7.3 Consumer Decision Making


Utility theory, satisficing and Prospect theory
1. Utility theory by Neumann and Morgenstern (1944)
The utility theory suggests that consumers make rational (logical) decisions based on the likely outcomes of their actions. In other words, consumers make decisions to purchase products based on how they utilise each product.
Example 1: if buying milk, the likely outcome is product usage. So in this case, it's drinking milk.
Example 2: If purchasing a car, the utility theory predicts that customers evaluate every car against all relevant variables (cost, size, make, etc.) and select the one that scored highest on all these variables.

It's unlikely that customers are as rational as the model predicts. Further, customers may not even be aware of the decision-making process.

2. Satisficing model by Simon (1956)
Consumers get approximately where they want to go and then stop the decision-making process.
Car example: You look at a few cars and stop the process when you find one that is good enough.

Although the model explains behaviour better than the utility theory, it's not a good predictor of consumer behaviour in a retail environment.

3. Prospect theory by Kahneman and Tversky (1970s)
This theory is based on value and endowment. This means that people are more likely to base decisions on the perceived likelihood of gains rather than losses.
Example: We buy a lottery ticket because we might win, not because we are highly likely to lose.

Decision-making strategies
1. Compensatory strategies
Consumers allow the value of one attribute to compensate for another.
Example: The right colour choice compensates for the fact that it is not the desired make.

Richarme: attributes of the decision-making process may be of equal weight or different weights.
Equal Weight Strategy: when one attribute equally compensates for the other.
Weighted Additive Strategy: For example, the right colour is far more important than having the right make.

2. Non-Compensatory strategies: Each attribute is individually evaluated.
There are 3 types:

  • Satisficing — the first product to meet basic requirements is chosen without further consideration.
    Ex. If an essential household item such as a kettle fails, one might simply buy the first kettle they see.
  • Elimination by Aspects — a cut-off value is set for the most important variable and everything meeting it is under consideration. The remaining items are assessed against the next variable.
    In other words, Elimination by Aspects includes starting at the most important attribute and then progressively eliminating options.
  • Lexicographic — looking for the best product on the most important attribute.

3. Partially Compensatory strategies

  • Majority of Conforming Decisions — 2 products are evaluated against all relevant attributes and the best is retained. The one retained is compared to the next one and so on until the best product remains.
  • Frequency of Good & Bad features — products are simultaneously compared to the cut-off values for their attributes. The product which exceeds the cut-off value remains.

Marketing Theories
Consideration and InvolvementRicharme

  • Consideration is forming an initial subset of items to consider.
    For example, we may be able to name a great many makes of cars but when we start considering purchasing a car, the shortlist will be a much smaller subset of this depending on the cars that are affordable, and on sale nearby.
    We are focusing our cognitive effort on the cars that belong to the final subset that might be considered.
  • Involvement relates to the amount of cognitive effort applied to decision-making processes. This is directly proportional to the importance of the decision.

    The above 2 marketing theories further explain consumer decision-making strategies.

Application to real-life
Richarme: The theory is useful for researchers who want to know the different variables affecting consumers' decision-making processes. Further, retailers can understand and manipulate the decision-making process.

Retailers can also try to enhance the shopping experience for their customers by allowing them to directly compare products on shopping websites.


  1. Application: Jedelski et al.'s (2002) study aimed to investigate whether websites allowing easy product comparison would lead to better decisions.
    Procedure: Participants were briefed on the difference between compensatory and non-compensatory strategies, and gave them 2 websites to purchase items from.
    • Website 1 allowed side-by-side comparisons of items in line with their chosen most important attribute.
    • Website 2 gave the same information, but in a manner which made comparison difficult.
    Results: Compensatory strategies were used more often with Website 1 than 2.
    Conclusion: Understanding decision-making strategies improve decision-making strategies.
  2. Utility theory assumes all consumers are rational decision-makers, but this is not true.
  3. Individual Debate
    The strategy for decision-making can vary according to individual preferences.
  4. Situational Debate
    Different strategies are used for purchasing different items. Compensatory strategies for household purchases not for luxury items.

Choice Heuristics
Availability, Representativeness
Heuristics = mental shortcuts which help make decisions quickly without spending a lot of time researching information.

