Wednesday, February 23, 2011

Understanding suspense by looking at films

I think one of the best ways to understand suspense and expectation is with films. When the trailer for a film comes out, we have a peak of suspense because suddenly we are anticipating the experience of the film. This suspense spike lasts for a short while, and then goes dormant – until the film starts playing TV spots and other advertisements. We go to the see the film – and our suspense is the highest right before the film starts. From then on, we are now in the experience that we were anticipating, and based on our expectations, we will leave surprisedsatisfied, orsacrificed.

Suspense Curve with Trailers and Films
This above diagram somewhat sums up what we experience for most films. But there are times when quite drastically, the trailer created more expectation than the film delivered, and we have sacrifice – the classic line being “all the best bits were in the trailer!”
This lesson of “the best bits were in the trailer” is first lesson in expectation management: what is your end?
In order to know what you have to give, you must take stock of what you’ve got. What is the sum total that your customer leaves with? Equally, you need to know what you will receive. You’ll see why in a moment.
Consider the following three scenarios:



1. A B-grade movie knows that it has a cheap storyline to give, and short-term ticket sales to receive. Given that its delivery is weak, it front-loads the best scenes into the trailer on purpose, to create expectations (suspense) that it knows it can’t meet, in order to get people in the cinema so the studio can receive its money. Important: even when they give their end away at the beginning, people still come thinking there is more. No matter how much you give away up front – people will always think there is more, even if there isn’t. So the studio makes lots on ticket receipts over a short period of time. They don’t even think about getting long-term revenue here. This is sacrifice.
2. Apple know that they have a great, game changing product to give, and years of sales and increased market share to receive. Therefore they release information to create a lot of expectation and buzz, but not more than they can deliver on. There’s the suspense. Then, they back-load the expectations. This means they keep the best stuff hidden They know that if they exceed expectations and surprise, they have done well, and even if they just meet expectations and satisfy, they have a very loyal fan base to receive sales from. They are actually more concerned with creating lots of suspense at the beginning to get people talking, than delivering a surprise on that suspense. This is surprise, or satisfaction.
A new café opens in town that already has a number of cafés. They know that they have a ‘newness’ factor, a unique selling point, and high standards from newly trained staff to give, and sales, a customer base, and permission to build long term loyalty with those customers, to receive. Given that they are new in town, they must not only satisfy expectations, but exceed them (surprise), to get customers to return and prefer them over their competitor. The first months are critical for them to continually surprise in order to lock in long term loyalty. This is surprise.
All this might seem like common sense, and it is. What each scenario highlights, however, is thatonce you know your end, and what you will receive, you know how much of the end to show, to get what you want to receive. A general rule of thumb is:
  • If receiving short-term gain, front-load the expectations, in order to get numbers and sales in a spike.
  • If receiving long-term gain, back-load the expectations, so that your over delivery gets people back for the next thing you do.
If you find that you’re running events, for example, but no one is coming back – then this may well be why. Why should they come back if you only satisfied them but didn’t surprise them? Given all the competition, if you only satisfy, then you lack uniqueness.
This is only a small part of expectation management, but it’s the start. Below I’ve given you a summary, and have asked specific questions from you. I’ll answer every question you ask to help you get better at this.

Actionable Summary

  1. Work out the sum total of what you have to give. This is called ‘the end‘.
  2. If you want to get short term gain, give away the end up front.
  3. If you want long term gain, give some of the end up front, but keep either a good portion, or the best portion, stored up.

Leading Questions

  • Do you identify where you are failing, because you are not managing expectations?
  • Does this model work with content marketing online? I’ve got a hunch that it does, but keen to hear your experience.
  • Have you proven this model wrong? Are you having long term success by front-loading experience?

No comments:

Post a Comment