Automating Your Museum

A five-part series

Stephen R. Toney
President & CTO, Systems Planning
Published on, February 2000

Copyright 2000


Part 5: Making Decisions

Choosing a collection management system (CMS) is all about making decisions. There is of course the big decision of which CMS to buy, but this decision is founded on thousands of smaller ones. To the extent that the smaller ones are made correctly, the big one becomes much easier. Therefore it is prudent to make even the small decisions by involving the right people and using a good process.

For space reasons, I will not cover decision theory, as it is based on statistical probability, and requires comparing various courses of action to costs or financial benefits; a typical decision theory scenario is the number of test oil wells to drill. These are not usually the kinds of decisions we have to make in nonprofit organizations. In our work, the major costs and benefits are hard to reduce to dollar amounts. Instead, we will concentrate on the people who make decisions and the way they make them.



Most important decisions in nonprofit organizations are made by groups of people. Therefore knowing something about how people work in groups can be highly beneficial.

As we hinted in article 1, groups have a life-cycle just as people do. The stages that a group goes through are usually given as "forming, storming, norming, and performing."

Progression through these four stages is a process that can take several hours to several months. A major determinant is the nature of the group's purpose and the clarity of its goals. Vague purposes require a great deal more Storming and Norming before the group can Perform. A group with a very limited, clearly defined goal may pass through the four stages rapidly. This suggests two ways to hasten the group's maturation: by providing it a clear written statement of its goals, and by having the group prepare it if it is not given.

Another major factor that influences the speed of a group's maturation is the interests, background, and social skills of its members. Rather than try to alter the individuals, the facilitator should employ techniques to emphasize their strong points and minimize their weak ones.

Some groups may never reach the Performing stage. It is particularly important, as you may image, to get past the Storming stage. Sometimes a good facilitator can help the group mature, or a change in the group's membership may be required.

I wish to emphasize that these stages are necessary parts of the maturation of any group. Often we hope to avoid the Storming stage by selecting only congenial people. However, groups cannot be made up of people that think alike: if everyone is singing the same notes, the music is not very interesting. In fact, diversity of members' styles has been shown to improve results. So if your group avoids all conflict, perhaps its needs some new blood.


Group Decision-Making

Why all this emphasis on groups? One reason, as we said, is that in museums, most important decisions are made by groups. Thus it is to everyone's benefit that the groups are making decisions in the best possible way.

The other reason is that groups at the "Performing" stage tend to make better decisions than any individual, especially where the problems are large, complex, and unclear. We are more creative, have more ideas, and come to better solutions.

Groups can make decisions in many ways. Here are four good ways to make decisions. Each has its uses, as well as its drawbacks.

As we discussed in article 1, the kind of decision-making process to use depends on the problem to be solved. If the problem is low-cost, without long-term considerations, and can be easily changed (what kind of coffee to buy for the staff lunchroom), make a quick decision using the autocratic or democratic styles. If the problem is high-cost, with significant long-term considerations, and not easily changed (what kind of CMS to buy), you should not settle for anything less than consensus.

Because consensus cannot evolve in an immature group, a consensus decision not only takes time to arrive at, but it takes time to evolve the group to the point where it work on the problem effectively.

The statement that consensus results in the best decisions applies even to technical problems. There is a standard problem in management training that consists of groups selecting which items from a list would be most important to survival in a desert or in the arctic. It has been shown that a group that allows one person to make the decisions -- even if that person is an expert -- makes worse decisions that a consensus among non-experts.

Here are some bad ways to make decisions:


Managing Group Decision-Making

One way to speed up the evolution of the group is to impose structure and discipline. This is best done by a facilitator, since the structure must be imposed from outside the group.

The most common structure that a facilitator imposes is to separate idea generation from idea evaluation. Since facilitators tend to love aphorisms, they like to say "You can't get hot water out of the tap while the cold water is running."

There are three steps to this approach:

1. Problem definition

Before ideas can be generated, the problem must be clearly defined. It is pointless to generate ideas about the wrong problem. Agreeing on the problem is a good way to have the group begin to work together. As with any problem statement, it should be in writing.

2. Idea generation

Idea generation can be fun, but only if criticism and evaluation are turned off temporarily. No ideas should be criticized during idea generation. Instead, participants should use ideas as springboards for new ideas.

When idea generation is working, group members will build on each others' ideas. Ideas will begin to flow faster and faster. Many of the ideas may be impractical, but they may spark in someone else a really good idea; that is why no criticism is allowed. You don't want to discourage the free association that generates new ideas. The goal is to have as many ideas as possible. The more ideas you have, the more likely it is that you will have some really good ones.

Brainstorming is the best known technique for idea generation, in which participants call out their ideas while the facilitator records them. Nominal group techniques, in which ideas are written instead of spoken, may be more effective in new groups in which the members have not come to the stage of trusting each other.

The flow of ideas cannot be interrupted without harm. It is best to isolate the group, whether the session is half a day or an entire week.

3. Idea evaluation

Idea evaluation requires a different mood and mode of thinking from idea generation. It is left-brained instead of right-brained.

