A deal sourcing with AI

Did you know that AI will forever change the M&A process? Research shows that AI…

know more

Did you know that AI will forever change the M&A process?

Research shows that AI will significantly reduce the time spent on sourcing by 80% and due diligence to less than a month by 2025, a significant improvement from three to six months in 2020. Evidently, the adoption of technology, that structures data, is becoming increasingly important for businesses seeking to streamline their M&A processes. The acceleration of digital transformation is no longer solely focused on achieving a competitive edge. In today’s landscape, it has also become a factor in building business resilience.

delphai is disrupting the B2B deal sourcing market by providing businesses with access to an extensive pool of potential targets and empowering them to make informed decisions based on firmographic data.

This blog features insights from an interview between Nikolaus Grefe, an experienced professional in the field of data science and currently serving as head of Strategy & Business Development at delphai, and Sebastian Zang, an experienced software industry manager who has held executive positions at major consulting and software firms. The interview is featured in the article titled “Digitalization of M&A: Interview with Nikolaus Grefe on Deal Sourcing with delphai”.

Machine-Readable Data: Scalable and Efficient

How can businesses be sure they have access to reliable and up-to-date data to enable their M&A sourcing processes?

The answer is simple: The usage of machine-readable data. Machine-readable data is a form of information that can be processed by a computer without the need for human intervention. This absence of human intervention allows for faster and more accurate processing and analysis of data. This reduces the possibility of human errors or biases, which can affect the quality of data analysis. In contrast, manual data entry involves the laborious process of individuals inputting information into a system by hand. 

The latter method is time-consuming, prone to errors, and limited in its scalability, as it can only cover a limited amount of industries and geographies. With machine-readable data, businesses can obtain more comprehensive insights, perform deeper analysis, and make better-informed decisions. Despite this, most other sourcing platforms still rely on manual inputs.

Nikolaus: Essentially, our approach differs in that we rely completely on the machine generation of data. delphai is thus able to process much larger amounts of data, especially extensive text data. Our market competitors continue to rely on manual data entry. In practice, this means that companies such as Dun & Bradstreet or Moody’s request employee and sales figures from the companies themselves by phone and e-mail. Everyone is aware that this is not a scalable process. 

Machine-readable data is a powerful tool that can help businesses enhance their decision-making processes and gain a competitive edge in their industries. In the current market environment, deal success is dependent on not only volume but also speed. Having the right data at the right time is crucial for securing deals before others take advantage of the opportunity.

A common method to obtain machine-readable data is to use web scraping tools that can automatically extract data from websites and store it in a machine-readable format. Another approach is to use APIs that allow businesses to access data from various sources in real-time.

Nikolaus: delphai integrates both public data sources such as company websites, press releases, job advertisements, conference participations, and the like, as well as data from strategic partners from these, we extract and generate our own machine-readable data, such as customer-supplier relationships, acquisitions, or corporate financial data. This diversity of sources and data enables delphai to create an accurate fingerprint of a company. Our search engine is thus able to find more and better companies within seconds than any other software.

The challenges of AI-based evaluation of information

The use of AI to evaluate information is becoming increasingly popular but it also presents several challenges.

One of the primary challenges is the quality and reliability of the data used to train the AI algorithms. The accuracy of the results depends heavily on the quality of the data input. The quality of data dictates the accuracy, completeness, consistency, and reliability of the output. It is the foundation for making informed decisions and ensuring successful outcomes in any data-driven process.

If data quality is not ensured, it can lead to incorrect or unreliable information, which can result in poor decision-making including missing opportunities, incorrect valuation of deals, and damaged reputation due to failed deals or investments.

However, there appear to be multiple solutions to this challenge.

Nikolaus: First and foremost, delphai controls its sources. We also provide provenance for each data point so that users can check whether they meet their own requirements. We are talking about data provenance or data observability, a central pillar of our product promise. Furthermore, we are developing AI methods that assist in assessing the credibility of data. Since this capability is not only central for our users, we are supported by a research grant of one million euros from the German Federal Ministry of Education and Research to develop methods for determining the accuracy and authenticity of textual information.

However, as competitors still heavily rely on manually collected data, provenance and the ability to assess credibility are heavily compromised, which can lead to inaccuracies and inconsistencies. It does not just end here. Another typical obstacle in the AI-based evaluation of information is the challenge of relevance. 

When evaluating information, it’s essential to differentiate between relevant and irrelevant data, as irrelevant data can skew the analysis and lead to inaccurate decisions. In the case of company websites, it can be challenging to distinguish between information that is directly related to the company’s services versus information about customer success stories or other unrelated topics. 

AI-based systems need to be designed to accurately identify and extract relevant data, while filtering out irrelevant information, to be accurate and useful.

Nikolaus: It is indeed an important part of our tech stack to check content in a text for relevance before linking it to a company profile. However, the problem starts even earlier. First, companies need to be identified in a source. The word “Apple” or the name “Miele” is not enough to properly link a news article to a company profile, for example, because of their ambivalence to the fruit or the name of the company’s heir. For this purpose, we have developed a series of NLP procedures that incorporate context to ensure that it really is a company.

The future of AI-based evaluation data

The future of AI data analysis seems bright as technology continues to improve and expand. With available data growing by 40% every year, AI-powered tools like delphai can provide valuable insights and help businesses make more informed decisions. 

The rapid pace of development in the field of AI means that there is a constant need to keep up with the latest research and channels to extract information more effectively.

Nikolaus: Through the further development of NLP technologies, we see new possibilities for collecting and structuring company data. Transcripts of earnings calls and podcasts are just the beginning. Shipping information from ports or satellite images of inventory could become just as much a part of our central hub.

At delphai, we believe that the future lies in low-code/no-code-based systems tailored to each company’s unique needs across various business units and functional areas. We don’t see a future in end-to-end vertical software applications as they create silos that obstruct collaboration and cooperation. 

Microsoft Power BI-based MI/BI dashboards are excellent examples of such systems, which can help build teams that work seamlessly across these boundaries. Hence, we don’t believe that single-use case, end-to-end vertical software has a future. 

For delphai this means:

To become the number one destination for firmographics and that our software is used by hundreds of thousands of customers beyond the M&A industry.

Nikolaus Grefe

Curious to learn more? Book your demo here