2 Heuristics:
1) Availability Heuristics
2) Representativeness Heuristics

  1. Availability Heuristics - cognitive bias where decisions are based on an example, information, recent experience, or hearsay that is readily available to you, even though it's not the best example to form your decision on.
    Hoyer’s Study showed that participants who read bad reviews of an appliance breakdown gave higher estimates of it breaking down than participants who were given statistics referring to the actual breakdown rate.
  2. Representativeness Heuristics - a mental shortcut that helps make judgments by comparing a current situation to a representative example.
    Manufacturers make use of the fact that customers use this heuristic, by making products look like an established leading brand in the market, so that consumer assumes they are of similar product quality.

Anchoring and Purchase Quantity Decisions
Wansink et al. examined the factors which might influence how many units of a product. Prior research suggested how point-of-purchase promotions could increase sales. The paper reported on 2 field and 2 lab experiments which showed that anchor point promotions (presented as multiple unit prices, purchase quantity limits, and suggestive selling) can increase purchase quantity.

Anchoring (a cognitive bias) is when people rely too much on pre-existing information, or on the first piece of information offered to them, when making decisions.

Experiment 1: A field experiment which compared multiple-unit and single-unit promotional pricing in 86 stores. They were randomly assigned to either the 'single-unit' or 'multiple-unit' promotion conditions.

A baseline score was calculated during the previous 6 months to note as the avg. weekly sales (there with no promotions during this period). The DV was the change in sales (in percentage) compared to this baseline.

The same shelf label size was used in all stores. The shelf displayed the original price, as well as the single-unit promotion price (75 cents) or the multiple-unit price (2 for $1.50). 13 items were included in the experiment, including cookies, candy, and cereals.

The results showed that multiple-unit promotional prices resulted in a 32% increase in sales over the single-unit control. A meta-analysis of all 13 tests indicates that the multiple-unit pricing effect is highly reliable (p<0.0001). Meta-analysis means when researchers combine findings from multiple studies to draw an overall conclusion.

The authors, however, were cautious when drawing conclusions. They were aware that consumer confusion might interfere with results. The confusion was that, consumers might have thought that they had to purchase multiple items in order to get the promotional price. Experiment 1 did not collect any self-report data, so it’s impossible to know if this was the cause, although they suggest this is unlikely to explain all of the increased sales.

The researchers also note that they have no way of knowing whether the increased sales were due to similar number of customers purchasing more items, or due to an increased number of customers.

Experiment 2: A field experiment looking at the effect of anchors which limit purchase quantities such as ‘Limit 4 per customers’. Research evidence showed that restricting customers to one item makes the deal look good, and that having low limits for purchasing a product seem to increase sales. Wansink et al. wanted to look at the effect of high purchase decisions, such as 12 items per customer.

They conducted a field experiment over 3 consecutive evenings in 3 Iowa, USA supermarkets. Each supermarket created an aisle end display of Campbell’s soups for 79c each, after a 12% discount as the regular price was 89c. Each supermarket presented a different limit notice each evening. Limits were: ‘No limit per person’, ‘Limit of 4 per person’, and ‘Limit of 12 per person’. Shoppers were observed unobtrusively. For each of the 914 shoppers, the data collected on them included: whether they purchased the soup, and how many cans they bought. Data for 8 shoppers, however, were excluded as they bought above the purchase limit.

Results showed that ‘no limit’ shoppers purchased an average of 3.3 cans. ‘Limit of 4’ buyers purchased an average of 3.5 cans. ‘Limit of 12’ buyers purchased an average of 7.0 cans. The ‘limit of 12’ sign increased sales per buyer by 112%.

Experiment 3: Researchers examined the effect of anchor-based slogans such as ‘Snickers bars – buy them for your freezer’. They also examined the effect of these anchors when accompanied with or without a price discount.

120 undergraduates participated in a shopping scenario study. Each participant was offered 6 well-known products at one of 3 price levels: a convenience store price (no discount), 20% discount, or 40% discount. All subjects were given suggestive selling claims, but some had no product quantity anchors, whereas some did.

