A study published in May articulated the findings of a Turing Test applied to three different artificial intelligence platforms ( https://www.livescience.com/technology/artificial-intelligence/gpt-4-has-passed-the-turing-test-researchers-claim). For those not aware of what a Turing Test is, a Turing Test, named after British Mathematician and Computer Scientist, Alan Turing (https://www.britannica.com/biography/Alan-Turing), is a test where a human engages with another party and is asked to assess whether the party is human or computer. The computer passes the Turing Test, if it can be perceived as a human interactive partner. The recent test resulted in GPT-4, the artificial intelligence that powers ChatGPT, convincing a human participant 54% of the time it was another human.  Interestingly, a human engaging with another human only thought the other party was human 67% of the time.  This leads to obvious follow up questions related to whether the results were stacked to be close by not having parties with strong knowledge, experience, and communication skills representing the human participant in the study.  Nell Watson (https://www.nellwatson.com/), an AI researcher at the Institute of Electrical and Electronics Engineers (IEEE), commented about the study that “Raw intellect only goes so far. What really matters is being sufficiently intelligent to understand a situation, the skills of others and to have the empathy to plug those elements together.”  I could not agree more with Ms. Watson relating to this assessment.

As a 30+ year M&A professional, I have long recognized the value of research and knowledge in the facilitation of transactions involving the sale and purchase of privately held companies.  The best deals are made when parties negotiate in an environment where a commonly held set of facts exists, empathy is present for the other side, and options are available to overcome differences in perspectives.

The following are three areas where artificial intelligence falls short today and likely for a long time in performance when compared with a knowledgeable, experience, highly skilled mergers and acquisitions intermediary:

Business Valuation

For as long as entrepreneurs have been selling businesses, people have been looking for “rules of thumb” and quantitative methodologies for pricing privately held companies.  I have seen models that take a percentage of gross sales to value businesses.  X Revenues times Y Percent = Company Value.  A computer could do this calculation in milliseconds.  The problem is this approach ignores labor and occupancy rates in a specific area (Pittsburgh vs. Seattle, Portland vs. Manhattan, etc.), the contribution of ownership in terms of vision, productivity, and time, plus many other factors.  How about another common approach EBITDA times a multiple?  The question is how is the multiple calibrated. Does it take into account competition, reoccurring customers, tax rates in an area, motivation behind the desire to sell?   IBA calibrates its multiples to two places to the right of the decimal point.  The difference in value of a company with $2 million in EBITDA multiplied by 4, 4.33, and 4.67 is significant.  A multiple of 5 applied in this situation employs a blunt instrument rather than a precision scalpel.  The only way to properly price a business is to have a M&A intermediary involved who is actively engaged in facilitating transactions in the relevant industry, geographic area, and of the approximate size of the company being valued.  If you had a respiratory issue impacting your breathing would you have a podiatrist, diagnose and put a treatment plan together?   If you were charting a navigation course for a boat to travel throughout the San Juan Islands, would you seek assistance from someone who had traveled those waters numerous times or someone familiar with the Florida Keys.  There is no substitution for knowledge & experience in the relevant industry, geographic area, and deal size in the valuation of a business.  It also should be noted that comparable sale data is incomplete related to business valuation as the price & terms of most business sales are not reported in the public domain and for those reported, the specifics of the situation are commonly unknown (e.g., Did the seller have cancer and need to sell yesterday or was the location or intellectual property of a company a “must have” for a strategic acquirer) outside of the firm that did the deal.  This lack of quality data impacts the ability to produce computer analytics with a high degree of accuracy.   It should be noted that IBA’s 4300+ successfully completed transactions are not generally available for reference by other firms or the industry.  This type of firm held proprietary data set is the standard and not the exception in business brokerage.

Negotiations

Getting parties to “YES” in M&A transactions requires knowledge, experience, and skill.  Agreement on price is just the starting point.  Next financial terms like escrow holdbacks, seller financing, earn-outs, salary compensation, dividends, and stock options need to be negotiated to the parties’ mutual satisfaction.  Playing a computer in chess, the computer will default to standard responses to a series of moves. This type of methodology works in situations where the motivation is straight forward (Win the chess game) and will against all, but the strongest chess players result in at its maximum level, computer victories.  This methodology of providing response scenarios does not work in negotiating business sales, as humans often take illogical positions (e.g., a seller may prefer to finance a larger portion of a transaction rather than receive cash for tax reasons or a buyer may push for more seller financing to have a right of off set to address potential trailing liabilities.)  The reality of business sale negotiations is that a neutral party needs to be able to listen to each side’s motivations and fears and come up with solutions to surmount them.  It is hard to do this without knowing the complete landscape for the deal and the perspectives of the parties involved in the negotiation.  To quote from the musical, Hamilton, the following is what a knowledgeable, experienced, highly skilled business broker brings to a deal that a computer cannot in getting a buyer & seller to shake hands and reach agreement, “No one really knows how the parties get to YES. The pieces that are sacrificed in every game of chess.  We just assume that it happens, but no one else is in the room where it happens (https://youtu.be/_vIjtHjoXhE?si=c-KvspKGXSISB_vh)”.

Deal Facilitation

There is an old expression, it is not what you know, but who you know.  A computer can pull up a list of lenders who finance business acquisitions, but is information available on who finances deals in a specific industry in a specific geographic area.  Does the list also convey which bank is aggressively seeking to finance transactions and which one has put their loan activity on hold because they are in the middle of corporate merger with another bank?  M&A intermediaries frequently have this second set of information. They know where the reefs are submerged under the water approaching the harbor and where the channel is that is safe to navigate.  Business brokers also have relationships built through shared experiences with attorneys, accountants, and other professionals.  People who will see each other again often play nicer in the “sandbox” than ones who will never pass ways again.  It is not uncommon for buyers & sellers to “horse trade” and swap unrelated concessions on a pathway to deal completion.  An extra percent or two on a seller promissory note in terms of interest rate can be traded for an extra year in the non-competition agreement given to a buyer by the seller. This is not a linear, but multi-dimensional negotiation.  An activity much easier to achieve for a M&A intermediary who has listened to the parties than a computer that may not have processed the significance of a sidebar conversation weeks before in finding a way to get the buyer & seller out of the mud and back on track toward transactional completion.

Artificial intelligence is not going away.  It is going to enhance productivity in a spectrum of industries including business brokerage.    However, in a profession that sells individually unique products in deals between parties who have never been involved in a high stakes agreement before that will impact both of their lives significantly, we are far from the day when a computer can replace the knowledge, experience, and skill of a top tier intermediary.

IBA has been successfully selling privately held companies and family businesses since 1975 in the Pacific Northwest.  As a firm, IBA has gotten buyers & sellers to “YES” over 4300 times.  We welcome the opportunity to learn about your business, exit strategy objectives, and provide information and strategies to help you achieve your goals.  We will only engage on a project, if we believe we can successfully complete it.  IBA charges no fees until you have been paid and the business is sold.  All conversations with IBA are held in strict confidence.  Please contact us if you would like to schedule a time to meet at your business or any of our seven offices in Washington & Oregon. Name a computer that can travel to you for a meeting, shake your hand, and offer a friendly smile.

IBA, the Pacific Northwest’s premier business brokerage firm since 1975, is available as an information resource to the media, business brokerage, mergers & acquisitions, real estate, legal, accounting, banking, and wealth management communities on subjects relevant to the purchase & sale of privately held companies and family businesses.  IBA is recognized as one of the best business brokerage firms in the nation based on its long track record of successfully negotiating “win-win” business sale transactions in environments of full disclosure employing “best practices”.