What is the Delphi Method – Pros, Cons, and Examples [2026]

Some decisions are too complex for spreadsheets and too important to leave to one person’s judgment. When a project faces uncertain risks, unclear requirements, or competing priorities, teams need a structured way to tap into expert knowledge without letting the loudest voice in the room dominate the conversation.

The Delphi technique is a structured method for building consensus among a panel of experts through multiple rounds of anonymous questionnaires and controlled feedback. Developed during the Cold War for military forecasting, the method is now widely used in project management, healthcare, research, and organizational strategy to make reliable decisions where hard data is limited.

This guide covers the Delphi technique in more detail, including how teams use monday.com’s AI Work Platform to run structured consensus processes.

Key takeaways

• The Delphi technique builds consensus through anonymous expert panels and iterative feedback rounds
• The technique has 4 defining characteristics: anonymity, controlled iteration, statistical aggregation, and expert selection
• You can apply the Delphi technique in project management, risk identification, forecasting, research, and HR strategy
• Variations of the Delphi technique include Modified Delphi, Real-Time Delphi, and Wide-Band Delphi for different contexts
• Modern tools like monday.com’s AI Work Platform support the technique using voting, forms, dashboards, and real-time intelligence 

What is the Delphi technique?

The Delphi technique is a forecasting and decision-making method that collects expert opinions using several rounds of anonymous questionnaires, with controlled feedback between rounds. The goal is to reach a reliable consensus in complex situations where quantitative data alone doesn’t lead you to an answer.

Based on the principle that structured group judgment outperforms individual expert opinion, the Delphi technique (also known as the Delphi consensus method) works by isolating each expert’s thinking from social pressure while still giving them access to the collective picture. Experts revise their views after seeing anonymized summaries of the group’s responses. This iterative cycle typically takes 2-4 rounds, with 3 being the most common.

Rather than relying on a majority vote, facilitators measure differences in opinion using the median response and interquartile range (IQR) across rounds. When the IQR falls within a pre-defined threshold, the group has reached statistical consensus. This approach produces more nuanced, defensible results than open debate or simple polling.

A brief history of the Delphi technique

Named after the Ancient Greek Oracle of Delphi, the method was developed by Olaf Helmer and Norman Dalkey at the RAND Corporation in the early 1950s. General Henry Arnold had commissioned a report on future technologies that could be useful for military planning. Traditional forecasting methods proved inadequate for questions where established scientific theories did not exist, so the team turned to structured expert consultation.

The Delphi protocol proved effective for generating reliable predictions about possible threats and emerging technologies. Through the 1970s and 1980s, the method spread from military forecasting into healthcare, social policy, and education. Today, it forms the basis for clinical guideline development, national standards processes, and technology foresight programs across dozens of countries.

Key characteristics of the Delphi technique

5 features distinguish the Delphi technique from other group decision strategies:

• Anonymity: Participants submit their opinions without knowing who said what. This prevents the highest-status or most senior person from anchoring the group’s thinking. By removing authority bias and the halo effect, anonymity encourages honest, independent judgment from every panel member.
• Controlled iteration: Rather than a single open discussion, the process moves through structured rounds. Each round builds on the last, giving experts a chance to reconsider their positions in light of new information rather than defending an initial stance.
• Anonymized feedback: Between rounds, facilitators compile and distribute summaries of the group’s responses. Experts never see each other’s raw answers. They see the aggregated picture: where the group clusters, where views diverge, and what reasoning supports different positions.
• Expert selection: Participants are chosen for domain expertise, not drawn from a random sample. A well-composed panel typically includes 10 to 30 experts with diverse perspectives on the problem, avoiding over-representation of any single viewpoint.
• Statistical aggregation: Facilitators summarize each round using the median response and interquartile range, giving experts a clear picture of where the group clusters and how far outlier views sit from the center. This replaces subjective interpretation with measurable convergence.

How does the Delphi technique work?

To run a Delphi study, follow these 7 steps:

1. Appoint a facilitator

Choose a neutral person who understands the domain and can design effective questionnaires, summarize responses objectively, and manage the logistics of multiple rounds. Crucially, the facilitator doesn’t contribute their opinions.

2. Define the research question

Narrow the problem to a specific, answerable question. Vague questions produce vague consensus. For example, “What are the top risks for our infrastructure project in Southeast Asia?” is more productive than “What could go wrong?”

3. Select the expert panel

Identify 10 to 30 professionals with relevant domain expertise. Aim for diversity of perspective: different specializations, geographic contexts, and organizational roles. A homogeneous panel will converge quickly but may miss critical blind spots.

4. Design and distribute the Round 1 questionnaire

In the classical Delphi method, Round 1 is open-ended: experts generate ideas, risks, or predictions without constraints. In the Modified Delphi technique, Round 1 starts with a pre-established list of items derived from prior research. Collect all responses anonymously.

5. Analyze and summarize responses

Calculate the median, IQR, and key themes across responses. Identify areas of convergence (where experts agree) and divergence (where they disagree). Prepare a structured summary for the next round.