Example of a suggestive selling claim without a product quantity anchor: Snickers bars – buy them for your freezer,
Example of a suggestive selling claim with a product quantity anchor: Snickers bars – buy 18 for your freezer.

Participants were not told whether the price was discounted or not. They were asked to provide for all products their intention for their respective purchase quantity decision.

Results suggested that both the suggestive anchor slogans and the level of discount increased purchase quantity intentions. However, authors claimed that supermarket shoppers might resist point-of-purchase (external) anchors by using self-generated (internal) anchors.

Experiment 4: The fourth study provided evidence that anchoring is the psychological mechanism driving the results of the previous experiments.

Course credits were given to 139 undergraduate students from a large university who took part in the study. Each participant was told they were involved in a shopping study for a local grocery store, and was given a shopping scenario involving 25-30% discounts on single servings of well-known products (e.g. Snickers candy bars). There were 4 purchase quantity limits: a no-limit control, limit 14, limit 28 and limit 56.

After studying the details of the promotional deal, subjects were immediately asked to answer the question, ‘How many units of this product do you usually buy at a time?’ After answering, the students indicated their intended purchase quantity for the item.

After seeing the deal, subjects were asked ‘On each of the lines below, please write down a different situation in which you might imagine yourself consuming this product.’ They were also asked, ‘How many do you think you might use in the next month?’ Lastly, participants provided their intended purchase quantities.

In the no-limit control condition, the results supported the anchoring model by showing that both low and high internal anchors can overpower the effects of external anchors.

Pre-cognitive decisions
Purchases are driven by a combination of consumer preference and price. Knutson et al.’s (2007) study on the neural predictors of purchases used fMRI scanning technology to investigate brain activity during purchasing decisions.

Subjects were scanned while engaging in a novel task. The SHOP (Save Holdings or Purchase) task consisted of a series of images. First, each subject saw the product for 4 seconds, then the product with a price for 4 seconds, and thirdly, a screen which asked them to choose their whether to purchase this product at the given price, was also shown for 4 seconds. After this trial, they fixated on a crosshair for 2 seconds before the next trial began.

The researchers predicted that:

  1. During the task, the preference would be shown by the neural circuits linked to anticipated gain getting activated.
  2. During the price presentation phase, excessive prices would activate the neural circuits linked to anticipated loss.
  3. The activation before the purchase decision being made would indicate whether individuals would choose to purchase a product or not.

Participants were 26 healthy right-handed adults; 12 were females and 14 were males; their age ranged between 18 and 26. They were all screened for the use of any psychotropic drugs (alcohol, and caffeine, etc.) and ibuprofen, substance abuse in the last month, and any history of psychiatric disorders. This was done before collecting informed consent. A further 6 participants were excluded as they bought less than 4 items, and a further 8 participants were excluded due to excessive head motion during the sessions.

Participants were paid $20 for their participation. To ensure engagement in the task and hence validity, they were told that one trial would be selected at random to count ‘for real’. If an item was purchased in the trial, they would pay the price (out of their $20) and would be shipped the product. Products ranged from $8 to $80, but were discounted by 75%.


  • Subjects purchased a mean of 23.58 items out of 80 products shown.
  • There was no significant difference in the number of products purchased by men and women.
  • Participants’ reaction time did not differ between products that were purchased and not purchased. However, for purchased products, the reaction time negatively correlated with preference. In other words, the more it was preferred, the less time it took for participants to decide whether to purchase it or not.
    This suggests that the reaction time indicates some form of response conflict (I want it, but it is too expensive).
  • Findings showed that there are distinct circuits which anticipate gain and loss.
    Product preference activated the nucleus accumbens (NAcc), and expensive prices activated the insula and deactivated the mesial pre-frontal cortex.
    In a nutshell, the findings suggest that the activation of distinct brain regions related to anticipation of gain and loss can be used to predict purchasing decisions.

