Machine learning and AI in dealmaking – The future of M&A?

In the M&A news cycle of major market trends, big acquisitions and shock administrations, the biggest changes, can sometimes get lost as they occur just below the surface. Behind the scenes of the global M&A market, major technological advances are permanently changing the way dealmaking can and will be carried out.

Key among these advances are artificial intelligence (AI) and machine learning - technologies that are increasingly being applied to dealmaking and prompting forecasts that they will drive and underpin fundamental changes in M&A.

As M&A processes advance into the 2020s, such technologies appear poised to deliver huge increases in both the speed and efficiency of dealmaking. Furthermore, this steady march forward has only been accelerated by COVID-19, which has prompted rapid tech and digital uptake, while forcing many dealmakers to adjust to operating remotely.

Reflecting this, in the recent Datasite report The New State of M&A: A Global Perspective, which surveyed more than 2,200 dealmakers from a range of industries and regions, 30 per cent of respondents named AI and machine learning as the technologies which would have the most transformative impact on M&A in the next five years.

For those that are perhaps new to the idea of AI and machine learning as a fundamental component of the M&A process, in this insight we’ll detail how they can be applied to dealmaking as well as the core advantages that utilising these technologies can deliver.

Technologies that can be used in M&A

Rule-based systems – Perhaps the most well-known form of artificial intelligence is the rule-based system. These are systems in which expert users set up rules and criteria which the system then uses to make decisions, classify information and data or retrieve specific information.


Machine learning – Machine learning begins with a generic algorithm, which then becomes more specific as it is fed with large volumes of data. There are three basic methods of feeding a machine learning algorithm:

Supervised – In which the data set used has been selected, classified and labelled by experts in the relevant field, eventually creating an algorithm which can automatically analyse and classify new data when applied to it.

Semi-supervised – The data set is comprised of some labelled data, typically with a larger amount of unlabelled data, from which the algorithm learns in order to make future predictions.

Unsupervised – The algorithm is trained using an unlabelled data set. Under this method, the generic algorithm will still be able to identify groupings of similar documents, but users will still need to manually classify these groupings afterwards.


Data Analytics – Once a machine learning algorithm has been programmed, it can automate many of the processes typically carried out during analysts in the M&A process. With machine learning, data analytics can massively improve the speed and scale at which analysis is carried out, providing granular insights into things such as a target company’s financials.

This can greatly increase an analyst’s capabilities and free up time for their own analysis to go deeper into areas such as a target company’s performance, sectoral trends or potential post-deal synergies.

What can AI/ML bring to the M&A process?

Due diligence
As anybody who has ever been involved in an M&A deal (whether on the buy side or the sell side) can attest, organising and preparing files ahead of the due diligence process is one of the most time-consuming elements of a transaction.

Utilising AI and machine learning solutions, dealmakers can benefit from algorithm-based systems that can learn from data and rapidly sift through huge amounts of content. An algorithm taught from a supervised or semi-supervised data set can read through, sort and upload files so that they are ready to be shared with the other party in a secure data room. In this way, machine learning algorithms can feasibly cut a process that often takes months down to mere minutes.

AI can similarly be used to cut down other time-consuming and menial tasks in the due diligence process, for example to automatically strike sensitive words, phrases or information from documents throughout the deal. Not only can such automation drastically save time, it can improve the efficiency of data protection, helping to ensure compliance with regulation such as GDPR and reducing the risk inherent in M&A.

From a due diligence perspective, the key benefit of AI and machine learning seems to clearly be the time-savings that can be delivered. This is reflected by the fact that 56 per cent of respondents to the Datasite survey said that technological advances such as AI would help to reduce the time required for due diligence processes from 3-6 months in 2020, to less than a month by 2025.

With quicker deals, buyers operating aggressive acquisition strategies can operate more efficiently, deploying their capital at greater speed and scale.

However, AI and machine learning’s due diligence applications aren’t strictly limited to making deals run quicker. With analytics powered by AI, buyers can more efficiently generate deep insights into a potential target company, through in-depth analysis of factors like cashflow, assets and the wider sector or market. In this way, AI and machine learning can contribute to the aspects of the M&A process that are generally seen as being most reliant on the “human touch”.

If you are considering target companies for acquisition, BSR has developed a research and reach-out facility to find suitable businesses and contact them on your behalf, in order to help you secure the perfect deal. Powered by smart technology and overlaid with manual research. Find out more here.

