Understanding the power of the Goldilocks Effect

I have a passion for design patterns. I like to understand the reasons why certain patterns work and dislike using conventions without knowing why they have become successful.

Recently I’ve found myself interested in the psychology of persuasion and the ways in which certain techniques have infiltrated the web. One such technique is the “Goldilocks Effect”. I’ve been aware of it for a while but had only ever thought of it as just another design pattern and hadn’t necessarily thought to understand the psychology behind it.

What is the Goldilocks Effect?

The term ‘goldilocks effect’ derives from the children’s story “Goldilocks and the Three Bears”. In the story Goldilocks decides, amongst other things, to eat one of three bowls of porridge; the first being too hot, the next too cold, but the final one she picks for being “just right”.

This, in a nutshell, is what the term goldilocks effect or ‘goldilocks pricing’ refers to. It is used to describe the practice of providing a premium as well as a budget option alongside the regularly priced product to make the standard option seem more appealing. A good example of this is the basic pricing structure adopted by most airlines. They encourage customers to see business class as good value by offering higher priced first class and economy options alongside it. A more everyday example can be seen in all mainstream coffee shops where the options range from small (tall) to large (venti), with regular (grande) in-between.  The goal of this type of pricing  is to push people who might usually buy the cheapest into buying a more expensive option.

Media Temple

“Third-class railway carriages in Victorian England are said to have been built without windows, not so much to punish third-class customers (for which there was no economic incentive), as to motivate those who could afford second-class seats to pay for them instead of taking the cheaper option.” Wikipedia

This technique for encouraging a more favourable transaction exploits our psychological aversion to extremes (a type of cognitive bias). Basically you can manipulate people into choosing the option that yields the greatest profit by providing them with three options (e.g. small, medium, large) as long as the item you wish to sell the most of is centered within the range.

Interestingly some economists have argued that the goldilocks effect constitutes a form of ‘pricing discrimination’ where premium pricing to encourage people to upgrade their purchases leads to providers intentionally worsening the quality of their ‘basic’ products to fit the model.

Odeon set 3 options across all audience types

I witnessed the goldilocks effect in action last year when user testing a travel insurance website. Following the quote process the participants where presented with 3 options, and even though most of them had mentioned price as a key factor in their decision making when purchasing insurance the vast majority commented that, even though there was very little difference between the products, they were more likely to purchase standard or premium over the cheapest option as they felt that by doing so they would be “even more insured”. Interestingly some users went as far as to show distrust, not of the brand, but of the lesser option.

Three is the magic number

Although in the children’s story Goldilocks is confronted by three options the pricing technique doesn’t necessarily have to be limited to that number. It does, however, make sense to limit the options as there’s less chance of overwhelming the customer with choice while also providing just enough for the technique to achieve its desired result.

It is possible to offer too much choice

Goldilocks and the four bears

While looking for examples I found a couple of instances were four products were shown. Rather than just show each option equally alongside the next, one of the central items would be highlighted as the ideal or ‘most popular’ choice. By visually separating out one of the central options they reduce the chance of choice paralysis. I accept that choosing from only four options is not exactly excessive but I do believe it has had enough of an effect for the companies to feel it necessary to highlight one.

Basecamp, a good example of exceeding the ‘magic number’

How to use the Goldilocks Effect successfully

So how can you utilise our natural aversion to extremes to encourage a transaction? I don’t consider myself an expert in this but I thought I’d share some of the ways I think you can use the goldilocks effect.

  1. Limit the number of options – ideally keep it limited to 3, although more can clearly work, too many could cause choice paralysis.
  2. Clearly show the comparison between each product – make the users choice clear by placing each option side by side
  3. Highlight the similarities and differences between the features and benefits – make sure the differences between each option is clearly displayed; what are the features or benefits? How is the higher price justified?
  4. Ensure product labels follow a consistent theme – SurveyMonkey for example use basic, pro and unlimited. Getsignoff uses audience types; freelancer, team or agency. With Getsignoff there is in-fact a forth option (free) but they choose not to place this alongside the paid for services.
  5. Use social proof to support the sell – as shown in the Basecamp example above highlight which option is the ‘most popular’ with your customers. By emphasising the choice made by the majority of people it reinforces that the middle option is the best, if everyone else is choosing it, it must be right.

I hope you’ve found this introduction useful. If you’d like to see more examples of the goldilocks effect in action I’ve put together a collection which I’ll be updating so If you know of any other examples please feel free to share them.

Update 27/10/2010

I’ve put together a collection of examples of the goldilocks pricing pattern on UI-Patterns.com so please check it out.


11 Principles of Interaction Design explained

This post isn’t intended to be an exhaustive list of interaction design principles, its merely an introduction to the subject. And I’m definitely not going to attempt to enter the lions den of defining what ‘interaction design’ is, that’s for another day!

1. Match Experience & Expectations

When using a product or service for the first time there is likely to be an element of learning needed to get to grips with it. This learning curve can often be an uncomfortable experience especially if the proposition doesn’t feel familiar.

By matching the sequence of steps, layout of information and terminology used with the expectations and prior experiences of the user the friction and discomfort of learning a new system will be reduced.

Matching your audience’s prior experiences and expectations is achieved by using common conventions or UI patterns. Des Traynor and Eoghan McCabe from Contrast put together a great presentation for FOWA Dublin last year all about conventions which is well worth a look. It introduces the importance of conventions but more importantly highlights when they should be broken.

2. Consistency

As well as matching peoples expectations through terminology, layout and interactions the way in which they are used should be consistent throughout the process and between related applications.

By maintaining consistency users learn more quickly, this can be achieved by re-applying in one part of the application their prior experiences from another.

