THE VALUE OF THE NOMINAL GROUP TECHNIQUE
By Harold Kurstedt
NGT is wonderful for idea generation and efficient meetings, but not quite so good for effective meetings and consensus.
Nominal Group Technique (NGT) is a common, valuable, often misused tool for group decision making. You misuse the NGT when you expect NGT to bring a group to consensus. Consensus requires identifying, acknowledging, scoping, confronting, and resolving conflict; and NGT alone will accomplish none of these activities. However, NGT is wonderful for free and abundant idea generation and for efficiently (not effectively) bringing a group to a relatively-well-documented decision. An efficient meeting runs smoothly and generates results. An effective meeting is on the right topic, involves the right people, occurs at the right time, and generates the right results.
In a world of wasted or even counterproductive group interaction in meetings, any semblance of process, progress, or result can help people feel productive, which is good. However, in this environment being lulled into a temporary feeling of agreement and commitment followed by no follow through on decisions made will make people even more frustrated than before, which is bad. Therefore, if you use NGT for what it is best suited for and are careful that people have realistic expectations, NGT can be wonderful. Furthermore, the individual steps of NGT are wonderful examples of steps you can custom tailor into a group decision making process to achieve specific results you want.
The steps of NGT sometimes are most useful when you use only one or two of the steps in a facilitation process to move a stagnated group forward. I add a few steps to the standard five-step NGT process so I can help some of NGT’s weaknesses. For NGT, helping weaknesses doesn’t resolve weaknesses. Your best bet is to recognize the strengths and weaknesses of NGT and use NGT in whole or in part to take advantage of the strengths.
Where does NGT fit into facilitation? In the 8-P model for planning meetings, the sixth P is for process. NGT is a set of steps that suggests a process for facilitation. NGT is a process tool for facilitation. You can use one or more of the NGT steps as you identify the process you want to use for facilitating the group. Also, the 8-P model requires before you begin the process (such as NGT) that you have clearly defined the issue to be worked on (the third P is for problem) and the people who will participate (the second P is for people). Example issues might be 1) What are the action items we need to accomplish to increase our sales by 20% this year?, 2) What skills do management systems engineers need to be successful?, and 3) How can we find better programmers in a highly competitive market for good talent? (Consider the 8-P model. The fifth P is for participation. When using the NGT, participation is free and open information sharing and controlled, facilitated participative decision making.)
Each participant needs to know his or her role in the meeting and what contribution is expected of him or her. When a willing and able participant doesn’t have a role, he or she will contribute something, and that something is usually disruptive. You also need to make sure all the perspectives of the issue under consideration are fairly represented among the participants. Unrepresented stakeholders and other critics will discredit the results of any decisions made by a group, thus discrediting the composition of the group. When assembling the group, consider all stakeholders. Any stakeholders not represented are not going to be committed to the result. The NGT is designed to get relatively equal and unbiased ideas from each participant in the group and to rank order the ideas.
The words nominal group mean a group in name only. To make an effective intervention in the workings of an organization supported by the group, you need to convert the nominal group into a real group. A real group is a group of people who share common interests and are able to communicate well so they can develop consensus around needed results that they are committed to following through with.
The potential problem with a real group is group think. Group think is when a group is so close and collaborative that a bad idea in the group isn’t challenged enough and the group risks going to Abilene (a popular story of the difficulty in group agreement that the group, in fact, chooses not to follow) and other dysfunctional behavior.
I find that 16 participants is the best number of people to be facilitated in group decision making. My nuclear engineering background suggests a magic number related to 16. (Four squared is 16, and two squared is four.) My facilitation experience suggests that more than 16 is difficult and less than 16 isn’t good representation–unless, of course, the number of participants less than 16 is everyone with a stake in the result. I have facilitated groups with as many as 40 people and as few as three. The problems with large groups are not moving the process at a fast enough rate to keep everyone’s interest and having trouble making sure everyone has ample opportunity to express their views. The problems with small groups are inadequate representation, and lack of mutual stimulation of creative idea generation.
Successful NGT (or any participative decision making tool) requires a convenor (usually a manager) who believes in participation and the value of group action. If potential group members believe any participative activity is an empty activity, he or she will find a way out of the meeting and send someone in his or her place who may not have the ability, understanding, or authority to represent his or her constituency. Then, you will have an empty activity. A good convenor provides a need, impetus, and resources to bring the group together and supports the group action before, during, and after the meeting.
I’ve never had a group with a bad result from the interaction of the participants. I’ve had participants who were outliers, with strange, angry, or misguided ideas. But the will of the group filters the ideas of outliers well. I recommend that the convenor agree to implement to group’s ideas at the outset–or at least agree to implement some fraction of the ideas, like 70%, 80%, or 90%. That means that the convenor believes that 7, 8, or 9 of about ten high-priority ideas generated by the group are as good as or better than what the convenor might generate. See the exercise at the end of this chapter for a good reason to believe that the group’s result may be the one to use.
Since a well-constituted group has broader and more specific knowledge than any one person, the odds are in favor of the group’s result. Since the group members or the constituencies of the group members will probably be the ones to implement the ideas or will have the ideas implemented on them, the group’s result clearly is the one to follow. The convenor may have some inside information affecting an idea or two, but groups will accept that response, even after the fact. That is, if I do my best to generate a result and you tell me the majority of what I did is meaningful and will be implemented but due to inside information you can’t share with me, you’ll make minor changes, I’ll buy in.
If the participants in a group believe most or all of their hard work will be implemented, they’ll be motivated and bring an even greater sense of responsibility to their work. I am positive the group will generate an excellent, responsible action plan. I recommend the convenor follow the plan.
Successful NGT requires a facilitator who only works the process and doesn’t get involved in the content of the discussion. The objectives of the facilitator are to move the process forward and to ensure that each participant believes he or she has had adequate opportunity to express his or her views. Our studies have shown that people feel more like consensus if they feel they’ve had adequate opportunity to express their views. No other variable seems to have much effect on their feeling of consensus–at least feeling consensus during the meeting and shortly thereafter.
Based on my experience, the big worry is that a short time after the NGT meeting there is little commitment to the NGT results and the meeting doesn’t yield anything tangible but the meeting itself. Especially after a well-facilitated NGT meeting that generates a feeling of accomplishment not usually felt in meetings and a feeling of having had opportunity to express views, a participant can become even more frustrated when the results lead nowhere after the meeting. Some consultants have been quite successful in leading good NGT meetings with a feeling of accomplishment and a feeling of consensus and then escaping before participants attach any responsibility to the consultant for no follow up or follow through.
The key to true consensus is not only acceptance, but agreement and commitment. Agreement and commitment imply that people are willing and ready to follow up and follow through. One consideration as a facilitator is to get the group to define realistic expectations on the group’s ability to follow up and follow through. To make something tangible happen, the group and its members need to be ready, willing, and able to carry out the actions implied by the decisions they make.
In my experience, I tell the group up front that NGT is good for idea generation and for efficient meetings and decision making. I also tell the group that idea generation is necessary but not sufficient for good consensus and that they have to work on consensus after the NGT part of the process. The NGT gets them part way there but not all the way to consensus. Groups understand and appreciate the truth and are thankful for any progress made in a meeting, thereby making NGT successful over the long haul. To move beyond the idea generation in NGT, the facilitator needs to be good at dealing with and resolving conflict and then moving on to consensus.
The NGT steps I’ll discuss are shown in Figure 1. The first five are standard NGT steps. Steps six through eight are steps I’ve added over the years to make NGT more effective.
~ This is part 1 of a paper published by Harold Kurstedt. Please e-mail us for the rest of the article.