Making a move and negotiations
With AI-derived insights, buyers will be better equipped and more quickly able to flag up potential risks and make a decision as to whether a deal is worth pursuing or whether they should move on to other targets. Again, the increased efficiencies delivered by AI and machine learning could help rapidly speed up this process, saving money as well as time.

Of course, if a buyer does still decide to make a move, identifying these risks also provides valuable information to take to the negotiating table and a stronger hand in talks. AI-powered data insights could, in this way, enable buyers to negotiate better payment terms or even a lower overall valuation for the deal.

With the current M&A market still suffering from widespread valuation gaps as a result of the COVID-19 pandemic, gaining these kinds of insights in a timely manner could prove invaluable in ensuring that negotiations don’t get bogged down.

Post-deal integration and value creation
Insights generated through data analytics and AI can also improve a business’ post-acquisition integration and performance, both in terms of identifying issues that will need to be addressed upon taking ownership and by highlighting potential areas of synergy with an owner’s existing business or other points of value creation.

Overall, when it comes to negotiation and integration, these naturally remain processes that lean heavily on human interaction. Whether that’s setting forth different points of view during negotiations over a valuation, or building relationships with new employees following an acquisition, these are inherently human processes that can’t be automated.

What AI and other technologies can deliver in this regard, however, are insights that can augment, smooth out and improve these processes, increasing efficiency and delivering better results during and after an acquisition.

Drawbacks and solutions – How can businesses implement AI and machine learning?

Naturally with any new developments that occur in the world of M&A, the issue of actually implementing such technological advances can be problematic, particularly when it comes to widespread availability and adoption.

The main drawback, particularly for smaller and medium-sized UK firms currently suffering from a funding gap, is the cost element. This may mean that, for the foreseeable future, bigger or more well-capitalised firms will be among the only buyers to benefit from the advantages that AI and machine learning can bring to dealmaking.

Another downside to adopting these kinds of technologies in the M&A process is the time that can be required to implement them. Setting up an AI rule-based system, for instance, will require someone with expertise in both building a rule-based system and the topic the system is being built for, something that many firms will simply not have access to.

Similarly with machine learning, an effective algorithm for an M&A process would typically require supervised machine learning. This, of course, is something that requires extensive, time-consuming work to label the large, fully classified data set that would be used.

As with the cost associated with machine learning and AI, some businesses simply may not have the capacity or available time to expend on setting up these solutions. Likewise, they may not have the budget to outsource this process to a dedicated expert.

As an unsupervised learning process requires user interaction to classify data after the algorithm has sorted through it, this is also a time-consuming approach that many companies may not consider worthwhile. The obvious alternative is to proceed with a semi-supervised approach, which will save time, but perhaps offer less accuracy and less value.

One possible solution for smaller companies or those with more limited budgets is to utilise AI and machine learning solutions developed by third parties, such as due diligence tools which can be applied to a wide range of cases and tasks or AI-enabled virtual data rooms.

Such solutions could prove to be both accessible and worthwhile in terms of automating some of the most time-consuming parts of the M&A process. While some fine-tuning may still be required to tailor these solutions to individual business deals, they are the sorts of AI and machine learning applications that may be the first to be widely adopted in M&A.

These are tools that can be readily accessed by firms looking to harness AI and machine learning for their dealmaking here and now, providing they have the budget and inclination to do so.

Furthermore, AI and machine learning are emerging technologies that are advancing everyday. While their most valuable applications may currently only be realistic options for some dealmakers, as time goes on they will only become more sophisticated and more widespread.

Already, we are seeing business brokers and corporate finance boutiques utilising AI to help produce client reports and other content. A few are using companies like Business Data to produce industry-specific AI chatbots.

For now, as demonstrated by the Datasite survey, the key change that AI and machine learning will deliver to M&A in the immediate future is massively speeding up the pace at which due diligence is conducted. Given the scale at which this looks set to happen, (with forecasts that due diligence will take, on average, under a month by 2025) this alone will clearly be enough to make AI and machine learning revolutionary forces in M&A.

Another thing that’s clear, however, is that this is only the beginning of their impact on dealmaking.

Also read: AI in M&A: How can buyers and sellers harness the technology?

Share this article

Latest Businesses for Sale

---

---

---

View more businesses for sale