General Evaluations:

  • The understanding of heuristics helps retailers as if a brand sells well, sales can be maintained by ensuring that customers can easily identify the one they usually buy. However, if a product doesn’t sell well, the understanding of customers’ use of heuristics can be used to break into this, by making the packaging look familiar, popular and trusted.
  • Wansink et al.’s study consisted of field experiments, hence raising the issue of the lack of control over extraneous variables which could have had influenced spending.
  • The study provided evidence that point-of-purchase promotions increase sales, and this gives a deep understanding of how these situational variables could have useful applications in retail.
  • Knutson’s study took place in a laboratory and would have had high levels of control, which increases reliability. However, the use of the fMRI would have had its limitations such as it can only capture a clear image if the person being scanned stays completely still.
  • The possible implications of this study’s findings are huge. If we can identify customers’ intentions without self-reports, then we remove all of the cognitive processing which might interfere with the subsequent decision made.
  • Wansink et al.’s and Knutson’s findings are most likely generalisable to all cultures even though the studies are not conducted cross-culturally.

Intuitive thinking and its imperfections
Thinking, fast and slow. System 1 & System 2.
Kahneman’s (2011) book ‘Thinking, Fast and Slow’ which presented his theory of thinking.
System 1 thinking corresponds to ‘thinking fast’. It’s described as intuitive and automatic.
System 2 thinking corresponds to ‘thinking slow. It’s described as controlled and deliberate.

Kahneman and Tversky claimed that most people are fast thinkers (system 1) most of the time, and that people rarely use slow thinking when making decisions. This makes predicting the decision-making outcome of people difficult.

Kahneman also discusses that heuristics bias our thinking, and claims that irrelevant material given to us can influence our answers, resulting in a biased answer.

Please check the textbook for several examples of the concept discussed by Kahneman.

Choice blindness
Choice Blindness – People don’t often notice when they are blind to their own preference as sometimes, they don’t notice that they have been presented with something they did prefer.

Hall et al. conducted a study into choice blindness for the taste of jam and the smell of tea. Shoppers (180 in total; 118 were females) at a supermarket in Sweden were recruited as they passed a tasting venue (opportunity sampling). Participants were asked to sample different varieties of jam and tea and to decide which one they preferred. They were told they’d be given their chosen item as a gift at the end of the taste test.

Immediately after making their choice, they were asked to sample the chosen item again and to verbally explain why they chose it. The experimenters had switched the contents of the sample containers (deception) so that the rejected item was now the preferred item.

Stimulus material: 3 pairs of jam & 3 pairs of tea selected from a pre-test in which independent participants rated the similarity of 8 pairs of jam and 7 pairs of tea.

To manipulate the choice, the bottoms of 2 small containers were glued together and wrapped it to create the illusion of a single container. For each participant, either the tea or the jam condition was manipulated, and the type of manipulation was randomized for each trial.

After the tasting, participants were asked whether they felt anything was unusual. Participants were then debriefed and given a chance to indicate whether they suspected a manipulation.

Detections were classified into 3 categories:

  • Concurrent detection – voiced concern after immediately detecting manipulation.
  • Retrospective detection – claimed to have noticed manipulation but reported at the end.
  • Sensory-change detection – did not report detecting manipulation, but described it tasted/smelled somehow different the second time.


  • The participants in the tea condition that had the gift incentive had a lower detection rate (19.6%) than participants that did not receive a gift (46.3%).
  • Result showing choice blindness: No more than a third of the switches were detected in jam conditions even if the tastes were different. No more than half of the switches were detected in tea conditions even when the smell was different.
  • Situational difference: Authors realise that this was a low-risk decision with no negative consequences, and that consumers are more likely to recognise manipulation in a decision with higher stakes. More research is required in this area.
    It is interesting to note that participants who were offered a gift were less likely to notice the manipulation than other participants. This shows that choice blindness remains strong even if there are consequences.

Advertising and false memory
Experiment 1
Background: Research shows reconstructive memory affects how people remember past events. Braun LaTour at el. (2004) tested whether this can be applied to advertising. The question was: will any experience post advertising influence the recollection of the advert?
Reconstructive memory: theory suggesting that the act of remembering is influenced by various factors such as cultural beliefs, expectations, and stereotyping.

Aim: To investigate whether true and false autobiographical advertising would be processed and remembered in the same way.

Research Hypothesis: Participants who recognised the deception would be less likely to create false memories than those who didn’t recognise the deception.

Sample: 66 undergraduate students from a US university; 32 females, and 34 males with an average age of 21. They were randomly allocated to 1 of 2 conditions. Independent measures design was implemented.

