The role of data science in the target analysis process

The role of data science in the target analysis process

So far, the field of mergers and acquisitions (M&A) does not seem to have made the shift to digital. Yet machine learning and artificial intelligence could be used in several phases of the process for evaluating a target company.

Our analysis will focus on the role of data in mergers and acquisitions (M&A) and on how it is used in this highly specialised field. Data is present at all levels and serves a purpose at every stage of an M&A project. It proves to be a key element enabling the buyer company to make the best possible choices.

What avenues are there for improving the M&A process when weaknesses are encountered in the information collection process? Which phases could be vastly improved to save precious time in the valuation process? What solution using data science would give an M&A company an edge over its competitors?

With digitalisation and the rise in the quantity and quality of data available, the M&A process will have to evolve to incorporate analyses based on mass data collection and perform more comprehensive and precise valuations.

The role of data and artificial intelligence in M&A deals

While most industries have made the shift towards using machine learning techniques to improve their value creation capability, it seems the field of mergers and acquisition (M&A) has not made the move and is now lagging behind its counterparts.

To retain a competitive advantage when it comes to technological and organisational advances, companies specialising in M&A tend not to reveal information about their processes.

It is important to emphasise that acquisition processes are controlled by strict protocols at each stage of the work chain to ensure that the target company valuation is as accurate and comprehensive as possible.

Incorporating the data science tool called optical character recognition (OCR) would be a good solution for this identified problem. Indeed, OCR turns files containing images and printed or typeset text into usable text files.  Major progress is being made in the scientific community with regards to OCR which could reduce the administrative and repetitive processing time by 75%.

Data science and M&A

In M&A, data is currently still used in ways very similar to about ten years ago. The field has yet to get on the data science train. The tool most commonly used to process information is still Microsoft Excel, which of course remains a must in finance as it can be used to develop economic valuation models, for example. However, it offers insufficient automation, artificial intelligence and data visualisation tools, and it does not provide the ability to interface with the latest advances in machine learning and task automation.

Algorithms fall short when it comes to valuation

There are solutions available for making certain financial predictions about target companies for the purpose of M&A, but these software applications tend to fall short of the mark and do not yet give the companies that own them a clear advantage over their competitors. The first reason is the lack of configurable settings when inputting data into the software. This makes the forecasting too general and not precise enough. Moreover, associates who use these solutions have trouble trusting them due to the lack of information concerning the way they work and how the predictions are generated. Today, with these algorithms called “black boxes” it is not possible to explain to the superiors checking the valuation how the result was obtained, yet data explainability is a major focus of the General Data Protection Regulation (GDPR).

Finally, we have demonstrated that certain machine learning and artificial intelligence techniques could be applied to the field of mergers and acquisitions at several stages of the target company valuation process. From straightforward task automation to algorithmic prediction based on rich databases, numerous possibilities exist and are eagerly awaited by professionals in the field.

With the growing digitalisation of the different functions within companies and the rise in the quantity and quality of data available, M&A processes will have to evolve and massively integrate data analysis to achieve more comprehensive and precise valuations.

To read the full article you may consult, Revue Banque n°842 (27/02/2020, Le rôle de la data science dans le process d’analyse des cibles)

Dhafer SaidaneDhafer Saidane, Professor of Finance, FAIRR Research Centre, SKEMA Business School - University Côte d'Azur, France

All author's posts

Mathieu Da SilvaMathieu Da Silva, Inspecteur à l'Inspection Générale, La Banque Postale, Alumni SKEMA Business School

All author's posts

Close Menu