Delphi method

The Delphi method, also known as the Delphi technique, is a process for gathering expertise or opinions from people with specialized knowledge of a particular subject for the purpose of making predictions or forecasts. The information is gathered anonymously from pre-determined experts through a series of questionnaires. The goal is to arrive at a consensus of opinion from the experts on the topic in question. This information is then used in a decision-making process by the company or other entity that initiated the process. The Delphi method is a cost-effective way to gain the expertise of a large number of subject matter experts. It has proven to be at least as effective as other ways of gathering such information; however, the process can take a long time to complete.

Background

The Delphi method is named after the ancient Greek oracle at Delphi. The oracle was consulted by leaders and commoners alike when they needed to make decisions. The prophecies made by the oracle were believed to have divine influence.

The method was developed in the 1950s by Olaf Helmer (1910–2011) and Norman Dalkey (1915–2004), two Rand Corporation employees. The process was originally designed as a way to forecast the effects of a nuclear war. It is a non-quantitative approach to reaching a consensus, which means it relies on information that is not specifically measurable. The process uses the anonymous opinions of a number of experts and allows them to change their minds based on the reasoning shared by others. The expectation is that by bringing together the thoughts and expertise of a large number of people, a consensus based on the best available knowledge will be reached and can be used as a basis for decision-making.

Overview

The first step in the Delphi method is for a facilitator to choose the experts who will participate. This decision is based on the topic to be considered and the areas of study or other expertise of the participants. Meanwhile, this facilitator, or a researcher working with the facilitator, creates the first questionnaire. This usually includes several specific questions along with some open-ended questions that allow the experts to add concerns that the survey originator did not consider. The facilitator also decides on a threshold for when the project will be completed. For instance, the facilitator may decide that the group will stop when 85 percent of the respondents are in agreement.

The questionnaire is then sent to the preselected participants. Originally, this was done by mail, but in the 2020s, it was nearly always done by email. The respondents complete the questionnaire and return it within a specified time period. When the surveys are returned, the facilitator reviews the results and prepares a second questionnaire to follow up on the first. The second survey is sent to the participants with a summary of the responses to the first questionnaire, which are reported anonymously. The participants review the results of the first round and then respond to the second questionnaire. This process of review and resurveying, accompanied each time by the anonymous feedback from the previous round, is repeated as many times as necessary for the predetermined threshold of consensus to be reached or until the facilitator determines that consensus is not possible.

Each previous round's feedback summary highlights the areas of disagreement. The respondents are free to reevaluate and change their opinions based on the input from their colleagues. The facilitator may also incorporate questions based on the input from the experts to the open-ended questions, which may also affect the answers of some of the respondents. Once the consensus level is reached, the facilitator reports the results to the group and to those who requested the process. If a consensus cannot be reached, the facilitator may reexamine the available information on the subject at hand and see if enough of a consensus was reached to address the issue or if another process needs to be tried. When consensus is reached, the decision-makers can use it to help forecast or predict the outcome of a planned action or determine which course of action should be undertaken.

The Delphi method has a number of advantages. It allows many subject matter experts to be part of the process at a relatively low cost; while they can meet face-to-face, it is not necessary, and the entire process can be accomplished via email. This low-cost factor means that more experts from a wider range of related fields can be included if desired, improving the pool of information for the decision-making process. Because each respondent answers the questions independently and anonymously, a greater opportunity exists for each to share their true thoughts and opinions without the influence of others.

The process also has some disadvantages. The need to repeatedly survey the participants means the process can take some time, which may not always be desirable. There is also no guarantee that the chosen experts will want to take part or give the process an adequate level of attention. The written method of responding may not be ideal for some participants, whose expertise may be in other areas besides written communications. The success of the process is also somewhat dependent upon the quality of the facilitator. The facilitator can knowingly or unknowingly create a bias in the responses by the way survey questions are formed. The quality of the summary and the extent to which the facilitator investigates the points of dissension can also affect the outcome.

To maximize the benefits of the Delphi method in solving complex issues, some practitioners recommend pairing the technique with another analysis, such as qualitative analysis. In some cases, a modified Delphi method may be most appropriate. These methods may incorporate additional or different feedback structures or additional analysis. Many professional fields use the traditional or modified Delphi method, including healthcare, environmental science, education, and technology.

Bibliography

Blackwood, Rosalind, and Claire Currie. "The Use of Delphi Methods." HealthKnowledge, 2016, www.healthknowledge.org.uk/public-health-textbook/research-methods/1c-health-care-evaluation-health-care-assessment/use-delphi-methods. Accessed 2 Nov. 2025.

“Delphi Method: Definition, Stages, Pros, Cons, Examples.” Indeed, 6 June 2025, www.indeed.com/career-advice/career-development/delphi-method. Accessed 2 Nov. 2025.

"Delphi Method." Rand Corporation, www.rand.org/topics/delphi-method.html. Accessed 2 Nov. 2025.

Kea, Bory, and Benjamin Chih-An Sun. "Consensus Development for Healthcare Professionals." Internal and Emergency Medicine, vol. 10, no. 3, 2014, pp. 373-83, doi.org/10.1007/s11739-014-1156-6. Accessed 2 Nov. 2025.

Nasa, Prashant, et al. “Delphi Methodology in Healthcare Research: How to Decide Its Appropriateness.” World Journal of Methodology, vol. 11, no. 4, 2021, pp. 116–29, doi.org/10.5662/wjm.v11.i4.116. Accessed 2 Nov. 2025.

Twin, Alexandra. "What Is the Delphi Method, and How Is It Useful in Forecasting?" Investopedia, 8 Sept. 2025, www.investopedia.com/terms/d/delphi-method.asp. Accessed 2 Nov. 2025.

"What Is the Delphi Method – Pros, Cons, and Examples." Monday, 27 Oct. 2024, monday.com/blog/project-management/delphi-technique. Accessed 2 Nov. 2025.