Procedure: Researchers told participants false information about a non-Disney character and expected this to change what participants remembered about their experience at Disneyland.
The advertisements had a vignette to imitate the Disney experience.
The true advert had a picture of Mickey Mouse and the researchers made a reference to the participant shaking hands with Mickey at Disneyland.
The false advert had a picture of Bugs Bunny, who is a Warner Brothers character, and the researchers made a reference to the participant shaking hands with Bugs Bunny at Disneyland.

  • Read and evaluate the advert.
  • Rate their attitude, affect and likelihood of visiting Disneyland in the future.
  • Talk about their past experiences visiting Disneyland, and whether they had seen certain characters at the park.

They were then debriefed and informed about false memory research.


  • Although a small number of participants identified that Bugs Bunny shouldn’t be on an advertised for Disneyland, there was largely no difference between how true and false adverts were processed.
  • There was no difference in which the true and false adverts affected participants’ attitudes towards Disney.
  • Participants who received the false Bugs Bunny information were more likely to recall Bugs Bunny memories (22% to 7%).
  • Some participants in the true condition confused Bugs Bunny with a Disney memory.
  • The false advert did seem to affect memory, since a very small number of participants believed the advert changed what they remembered.

Experiment 2
Experiment 2 was similar, but had 3 false conditions:

  • False information was provided as a picture.
  • False information was provided in words.
  • False information was provided as both words and pictures.

100 students from a different US university were randomly assigned to 1 of 3 conditions.

  • The verbal-only condition produced the largest number of ‘Bugs detectors’ (participants who recognized this was false information).
  • Pictures had a stronger effect than words, since the 2 conditions with Bugs Bunny’s picture produced the greatest number of false memories.
  • 2 Ways to interpret these findings:
    1. Participants in the verbal condition processed information more deeply than in other conditions. Hence, they’d be more likely to detect deception, and less likely to create a false memory.
    2. Participants did not process the words deeply enough – they overlooked the verbal mention of Bugs since it wasn’t as prominent as the picture of Bugs, hence they developed false memories less.

Experiment 3
Experiment 3 had a memory test at the end of the experiment to distinguish between the above 2 explanations (aim).

Research Hypothesis: The researcher predicted that if information is more deeply processed, then the memory of it will last longer and participants will be more likely to remember it. If the second explanation is more accurate, then we would predict that less information would be recalled.

Sample: 110 participants were randomly allocated to 1 of 3 conditions. Participants were in a computer lab in groups of 30.

Procedure: They were given the advertising information and asked to give feedback. They were given 10 minutes to write about their first childhood experience at Disneyland, and then were asked to complete a computer task which had 2 parts.
First, they had to recognise items from their own childhood visit to Disneyland, to check whether they would identify Bugs Bunny as being part of their childhood experience, and how confident they were they met him.

Second, they had an Implicit Association Test (IAT) which aimed to check whether participants would categorise Bugs as belonging to the Disney or ‘Other’ theme park category.

Finally, participants gave a written memory task in which they recalled everything they could remember from the Disney ad they saw earlier and were asked questions including ‘What character did the child shake hands with?’.


  • Results confirmed experiment 2’s findings in that more people detected deception in the verbal-only condition, and more false memories were created in the pictorial conditions.
  • Participants remembered significantly more items in the pictorial condition. They remembered an average of 6.2 items in the ‘both’ condition, 5.1 in the pictorial-only condition, and 4.7 in the verbal-only condition.
  • 88% of participants in the ‘both’ condition recalled Bugs Bunny, 76% in the pictorial-only condition, and 47% in the verbal-only condition.

General Evaluations:

  • Application to the retail environment: The study by Hall et al. on choice blindness involved clever manipulation of the jars and the majority of people didn’t notice when the products were swapped. Retailers can benefit from this, as they realise packaging is very important. If a less costly product was packaged in a similar way an expensive product was packaged, there’s high chance customers would buy it.
  • The study by Braun LaTour on false memory shows that information received after an event can alter our memory for that event. The concept’s application to the retail environment is relatively new.
  • The 2 studies show that situational variables clearly influence our behaviour, whilst the discussion on the way we think, brings focus on individual cognitive processes.