6. Conduct further rounds

Distribute the next questionnaire alongside the previous round’s summary. Experts review the group’s aggregated picture and revise their responses. Repeat until the pre-defined consensus criteria are met.

7. Define consensus and close

Consensus is typically defined as the IQR falling within a pre-specified threshold. For example, on a 9-point scale, the study might close when the middle 50% of responses fall within a 2-point range. State stopping criteria at the start of the study so closure is objective, not arbitrary.

If you haven’t reached a consensus after 4 rounds, facilitators typically analyze persistent divergence rather than continuing indefinitely. Documented disagreement is a legitimate and valuable output.

Different applications of the Delphi technique

The Delphi technique adapts to any context where expert judgment matters more than historical data. From project scoping to clinical guideline development, the method provides a structured way to surface insights that quantitative models alone would miss. Below are the most common applications across industries.

Delphi technique in project management

In project management, the Delphi technique is a good fit wherever decisions depend on expert judgment rather than historical data. 3 applications stand out:

• Scope definition: When project requirements are unclear or contested, a Delphi panel of stakeholders can clarify what belongs in scope and what doesn’t. This is especially valuable in early-stage projects where ambiguity around deliverables is a leading cause of scope creep.
• Estimation: The Wide-Band Delphi technique, developed specifically for project cost and effort estimation, uses the same iterative anonymous structure to generate reliable estimates when historical data is thin. It’s widely used in software development and construction planning.
• Vendor selection: For complex procurement decisions with multiple evaluation criteria, a Delphi consensus process among cross-functional evaluators produces more balanced, defensible supplier recommendations than a single-round scoring exercise.

Delphi technique in risk management

The Delphi technique is particularly effective for risk management because it identifies threats that quantitative models alone would miss. Expert panels can spot emerging risks, second-order effects, and low-probability scenarios that don’t necessarily appear in historical data.

In a typical Delphi risk assessment, experts anonymously rate each risk’s probability and impact over multiple rounds. The anonymity is critical: without it, a senior leader’s initial assessment can anchor the entire group, causing the team to underestimate risks that the leader dismissed. By using anonymous rounds, the process captures genuine expert disagreement and produces a more honest risk register.

Delphi technique in forecasting

The Delphi method of forecasting uses structured expert consensus to predict future trends, technologies, or events, particularly in areas where historical data is incomplete or unavailable. This is one of the method’s oldest and most established applications.

Common forecasting applications include technology foresight (which emerging technologies will affect a market within 5 years), economic and market forecasting (where quantitative models lack sufficient training data), and scenario planning and predictive project management for strategic decisions that depend on unpredictable external factors. The method is especially valuable when organizations need to plan for futures that current data cannot model.

Delphi technique in research

Academic and policy research is one of the most common global applications of the Delphi technique. Researchers use it to build consensus on clinical guidelines, educational standards, evaluation frameworks, and policy recommendations.

When a question requires expert judgment rather than empirical measurement, Delphi provides a structured, transparent method for arriving at defensible conclusions. It’s a good fit for monitoring and evaluation contexts where stakeholder perspectives must be systematically integrated.

Delphi technique in HR and organizational strategy

In human resources, the Delphi technique helps organizations define job competency frameworks, set salary benchmarking ranges, and develop HR policies where stakeholder opinions diverge.

As HR decisions often involve subjective judgment across multiple departments, the structured anonymity of Delphi produces more balanced outcomes than traditional committee discussions. This is an underserved application with significant potential for teams managing workforce planning across distributed organizations.

Variations of the Delphi technique

The Delphi technique continues to evolve, with new variants adapting the method to faster timelines, digital tools, and AI-assisted analysis.

Modified Delphi technique

The modified Delphi technique differs from the classical method in one fundamental way: rather than starting with open-ended expert ideation, Round 1 begins with a pre-established list of items derived from a literature review or prior research. Experts evaluate and rate existing items rather than generating new ones from scratch.

This approach shortens the process significantly, often requiring only two rounds instead of three or four. It is appropriate when a preliminary framework already exists and the goal is to validate, prioritize, or refine rather than discover. The modified Delphi technique is widely used in healthcare and nursing (developing clinical practice standards), education (curriculum framework validation), and policy research. It is sometimes called the abbreviated Delphi or consensus conference method.

Real-time and online Delphi

Real-Time Delphi removes the sequential round structure entirely. Instead of waiting for all experts to respond before starting the next round, participants enter responses through a live platform and see aggregated results update continuously. Experts revise their estimates at any time as they watch the consensus evolve. This compresses multi-month studies to days and is increasingly used in technology foresight and policy research.

The broader shift to e-Delphi, conducting studies through web-based survey platforms, has removed geographic barriers, shortened round turnaround, and lowered logistical costs. Most modern Delphi studies are now conducted online by default.

