What choice architecture means for marketing

Architecture for marketing

Behavioural economics has given us the concept of choice architecture. What it means is simply that the way a choice is presented influences how it is made. Given this, choice options can be designed to influence the way people choose as we would like. Nick Southgate explains how designing choices based on this knowledge is important to marketing

When Richard Thaler and Cass Sunstein coined the term ‘choice architecture’ in Nudge, they did not intend it to become just an academic term for dry debate. They were laying down a challenge for people in positions of influence to become Choice Architects.

Thaler had already done this with the Save More Tomorrow pension in the US. Everyone knows people are bad at taking out pensions. Many of us have created advertising meant to reform the attitudes that stop people taking out pensions. People, however, often remain resolutely committed to not reforming their attitude and not changing their behaviour.

What the Save More Tomorrow pension does is remove the behavioural barriers to taking out a pension, with dramatic results.

The first piece of choice architecture is to enrol automatically any employee in their company scheme (while still allowing people the choice of opting out). When enrolment is the default choice most people enrol. One company introducing the scheme found that in the 12 months before automatic enrolment, participation was at 23 per cent. In the 12 months after, this rose to 78 per cent. Once the employees are in, employers assume they want to stay in and avoid giving people the opportunity to fall out of the scheme.

The second piece of this pension's choice architecture is far more subtle and ingenious. One of the things people worry about when they take out a pension is that their take-home pay goes down from that moment by the amount of the deduction. Although, in the long term, they know a pension will make them richer, in the here and now it makes them poorer. This immediate impoverishment crowds out any long-term gain.

The answer is to make sure people never feel this loss. How? Don't start taking money until they get a pay rise. This way their take-home pay always goes up (which they like) and the deduction is taken from ‘money they never had’ (which they don't miss).

In one study the group offered the Save More Tomorrow option increased their savings from 3.5 per cent to 13.6 per cent of pay. This contrasts with those only offered traditional advice. They averaged savings of 9 per cent of pay. Redesigning the choice made people more than 50 per cent better at saving.


We like to think that people choose the brands we advertise deliberately. In reality, people spend a lot of their time avoiding making active choices. Instead they take the choice that is no choice at all – the default choice.

Defaults, though, are not only influential with simple and mundane choices. When faced with complex decisions – whether moral, medical, commercial, social or personal – people are just as likely to look for cues to norms and typical behaviour to understand what to do. People, not unfairly, feel that typical choices reflect a collective evaluation of the facts and represent a strength of conclusion they could not reach alone.

This, for example, is why Welsh health minister Edwina Hart announced her intention in December 2009 to change legislation on organ donation in Wales. The new rules will presume that individuals are willing donors (although family will still need to give final permission).

This ‘soft’ opt-out will increase greatly the availability of organs and follow the higher patterns seen in Spain and Belgium.

People, who are broadly supportive of organ donation in principle still stumble at choosing it, maybe feeling they could have misunderstood complex issues. A default option helps them accept a situation with confidence.

Similarly, when faced with difficult trade-offs between price plans or offers, consumers are hugely reassured if there is a guarantee that their option will default to the cheapest available in the absence of any decision. They do not want to feel that defaults will work to cost them money. People look to them as safety nets.

Two things are worth noting here. First, no choice architecture is fixed; people can and do learn how to interpret, resist and play systems. If they feel the default choice is wrong, they can change their behaviour. For example, how many people have learnt to be more wary of the default choices on software, having inadvertently installed an unwanted toolbar?

Second, default choices are strong but not insurmountable. A brand seen to abuse such default choice will be punished in the long term by some, or all, of their customers. Low-cost airlines are prime examples. A growing constituency of people are unwilling to comply with the disadvantageous revenue-boosting charges low-cost airlines impose on default behaviours – for example, fees for using credit and debit cards.

While some may enjoy working out how to get the lowest possible fare, other people feel that these are charges for ‘doing what is normal’ and seek to avoid this way of doing business altogether.


Many globally successful brands have thrived by limiting their offer to a single product, often in a single format. Take WD-40, Red Bull or Angostura Bitters, for example. These brands are strong because they enforce simple choice architectures. Such single-mindedness also focuses the businesses’ support of the brand.

Likewise, McDonald's shrunk the traditionally labyrinthine American diner menu down to only four or five items. The previous customer benefit of choice and food cooked to order became unnecessary and was replaced with an emphasis on speed. McDonald's global footprint is testament to how doing a few things reliably and predictably can trump previously dominant category values.

This wisdom points to a clear area of application for choice architecture – the analysis and desirability of brand extensions. When brands proliferate choice they risk the adverse cost of presenting a poorly conceived choice architecture.

Understanding this as a cost on a brand acts as a useful counter to the belief that brand extensions are a ‘free’ way of exploiting a brand and increasing its profitability. The costs of brand extension are well known; they risk switching people to the new variant, encouraging cannibalisation, or confusing people to the point where they switch away from the brand altogether.

Consider, for example, a food brand that has an original product high in fat. In response to the consumer desire for a healthier lifestyle the brand introduces a low-fat variant. The brand hopes to benefit from increased sales as a result of increased consumer choice.

However, it may have failed to factor in a parallel, widely held consumer belief that low-fat variants may not taste as good. The brand's choice architecture now contains two products that both say unflattering things about each other – the new variant says the original is unhealthy; the original reminds consumers that the new variant will not taste as good.

The company behind the brand should not be surprised if the net public regard for the brand diminishes overall, and some consumers opt out altogether. Consumers react not only to the new product but also to the new information about the relative values at play in the market.


Behavioural economics provides many examples of the impact of extra choice on how people choose. The most celebrated example is that produced by Dan Ariely and based on a subscription offer to The Economist. However, this example needs careful consideration before it is applied to marketing and advertising. The simple lessons it appears to offer are, in fact, complex and therefore worth examining at some length. Here is a recap of the experiment.

