Services As Structure

Legendary film director Howard Hawks once said that to make a great movie all you need are three great scenes and no bad scenes. It’s a pretty simple formula for success. What if we looked at service design through the same lens? It’s not a perfect analogy, but each service encounter is composed of touchpoints. Is it enough to create three great touchpoints and no bad touchpoints? It might not be a bad place to start.

Here’s another parallel from the world of screenwriting:

The first fifteen pages are the most important of any screenplay… and the final fifteen minutes are the most important of any movie.

This corresponds to what James Heskett calls “service bookends” in his book on Service Breakthroughs:

The stage for the service experience is set in the first few minutes of the situation. Once the tone has been established, it is difficult to change a customer’s impression of what follows.

Last impressions count too. That last few minutes of a service experience may cement the final impression of the event, which influences a customer’s willingness to make a repeat purchase or provide positive “word-of-mouth” selling to a potential customers. These first and last parts of the service encounter are the “service bookends.”

The final structural analogy is based on the peak-end rule. Basically that the best or worst part of an experience and the end of that experience are what influence our perception of an event. From a screenwriting perspective it’s apparently critical to minimize the time between the peak and the end, otherwise the audience gets bored. One of the best examples of trimming down that gap is Hitchcock’s North by Northwest, where the climax of the film happens a mere 43 seconds before the end of the picture. In less than a minute Hitchcock still manages to wrap up seven different plot points.

Retail could certainly learn from this. Finding a product to buy can be a great experience, but it’s always followed by an interminable wait in line to pay. Same problem with restaurants. Not only is the end of the experience (checkout/the bill) lackluster, but the long wait gives the customer too much time to recover before wrapping things up. How anticlimactic.

In William Goldman’s book Adventures in the Screen Trade (from which the 15 minute rule and the Hitchcock example are drawn) his central thesis is that screenplays are all about structure. Understanding some of the subtleties of screenwriting structure, such as the storytelling balance between acts and the structural differences between one act, two act and three act forms could prove to be fertile ground for service design.


  1. Great parallel. In my thesis paper I parallel the importance of first impressions in music with the same importance in service design. I’ve read that in Marriott’s customer surveys, four of the top five factors that contribute to customer loyalty occur in the first ten minutes of interaction with their hotel.

    … I never considered last impressions though, but maybe I’ll include it in my paper since it seems to make a lot of sense!

    I wonder with the peak-end rule applied to services, if minimizing the time between the peak and the end would work all the time… or if designing for that space between the peak and the end to be enjoyable would make up for the fact that most of the time it’s a long and horrible process.

  2. Jeff

    Marriott applies this principle to the end of the service by providing express checkout. You wake up with your receipt slipped under your door and on your way out you just toss your key on the front desk. Flawless!

    The last 15 minute quote is attributed to Paul Newman, and I think from a film perspective it’s probably more important than most services because people don’t have to stick around for the end of the movie. If they get bored, they just walk out.

    There’s a little more about the peak-end rule over on the Experience Matters Blog that talks about exploiting this structure to offset negative experiences.

    I think it’s close to what you’re suggesting.

  3. Hey Jeff

    I just completed a blog entry on the exact same concept (see url). I didn’t punctuate it with concrete examples, but I am convinced that this was of thinking can help in service design quite a bit. I am trying to implement it in one my current projects.

    Thanks for the examples!

  4. Kip

    Jeff and Carrie, thanks for the tidbit. I actually work for Marriott now so it’s good to see there’s a lot to learn from this hospitality industry.
    I actually just blogged on my pizza ordering experience with domino.com. It seems like a lot of people have, but yesterday night was my first experience with it. I really like how they took transaction to another level because it didn’t end my “facial engagement” in the Goffman sense.
    I don’t know if they made a conscious deliberation during the creation of the “Pizza Tracker,” but someone at dominos (or a vendor that worked with them) that the pizza ordering experience does not have to end with a credit card transaction … the key moment is actually getting the pizza and enjoying it! If you haven’t tried it yet, make it a dominos night and try it yourself 🙂 It was a lot of fun and felt more like “an experience” than just another dinner order.




Leave a comment