When business people face a decision, what if they behaved like scientists to get the best results? Can they count on more accurate predictions? Would they find it easier to focus on original ideas? Would they be more likely to develop a project beyond the initial stages?
In recent years, in both management practice and in academic debate, the fact is that companies are having to make decisions on new products or business ideas in conditions of growing uncertainty. This discourages them from committing sizeable resources ex-ante to more innovative business models or new product features, and instead encourages them to adopt flexible approaches based on market feedback, prototypes of ideas, staggered investments and adaptations to environmental changes. New theories on strategic management and economics have emerged on this subject, a classic approach that combines differing search heuristics (such as trial and error, effectuation, or confirmatory search), but this doesn’t seem to be enough to collect and process information that can inform entrepreneurial decisions in a constructive way.
Still today the academic literature and management practice struggle to determine whether methods actually exist that are effective for collecting and elaborating the information needed to support the decision-making process. To find the answer, we decided to explore this question from two perspectives by comparing two different approaches that can be used for decision making:
Our study, which recently appeared in Harvard Business Review and Forbes, empirically verifies the varying effects on performance that a company can achieve by applying a scientific approach to decision-making processes. We specifically looked at launching a new business model or product idea, compared the results of this approach to one based on heuristics, and then attempted to explain possible divergencies.
During our randomized control trial (RCT), we utilized a sample of 116 Italian startups and 16 data points over a year-long period.
We randomly sorted the companies into two groups, one treatment group and one control group, offering all companies a four-month entrepreneurship training program. Then we monitored the performances of the two groups over time.
Both the treatment group and the control group participated in ten sessions of general training, focusing on how to get market feedback and how to gauge the feasibility of ideas by searching, collecting and elaborating information before committing resources to actually developing these ideas. Specifically, the program consisted of four steps centering on market validation: business model canvas, customer interviews, minimum viable product and concierge or prototype.
As for the startups in the treatment group, we taught them to develop a framework for predicting the performance of their ideas, and to run rigorous testing on their hypotheses, just as scientists do in their research. With the control group instead, we let the companies follow their intuitions as far as evaluating their ideas using a method that normally produces fairly standard search heuristics.
Although we offered the same basic training to both groups, we did not provide decision-making criteria to the control group. More specifically, we taught the treated group to identify the problem, formulate theories, clearly define their hypotheses, run tests to prove or disprove their theories, measure their test results and finally make decisions based on the outcomes of all these techniques.
Our study highlights the fact that entrepreneurs who act like scientists get better results, more readily pursue new and original ideas, and are less likely to give up on a project in the initial stages. The results from our trial are consistent with the main hypothesis of our theory: a scientific approach actually improves the precision of predictions.
In our study, we focused on the decision-making process, which has taken on an increasingly central role in shaping the performance of companies that embark on new projects, in particular in the stream of research that ties together entrepreneurship and strategic management. We demonstrate that when entrepreneurs apply a scientific approach to their decision-making, their companies achieve better performance, which is an immediate advantage. In fact, entrepreneurs can ascertain, with some degree of precision, if and when their projects will generate low or high performance - and if and when it’s time to pivot and pursue other ideas. In other words, thanks to a scientific approach, entrepreneurs who have carefully considered and validated their theories on what their customers want, and thoroughly tested their hypotheses with experiments that reduce bias or imprecision in their market analyses, actually run less risk of false positives or false negatives.
In this research, we develop a model which proves that a scientific approach enables entrepreneurs to make better predictions, which explains why we observe different performance outcomes when we compare entrepreneurs in the test group with their counterparts in the control group. However, the time span of our RCT does not allow us to verify whether the companies in our sample eventually fail, or if the test companies fail faster without incurring higher costs. Another interesting question that we do not explore in our work is whether the scientific approach can also correct our inability to pursue false negatives. And last but not least, we show that a scientific approach helps larger companies make decisions, but we have not given any clues as to how this approach would play out within these complex organizations.
These are all questions that our study raises which future research will have to explore.
Strategic innovation in the digital era will improve the ability of any decision-maker, including those with limited knowledge of data analysis, to make informed decisions using data.
A journey through the digital revolution to learn how to capitalize on these enhancing technologies to gain competitive advantage in existing and new markets.
Ottimizzare la gestione di un ampio portafoglio clienti attraverso l’uso intelligente di diverse tipologie di dati, algoritmi, sistemi di CRM e sistemi di monitoraggio.