Faced with a two-way choice between an online subscription for $59 and a print and online subscription for $125, 68 per cent of people chose the cheaper offer of online-only. However, when a third option was introduced, the outcome changed. The third offer was a printonly subscription for $125. Needless to say, no one chose this offer. Everyone willing to spend $125 wanted to get the now ‘free’ online subscription.

What is counter-intuitive, at first, is that 84 per cent of people chose this option and only 16 per cent chose the online-only subscription. The third choice has the effect of reframing the others. For The Economist, this means the total value of subscriptions from 100 subscribers rises from $8,012 to $11,444 – an increase of 43 per cent.

What does this mean for the consumer? Presumably they are happy with their choices in both scenarios. They have chosen a subscription which represents good value for them. As they are asked to, people try to make the ‘best’ decision.

In the first set-up, most people choose to reduce the cost of the subscription by altering the media in which they consume it (getting the lowest price is considered the ‘best’ option).

In the second set-up most people pursue value for money, choosing the extra value of having two ways to access the magazine (getting more at a higher price is now the ‘best’ option).


Ariely calls these extra choices on the subscription form ‘dummy choices'. The phrase suggests the choice should have no real role and carries a hint that considering it is stupid.

This language is typical of the popularising texts of behavioural economics. There is a tendency to emphasise the ‘irrational’ in our behaviour and to frame our choices as wild and unexpected. It adds to the drama and theatre of behavioural economics, amplifying effects and emphasising just how much small changes can make big differences.

The truth for the serious marketing practitioner lies in the reverse direction when we consider how sensible and expected these decisions are.

Decisions can only be described as rational by prescribing a framework with its own set of presumptions that assumes a single optimal response. When people choose a non-optimal outcome they fall short of this rational model.

However, this does not make them ‘irrational'. It is as likely to suggest that the framework and presumptions used to understand the decision do not tell the whole story. In Ariely's example, the ‘dummy’ choice moved people from a ‘money-saving’ frame to a ‘valuemaximising’ frame for their subscription choice when they looked for the ‘best’ answer. Both are rational within their own terms of reference.

Many retail outlets are designed to create closed choice architectures as far as possible. Car dealerships usually only sell one brand of car, despite the fact that the consumer might find a stockist of multiple brands of a particular format (such as estate cars) more useful, as they do when buying used cars. As a result, most car buyers expect to visit more than one dealership.

Likewise, in the days of the brewers’ ‘tied estates’ their pubs typically only provided a single brand of stout, cider or lager – and at one time they excluded real ale altogether. This practice so offended beer drinkers that they formed the Campaign for Real Ale and their vigorous lobbying persuaded many pubs to carry ‘guest’ ales.

The popularity of shopping malls is, in part, because they allow people to move easily between retail outlets devoted to individual brands, especially in the clothing sector. Although we can work to optimise choice architecture within our own brand, it is more difficult for choice architecture, on its own, to influence how consumers relate to competing brands.

This is where the art of positioning kicks in. Although brand owners can act as choice architects, they need to remember that people are not obliged to choose within the architecture they present. They can simply choose to shop elsewhere.


There are fine examples of strong brands built around business models that enshrine choice architectures. At launch Daewoo promised buyers that its showrooms would have fixed prices and no salesmen. For people for whom the rigmarole of car purchase with pushy salesmen and haggled prices was a deterrent, this new form of choice architecture in the car market was powerfully attractive.

Direct Line promised customers a better price by cutting out the middle man, but at the same time it only sold its own products. What consumers saw, and still see, is a clear choice architecture that promises value while it rejects the traditional broker's promise of finding the perfect product through wide choice.

The easyJet approach, where you can bypass the travel agent – but only by visiting a site which sells only easyJet flights – is another example of people forsaking choice willingly in return for some other incentive. So, too, are schemes such as Amazon Prime and Ocado on Demand, where customers pay the company upfront for a free-delivery saving which will pay back only if they use the service heavily.

Argos presents a choice architecture that removes many of the supposed advantages of the shopping experience – no fancy display, no helpful floor staff and no direct interaction with products. Choice is limited to what can be seen in the catalogue or online. This apparently austere presentation of stock remains one of Britain's favourite ways to buy goods as diverse as jewellery, lawnmowers, computer games and sporting goods.

What Argos customers see and value is a choice architecture that prioritises what they find important in making a purchase.

The appeal of John Lewis's promise of ‘Never knowingly undersold’ is another example of a different choice architecture. When people decide to trust John Lewis they outsource all the stress and hard work of choosing which goods (white goods, small electrical goods, bed linens and cookware etc) are worth buying.

The presence of goods in John Lewis is taken as a promise of high quality and excellent value.


Architectures also explain one of the great phenomena of marketing in recent years – the growth of the price comparison site. These sites have produced some of the most liked ( and most disliked ( campaigns of 2009.

The online and digital environment means consumers demand the ability to choose between offers and brands. This creates a series of brands that exist only to enable these comparisons (bewilderingly, some are now offering to compare the comparisons).

The real power of choice architecture is not its ability to coerce but its ability to persuade people that this is the best way to choose. The marketer's skills are tested when positioning a brand or offer within the context the consumer experiences it – and making that brand stand out in the consumer's own terms.


Some car brands are better at choice architecture than others. The BMW numeration system is clear, as is the relative size of the various Fords. Both help buyers negotiate important variables – be they status or function.

In comparison the numeration of Volvos, Audis and Peugeots is opaque, communicating more to service mechanics and engineers than to simple buyers.

Nick Southgate is a freelance consultant.

[email protected]

This is adapted from an IPA brochure, We're All Choice Architects Now, 2nd edition, June 2010