An emerging development is AI-assisted Delphi, where large language models automate facilitator summaries, categorize free-text responses, and identify points of convergence and divergence between rounds. While promising for reducing the manual burden on facilitators, AI-assisted Delphi is still experimental, and methodological debates remain around whether synthetic expert panels can substitute for genuine domain expertise.

Wide-Band Delphi

Wide-Band Delphi was developed specifically for project cost and effort estimation. It applies the same iterative anonymous expert structure to numerical estimates: each expert independently estimates effort, time, or cost, reviews the group’s aggregated estimates, and revises over multiple rounds. It is widely used in software development and construction project planning where historical benchmarks are incomplete or unavailable.

Delphi technique advantages and disadvantages

Like any structured decision-making method, the Delphi technique comes with clear strengths and practical limitations. Understanding both helps teams decide when the method is worth the investment and when a faster alternative makes more sense.

Advantages

• Removes authority bias: Anonymity prevents the most senior or outspoken person from anchoring the group, leading to more honest expert input
• Enables distributed participation: The asynchronous format allows experts from different geographies and time zones to contribute without travel or scheduling constraints
• Promotes genuine reconsideration: The iterative structure encourages experts to update their views based on new information rather than defending their initial position
• More reliable than individual judgment: For complex, low-data decisions, structured group consensus outperforms any single expert’s prediction
• Highly adaptable: Variants like Modified Delphi, Real-Time Delphi, and Wide-Band Delphi adapt the method to different timeframes, contexts, and organizational needs

Disadvantages

• Time-intensive: The classical format requires multiple rounds over weeks or months, making it impractical for decisions that need rapid turnaround (unless using Real-Time Delphi)
• Risk of attrition: Experts may drop out between rounds, compromising the panel’s integrity and the reliability of later-round results
• Consensus is not guaranteed: Persistent expert disagreement may indicate genuine uncertainty rather than a flaw in the process. Facilitators must recognize when divergence is the legitimate finding
• Quality depends on expert selection: A poorly composed panel, one that is too homogeneous or lacks genuine domain expertise, produces unreliable results regardless of how well the process is run
• Facilitator skill is critical: Poor questionnaire design or biased response summaries can steer the panel toward a predetermined outcome, undermining the method’s value

Delphi technique vs Nominal Group Technique

The Nominal Group Technique (NGT) is another structured consensus method, but it works very differently from Delphi. NGT brings participants together in a single facilitated session where they silently generate ideas, share them in a round-robin format, discuss and clarify each item, and then vote to prioritize. The process is face-to-face (or synchronous virtual), transparent rather than anonymous, and completes in one session rather than multiple rounds.

Both methods aim for structured consensus, but they suit different contexts. The table below compares the two approaches across key dimensions:

How monday.com's AI Work Platform supports the Delphi process

Running a structured Delphi process, coordinating multiple rounds, aggregating expert responses, and keeping the panel engaged, has traditionally been manual, slow, and logistics-heavy. monday.com’s AI Work Platform provides the infrastructure to run a structured consensus process without the administrative overhead.

Vote column: Teams can set up anonymous voting rounds directly on any item on their board. No one sees who voted how, just the aggregated result. This mirrors the anonymity principle that makes the Delphi technique effective.

Forms: Each Delphi round involves a structured questionnaire. monday forms lets facilitators design and distribute multi-question surveys, collect expert responses on a structured board, and track who has responded, all without leaving the platform.

Dashboards: After each round, facilitators need to see where the expert panel clusters. Dashboards aggregate form responses and vote data visually, making it straightforward to identify convergence and spot persistent outliers across rounds.

Automations: The most manual part of running a Delphi study is round management: reminding experts to respond, notifying them when a new round opens, and flagging when a response threshold has been reached. Automations handle all of this without manual follow-up.

AI Blocks and monday sidekick: When experts provide free-text responses, as they typically do in Round 1 of a classical Delphi, AI Blocks can summarize and categorize responses by theme. monday sidekick notices patterns and points of convergence for the facilitator, reducing the synthesis work that traditionally takes days to hours.

monday agents: For teams using Delphi specifically for risk identification, the Risk Analyzer agent flags tasks and risks proactively, complementing expert consensus data with real-time signals from the project itself.

For teams that want a fully custom consensus workflow, monday vibe lets you build a purpose-built Delphi application tailored to your organization’s specific needs, no coding required.

Running expert consensus in a changing world

The Delphi technique’s core strengths, structured anonymity, iterative convergence, and statistical aggregation, are as relevant to modern decision-making as they were during the Cold War. What has changed is the tooling and the speed.

With Real-Time Delphi, online platforms, and AI-assisted synthesis, organizations can run structured expert consensus processes in days rather than months. For complex decisions in project management, risk, research, and strategy, the Delphi method continues to prove its value across industries and contexts.

monday.com’s AI Work Platform gives teams the infrastructure, anonymous voting, structured forms, dashboards, automations, and AI assistance, to bring the Delphi approach into everyday project workflows. Get started to see how structured consensus fits your team’s decision-making process.