A new dynamic in the decision-making process
Let's take the example of the debt management process that Claire and her team are facing. Traditionally, the team had to manually analyse large volumes of data, such as payment histories, debtor profiles and economic forecasts, to make informed decisions. Claire would then have to evaluate the different collection options and choose the most appropriate one, while ensuring that the process remained fair and humane.
Today, thanks to advanced analysis algorithms, some of this work can be automated. An algorithm can analyse far more data than a human being could: payment histories, demographic data, etc. Using these facts, the algorithm can classify accounts into simple categories, for example red-yellow-green, making it easier to process cases.
In this example, a system can analyse debtor data in real time, identify high-risk accounts and propose optimised collection strategies. This allows Claire and her team to concentrate more on strategic decisions and to have more human interaction with debtors, tailoring the approach to each situation, rather than spending time on repetitive, administrative tasks.
Technology can lighten the workload, but Claire must also ensure that her team retains an essential human dimension, particularly in communications with customers, in order to maintain a respectful and empathetic relationship.
The importance of human interaction in an automated process
While algorithms make it possible to automate a large part of the process, human intervention remains essential at several levels in the life cycle of a decision.
- Data collection and weighting: Software can assist in the collection and analysis of critical information, but it is humans who determine which data to collect and how to weight it.
- Formulation of alternatives: Advanced analytical models can eliminate the least viable alternatives, but decision-makers need to be trained to interpret these results and use decision-support tools.
- Experimentation and learning: In categories where human intervention is required, the results of test-and-learn experiments can improve the quality of decisions. Teams need to identify which tests to conduct and interpret the results.
The challenges of organisational transformation:
For Claire, the integration of algorithms into the decision-making process represented much more than a simple technological adoption; she had to carry out a genuine organisational transformation, in particular by taking into account the varied styles within her team. For example, Monique, who is very empathetic, tends to be more flexible in her handling of cases, always seeking to understand the reasons behind debtors' difficulties. Serge, on the other hand, believes that everyone should take full responsibility for the consequences of their choices and applies procedures strictly, with no exceptions. This diversity of approach made the transformation even more complex, as a balance had to be struck between the interpretation of decisions by the algorithms and the individual sensitivities of each team member.
One of the first challenges Claire faced was to redefine the roles within her team. For example, some members of staff, who were used to repetitive data analysis tasks, saw their role evolve towards more strategic functions, requiring a detailed interpretation of the results provided by the algorithms. Claire had to deal with these differences in style to ensure that the new technologies were accepted and used effectively by everyone.
The cognitive biases of her colleagues also influenced their ability to fully accept the algorithms' recommendations. For example, Monique, with her natural empathy, was often prone to confirmation bias, looking for evidence in the data to justify a more flexible and humane approach to helping debtors. She tended to favour individual cases and question algorithmic recommendations when they seemed too rigid or disconnected from the human realities she perceived.
For his part, Serge, who was very attached to rules and structure, sometimes fell victim to anchoring bias. When an algorithm suggested an action different from the usual procedures, he focused on his habits and on the first information to which he was exposed, refusing to fully consider the new solutions proposed by the analytical tool. He found it difficult to break out of the pre-established framework and consider that the algorithm could offer a valid alternative.
To overcome these biases, Claire set up awareness-raising workshops, helping everyone to become aware of their own cognitive biases and develop a more critical approach to algorithmic recommendations. The aim was to make Monique and Serge understand that, although the algorithm is a powerful tool, it should be used as a complement to their human expertise, and not as a unique or infallible solution.
In this way, by working on their biases, they could better navigate between human intuition and algorithmic analysis, striking the right balance in their decisions.
Conclusion
This project with Claire highlights the fact that optimising the decision-making process in business no longer relies solely on human experience or available data. Integrating algorithms and advanced analysis into the decision-making process not only optimises strategic choices, but also transforms the organisation. To maximise the benefits of this new era of decision-making, businesses need to adopt a systematic approach that combines the analytical rigour of algorithms with human intuition, while training their teams in the new skills required, such as communication, collaboration and knowing how to act under uncertainty. Decision-making has become an art that every decision-maker should practise.
Thanks to this approach, Claire has seen a marked improvement in collaboration within her team, with a confidence in technology that is still tentative, but growing. This highlights the crucial importance of the alliance between people and technology in overcoming today's organisational challenges.
To find out more :
Decision-making: making the right choices in an uncertain environment
