What Failure Can Tell Us

Suppose the Foundation puts 60 tickets on sale for a film screening.

Eight people buy them.

That looks like failure.

Those are very different conclusions.

And eight ticket sales cannot tell us which one is true.

That distinction matters if we are using arts programs as part of the Cultural Interest Study.

We should expect some things not to work

The purpose of the Cultural Interest Study is to learn what Duncanville residents will buy.

If every test succeeds, we probably are not testing very much.

That is not a problem with the study.

That is part of the study.

If we only produce things we already know people will buy, we learn very little about the edges of the market.

Research requires uncertainty.

We try something because we do not know exactly what will happen.

Then we watch.

A bad result can have many causes

This is where we need to be careful.

Low attendance does not automatically mean there was no interest.

National Endowment for the Arts research has found several reasons people may want to attend an arts event and still not go. Those barriers include lack of time, cost, difficulty getting to the location, and not having someone to attend with.

That means an empty seat can have several explanations.

Imagine we offer a jazz performance for $35 on a Tuesday night.

Only 12 people buy tickets.

We could conclude:

Duncanville does not want jazz.

But we would be making a much larger claim than the evidence supports.

What we actually know is:

That is a much more accurate statement.

And it gives us somewhere to go next.

Change the question before changing the conclusion

Suppose we try jazz again.

This time the ticket is $20.

The location stays the same.

The day stays the same.

Thirty people buy.

Now we have learned something.

Maybe price mattered.

So we test again.

Or perhaps we keep the $35 ticket but move the performance to Saturday.

Sales increase.

Now timing may matter.

Or we keep the price and date but move the performance to a different part of town.

Sales change again.

Each test helps us separate one question from another.

The National Institute of Standards and Technology describes experimental design as deliberately changing factors so we can observe how those changes affect the result.

The Cultural Interest Study does not need to turn every arts event into a laboratory experiment.

But the basic idea is useful.

If we change everything at once, we may learn nothing.

We need to know what we are testing

This means every Cultural Interest Study activation should begin with a question.

Not:

Let’s have a concert.

Instead:

Will residents buy 50 tickets to this kind of music at $20?

Not:

Let’s have an art exhibition.

Instead:

Will visitors purchase original work priced between $100 and $500?

Not:

Let’s offer a workshop.

Instead:

Will 20 people pay $35 for a two-hour class on a Saturday afternoon?

Those questions give us something we can measure.

Then, when the program ends, we have a result we can compare with the question.

The Centers for Disease Control and Prevention’s 2024 Program Evaluation Framework makes the same basic point in a much broader setting: evaluation should begin with context and clear questions, gather credible evidence, support its conclusions with that evidence, and then use the findings to make decisions.

That process matters just as much when the answer is no.

Eight tickets are still eight purchases

Return to our film screening.

We offered 60 tickets.

Eight sold.

It would be easy to focus on the 52 tickets that did not sell.

But the eight that did sell may be more interesting.

The event may have been weak as a business result.

It may still produce useful research.

Those eight buyers are evidence.

So are the 52 unsold seats.

The important thing is not to make either group say more than the evidence allows.

Failure can save money

There is another reason weak results matter.

They can stop us from making bigger mistakes.

Imagine we think Duncanville needs a permanent film theater.

That could require a building, equipment, staff, insurance, utilities, marketing, and years of operating costs.

Before making that investment, we could test film demand in temporary spaces.

Those results could support a case for doing more.

Or maybe repeated screenings struggle even after we test different films, prices, dates, and locations.

That finding has value, too.

A few inexpensive tests might keep us from making a much larger investment in an idea the market does not support.

We should not rescue every idea

This may be one of the harder parts of the study.

When we care about an idea, we naturally want it to succeed.

If ticket sales are weak, we can lower the price.

If that does not work, we can give tickets away.

We can spend more on advertising.

We can call friends.

We can find sponsors.

We can move the event.

We can keep changing things until the room finally fills.

Then we can call it a success.

But somewhere along the way, we may stop doing research.

The Cultural Interest Study has to be willing to hear no.

That last answer is useful, too.

One failure should not decide the market

The opposite mistake is just as dangerous.

One poorly attended event should not become proof that an entire art form has no market.

One concert cannot tell us whether Duncanville supports live music.

One exhibition cannot tell us whether residents buy art.

One workshop cannot tell us whether people will pay for arts education.

The result belongs to the conditions under which we produced it.

That is why patterns matter.

If we test something several ways and continue seeing weak demand, our confidence in the finding grows.

If changing one factor produces a very different result, we learn something else.

Over time, separate tests begin to form evidence.

That is what we are looking for.

Success means learning

This requires a different way of talking about the Foundation’s work.

A sold-out event can be successful.

So can an event that sells half its tickets.

Under the right conditions, an event that sells almost none may still be useful.

The question is whether we learned what we set out to learn.

If the answer is yes, the research worked.

Follow the evidence

The Cultural Interest Study is not supposed to prove that Duncanville needs more of everything.

It is supposed to help us find out what people will support.

All of those answers have value.

Because the Foundation’s job is not to make every idea succeed.

It is to learn enough to know which ideas deserve the next investment.

A weak result is not something we need to hide.

It is something we need to understand.

Subscribe to Ron Thompson’s Field Notes

Occasional notes on art, culture, creative economies, and the work of building cultural capacity from the ground up.

Subscribe to Field Notes

By sending the subscription request, you agree to receive Ron Thompson’s Field Notes by email. Your email address will be used to manage and deliver your subscription and will not be sold. You may unsubscribe at any time by replying “unsubscribe.”