Exhibition evaluation basics for museum teams

What is exhibition evaluation and when does it happen?

Exhibition evaluation is the structured practice of checking, with real visitors, whether an exhibition communicates what the team intended, before, during, and after it opens. The Smithsonian Institution's Office of Policy and Analysis, for instance, spent years conducting studies of Smithsonian visitors and of individual exhibitions for the Institution's museums (Smithsonian Office of Policy and Analysis), and museum practice generally organizes evaluation into three stages tied to the exhibition timeline: front-end evaluation before design begins, formative evaluation while prototypes are still changeable, and summative evaluation after the show opens to the public. Each stage answers a different question, and skipping one does not make the show safer, it just means the team finds out about a problem later and at higher cost to fix.

What does front-end evaluation check, and why does it come first?

Front-end evaluation happens before a single label is drafted or a single case is built. It asks what visitors already know, believe, or misunderstand about the proposed topic, and what language or framing will make sense to them. This stage typically uses short interviews, surveys, or simple prompts with people representative of the expected audience. The findings should feed directly into the exhibition brief, the same document described in drafting an exhibition brief that guides decisions, because a brief written without knowing what visitors already assume is more likely to talk past them.

Front-end evaluation is cheap compared to what comes later. A misjudged assumption caught here costs a conversation; the same assumption caught after fabrication costs a rebuild.

What does formative evaluation test, and why must it happen while things are still changeable?

Formative evaluation tests mock-ups, draft labels, prototype interactives, or a walk-through of the planned sequence, while changes are still inexpensive. Typical formative questions include: can a visitor find the object the label is describing, does the label's first sentence get read and understood, is the physical sequence of the galleries clear without staff intervention, and do visitors leave with the core idea the brief intended. Testing with even a small number of visitors, using think-aloud walk-throughs or short intercept interviews, routinely surfaces problems that no amount of internal review catches, because staff already know what the label means and cannot read it the way a first-time visitor does.

Run formative testing on label language established in writing interpretive labels in plain language specifically, since label comprehension is one of the most common and most fixable formative findings.

What does summative evaluation measure after opening, and what should a team do with it?

Summative evaluation happens after the exhibition opens and typically combines observation of visitor behavior (where people stop, how long they stay, what they skip) with exit interviews or surveys about what visitors understood, felt, and remembered. The goal is not to grade the exhibition after the fact with no further use; it is to record, in writing, what worked and what did not, so the next exhibition brief starts from evidence rather than assumption. A summative finding that visitors consistently missed the exhibition's core argument should change how the next brief states its argument and how early labels are drafted, not just how this particular show is remembered internally.

How should a small museum run evaluation without a dedicated research staff?

Most small and mid-sized museums cannot run full formal studies, but a minimal version of each stage is still possible and still useful.

Stage	Minimal version a small team can run
Front-end	Five to ten short conversations with people representative of the expected audience, before design starts
Formative	A walk-through of mock-ups or printed draft labels with a handful of non-staff visitors, think-aloud style
Summative	Simple observation notes on where visitors stop and skip, plus a short exit question or two

The point is not statistical rigor; it is catching avoidable misunderstandings before they are expensive to fix, and keeping a written record for the next project.

What should be documented so evaluation findings are not lost after the show closes?

Keep an evaluation log attached to the exhibition file: what was tested at each stage, what was found, and what changed as a result. This log should sit alongside the object selection log described in setting object selection criteria for a show, since both exist to let a future team reconstruct why decisions were made, not just what the final result looked like. Without this record, the same comprehension problems tend to recur in the next exhibition, because the lesson lived only in the memory of staff who may have moved on by the time the next show opens.

What is the first step for a team that has never run formal evaluation before?

Start with formative evaluation on the next exhibition's labels, since it is the lowest-cost, highest-value entry point: print a handful of draft labels, stand near them during a quiet hour, and ask a few visitors what they think the label is telling them. Compare their answer to what the brief intended. That single exercise, repeated across a few labels, usually reveals more about how the exhibition will actually read than any amount of internal review, and it builds the habit that front-end and summative evaluation can later extend.