An added bonus of keeping elements consistent is that you can then use inconsistency to indicate to users where things do not work the way they might expect. Breaking consistency is similar to knowing when to be unconventional as mentioned above.

3. Functional Minimalism

“Everything should be made as simple as possible, but no simpler.” Albert Einstein

The range of possible actions should be no more than is absolutely necessary. Providing too many options can detract from the primary functions and reduce usability by overwhelming the user with choices. To achieve the zen of ‘functional minimalism’:

  • Avoid unnecessary features and functions
  • Break complex tasks into manageable sub-tasks
  • Limit functions rather than the user experience

4. Cognitive load

Cognition is the scientific term for the “process of thought”. When designing interactions we need to minimise the amount of “thinking work” required to complete a particular task. Another way of putting it is that a good assistant uses their skills to help the master focus on their skills.

We need to understand how much concentration the task requires to complete it and create a user interface that reduces cognitive load as much as possible. A good way to reduce the amount of ‘thinking work’ the user has to do is to focus on what the computer is good at and build a system that uses the computers skills to the best of its abilities. Remember that computer are good at:

  • Maths
  • Remembering things
  • Keeping track of things
  • Comparing things
  • Spell Checking and spotting/correcting errors

5. Engagement

In User Experience terms engagement measures the extent to which a consumer has a meaningful experience. An engaging experience is not only more enjoyable, but also easier and more productive. As with many things engagement is subjective so the system your designing must engage with the desired audience; what appeals to a teenager is not necessarily what their grandparent would also find engaging. Beyond aligning with the appropriate users, control achievement and creation are key.

The user should feel like they are in control of the experience at all times, they must constantly feel like they’re achieving something and also be able see the results through positive feedback or alternatively feel like they’ve created something.

In his book ‘Flow’, Mihaly Csikszentmihalyi describes a state of optimal experience, where people are so engaged in the activity they’re doing that the rest of the world falls away. Flow is what we’re looking to achieve through engaging interactions. We should allow users to concentrate on their work, not on the user interface. In short keep out of the way!

6. Functional Layering

The Pareto principle (also known as the 80-20 rule), in the context of interaction design, is the rule that 20% of the functionality is used 80% of the time. Therefore we should make the most common or important functions easiest to find. We can do this by hiding or reducing the prominence of infrequently used or advanced functions.

  • Look for functions that are not essential to the core tasks, or which are shortcuts for advanced users.
  • Consider introducing default settings and preset choices for new users or people who either don’t wish to or aren’t experienced enough to access advanced functionality.

Functional layering allows experienced users to access advanced functionality easily without hindering beginners. The idea is that as a user progresses they will naturally discover how to access such features without being overwhelmed early on.

7. Control, Trust & Explorability

These three elements are fundamentally important to any system. If users feel in control of the process they will be more comfortable using the system. If the user is comfortable and in control they will trust that the system will protect them from making unrecoverable or unrecognised errors or from feeling stupid. Trust inspires confidence and with confidence the user is free to explore further.

8. Error Prevention, Detection & Recovery

The best way to reduce the amount of errors a user makes is to anticipate possible mistakes and prevent them from happening in the first place. If the errors are unavoidable we need to make them easy to spot and help the user to recover from them quickly and without unnecessary friction.

Error Prevention
Prevent errors by:

  • Disabling functions that aren’t relevant to the user
  • Using appropriate controls to constrain inputs (e.g. radio buttons, dropdowns)
  • Providing descriptive, clear instructions and considering preemptive help
  • As a last resort provide clear warning messages

Error Detection
Anticipate possible errors and provide feedback that helps users verify that:

  • They’ve done what they intended to do
  • What they intended to do was correct

Its important to remember that providing feedback by changing the visual state of an object or item is more noticeable than a written message.

Error Recovery
If the error is unavoidable provide clearly marked ways for the user to recover from it. For example provide “back”, “undo” or “cancel” commands.

If a specific action is irreversible it should be classed as critical and you should make the user  confirm first in order to prevent slip ups. Alternatively you can create a system that naturally defaults to a less harmful state. For example if I close a document without saving it the system should be intelligent enough to know that it is unlikely that I intended the action and therefore either auto-save or clearly warn me before closing.

9. Mousing

In my daily interactions ‘mousing’ is becoming less of an issue as I begin to rely more on touch screen interfaces such as my iPhone and iPad. However, in the classic desktop environment ‘mousing’ relates to the ease in which you are able to move between controls, which is described best through Fitts’ law.

“The time required to click an object is proportional to the distance and inversely proportional to the object size.” Fitts’ law

Fitts’ law is a model of human movement in human-computer interaction (HCI) and ergonomics which predicts that the time required to click an object is proportional to the distance and inversely proportional to the object size.

With key functions or sequential mouse-operated controls we need to maximise the size of the controls and minimise the distance between them. This not only improves efficiency but in certain instances can reduce the risk of error.

  • Consider what the most common or typical mouse movements will be on each screen.
  • Where possible, place elements that will be used together in close proximity to each other.
  • Be particularly conscious of transactions that require a combination of mouse and keyboard controls. Consider ways to optimise for both styles.

10. Affordance

Affordance is the quality of an object that allows an individual to perform an action, for example a standard household light switch has good ‘affordance’, in that it appears innately clickable. In short the physical properties of an object should suggest how it can be used. In the context of user interfaces, affordance can be achieved by:

  • Simulating ‘physical world’ affordances e.g. buttons or switches
  • or keeping consistency with modern web standards or other interface elements e.g. underlined links or default button styles.

11. Hierarchy of Control

The hierarchy of influence between elements should be clearly apparent. Generally, controls which affect an object, should be grouped with the object, such as zoom controls on a map.

Controls which influence a group of objects should be associated with the entire group, forming a hierarchy.

Mocked up example showing hierarchies of control