The first step should be to agree on what a good solution would look like. These are criteria that will be used to evaluate the ideas, such as the cost, the timetable, and the degree of risk. One good method of developing criteria is called QFA -- Quality, Feasibility, Acceptability. It recognizes that although ideas may be inherently good (Quality), they may not be doable in your museum for lack of talent, funds, or other resources (Feasibility); or they may be so unpopular or risky in your museum that they could never be implemented (Acceptability). Using QFA as a guide to developing decision criteria helps make sure you are considering not only the inherent value of an idea but the context in which it would be applied.

Next, a matrix is developed with the criteria along the top and the ideas along the left. The cells of the matrix can then be given scores for how well each idea matches each criterion .

As you proceed, new ideas or variations will arise. Add these to the list as well. Obvious stinkers can be removed from the list -- as we said, idea generation will result in lots of ideas that are obviously unworkable. But do not eliminate ideas too fast. The goal is not to emerge from the meeting with one idea or solution; instead, because potential solutions probably will (and should) be reviewed over a period of time by your group or by others, you want to have a range of solutions.

Finally you must perform a reality check. When you think you have selected the best ideas, make sure that they will actually solve the problem. Although I have emphasized the value of a good process, there is the danger that a group can get carried away by the process, so it is also necessary to make sure the results make sense.



Most museums use a consultant to help them plan for a collection management system (CMS). This is because a museum may acquire a CMS only once every ten years, so museum staff naturally have little experience with it. A consultant may be involved in such decisions several times a year.

For the same reason that the number of CMS vendors is small, so is the number of consultants qualified to help you with this process. In North America there are only a handful.

What can a consultant do for you? Let's clarify one thing right away. The consultant should not be making decisions for the museum, especially The Decision about the best CMS. It is not good even for the consultant to participate in consensus decisions as one of the team, because he or she may be thought of as someone to whom the others should defer.

Instead, the consultant supplies two things: process and information.

Choosing a consultant thus boils down to assessing his or her ability to assist with the process and to provide information. In other words, the consultant must be able to work with you and your team, and must have current and accurate information, delivered free of bias.

Both of these areas are hard to assess. You can and should evaluate the personal chemistry yourself, as well as the ability to communicate problems, intellectual honesty, and reliability, but references from trusted colleagues at other museums should be part of the process too.

However, like anyone, the performance of the consultant depends partly on the context. For example, if a museum doesn't value planning, the consultant cannot lead the museum to plan effectively. Or if the museum has a pattern of autocratic decision-making, you can't expect the consultant to develop consensus.

Consultants are expensive, so you want to use them where you will get the most return for the dollar. On some projects, a consultant simply gathers data and analyzes it in straightforward ways, but this is the 80 percent of the work that accounts for 20 percent of the value. The more of this you can do in-house, the less you'll have to pay the consultant. The more effective use of a consultant is in applying intelligence and experience to your unique situation; this is the 20 percent that can make the big difference between success and failure. This is what you really need and should pay for.

One problem I see often is that a museum will hire a consultant who is good in certain areas but not for the task they were hired for. The museum doesn't realize the difference, or feels that they are closely related. This is natural -- if you are not a technology specialist you tend to lump together those who are, but in fact technology is a tremendous range of specialties, and is becoming ever more fragmented. Make an effort to understand enough about the problem to find someone with relevant experience.


How To Manage Meetings

There is nothing worse in daily work life than boring, unproductive meetings. And there are few things better than a crisp, well-managed, productive meeting. What's the difference? If you've read the preceding articles, I'm sure you will not be surprised to hear that the difference is planning.

For a meeting, a plan is called an agenda. Seeing the agenda as a plan may help you realize that it must be more than a list of topics. This is as pathetic as would be an RFP consisting of a few bullet points.

Instead, an agenda should describe the results expected from the meeting. Consider the difference between these two examples:

Bad example

Good example

The bad example is just a list of problems or points to be addressed. The good example shows that the context is provided for each topic, and some thought was given to what should result from the discussion. In other words, it states the goals of the meeting.

You can see also that some thought was given to how much time should be spent on each topic. This is the only way to cover all the topics and still end the meeting on time. Otherwise the early topics will get most of the time and the later ones will be Decisions by Default or by Fatigue (see above).

One easy way to improve the value of an agenda is to distribute it before the meeting. Obviously there would be little point to doing this with the "bad example" above, but I hope you can see that if the "good example" reached your desk a few days before the meeting, you would be able to use it to review pertinent documents or to think about what you want to say in the meeting.

So a good agenda distributed ahead of time maximizes the productivity of the meeting, because people have had a chance to think about the issues ahead of time. Less time is spent on vague discussion and more on possible solutions.

When you distribute the agenda ahead of the meeting, include with it background documents or information. This can reduce the time needed in the meeting for transmitting information and increase the time available for discussion.

These ideas help the meeting be what it should be, which is face-to-face discussion, not information distribution which can be done by email or on paper. Especially you never want to spend time in a meeting with everyone reading documents -- this is a total waste of time.

The next biggest waste of time is having everyone writing documents. There is no surer way for a group to feel productive without being so. Instead, have one or two people draft the document, send it to everyone for email comments. You may be able to write the entire thing without any meetings at all.


Summary Of Major Ideas From This Series

Part 1: Making a Plan       Part 2: Managing the Project       Part 3: Technology       Part 4: Sample RFP

All contents of website, including HTML and JavaScript, copyright © 1996-2014 Systems Planning. MWeb, InFORMer, and CAPS are trademarks of Selago Design, Inc. MARCView and MARConvert are trademarks of OCLC, Inc.