Market Basket Analysis What is it?

Market Basket Analysis is a modelling technique based upon the theory that if you buy a certain group of items, you are more (or less) likely to buy another group of items. For example, if you are in an English pub and you buy a pint of beer and don't buy a bar meal, you are more likely to buy crisps (US. chips) at the same time than somebody who didn't buy beer. The set of items a customer buys is referred to as an itemset, and market basket analysis seeks to find relationships between purchases. Typically the relationship will be in the form of a rule: IF {beer, no bar meal} THEN {crisps}. The probability that a customer will buy beer without a bar meal (i.e. that the antecedent is true) is referred to as the support for the rule. The conditional probability that a customer will purchase crisps is referred to as the confidence. The algorithms for performing market basket analysis are fairly straightforward (Berry and Linhoff is a reasonable introductory resource for this). The complexities mainly arise in exploiting taxonomies, avoiding combinatorial explosions (a supermarket may stock 10,000 or more line items), and dealing with the large amounts of transaction data that may be available. A major difficulty is that a large number of the rules found may be trivial for anyone familiar with the business. Although the volume of data has been reduced, we are still asking the user to find a needle in a haystack. Requiring rules to have a high minimum support level and a high confidence level risks missing any exploitable result we might have found. One partial solution to this problem is differential market basket analysis, as described below. How is it used? In retailing, most purchases are bought on impulse. Market basket analysis gives clues as to what a customer might have bought if the idea had occurred to them . (For some real insights into consumer behavior, see Why We Buy: The Science of Shopping by Paco Underhill.) As a first step, therefore, market basket analysis can be used in deciding the location and promotion of goods inside a store. If, as has been observed, purchasers of Barbie dolls have are more likely to buy candy, then high-margin candy can be placed near to the Barbie doll display. Customers who would have bought candy with their Barbie dolls had they thought of it will now be suitably tempted. But this is only the first level of analysis. Differential market basket analysis can find interesting results and can also eliminate the problem of a potentially high volume of trivial results.

In differential analysis, we compare results between different stores, between customers in different demographic groups, between different days of the week, different seasons of the year, etc. If we observe that a rule holds in one store, but not in any other (or does not hold in one store, but holds in all others), then we know that there is something interesting about that store. Perhaps its clientele are different, or perhaps it has organized its displays in a novel and more lucrative way. Investigating such differences may yield useful insights which will improve company sales. Other Application Areas Although Market Basket Analysis conjures up pictures of shopping carts and supermarket shoppers, it is important to realize that there are many other areas in which it can be applied. These include:

Analysis of credit card purchases. Analysis of telephone calling patterns. Identification of fraudulent medical insurance claims. (Consider cases where common rules are broken). Analysis of telecom service purchases.

Note that despite the terminology, there is no requirement for all the items to be purchased at the same time. The algorithms can be adapted to look at a sequence of purchases (or events) spread out over time. A predictive market basket analysis can be used to identify sets of item purchases (or events) that generally occur in sequence something of interest to direct marketers, criminologists and many others.
A market segment analysis is one of the fundamental tools for a sales and marketing professional. Very few products can be all things to all people; hence, it is important to accurately analyze a market, and then choose the appropriate segment to target. A market segment analysis includes demographic and psychographic information as well as estimates of consumer purchasing power. This article explains how to conduct a thorough and accurate market segment analysis. Determine the appropriate metric to use to segment the market. Some of the most commonly used metrics include age, gender or income bracket. The specific metric will depend on the product. For example, if you are introducing a line of clothing, you may want to segment the market according to income since it will help determine if you should introduce luxury, everyday use or affordable clothing. Carry out a consumer behavior survey to determine their purchasing habits. You could conduct the survey yourself, or hire a professional consultant to do it. A good survey will identify how frequently customers purchase a given item, how much they spend on the item per year, and what factors cause them to purchase the item. Supplement the consumer survey with your own research. Nielsen is a marketing and advertising research firm that provides demographic information on consumers. For example, Nielsen data will determine how much adults ages 25 to 39 who make $50,000 to $100,000 per year and live in the Chicago, Illinois, metropolitan area spend on household items. There are multiple types of Nielsen subscriptions; a base subscription, which costs $550 per year, will provide the level of data needed to complete a market segment analysis. Download the data from the consumer behavior survey into Excel, and use the "Sort" function (under the "Data" tab in the tool bar) to organize the data according to the specific parameters you outline. Run a series of regressions in Excel using the specific parameters as independent variables and the predicted amount spent on the product as the dependent variable. You can locate the regression functionality by clicking on the "Analysis" tab, then the "Data Analysis" button in Excel. The regression will indicate which parameters will most strongly influence the willingness of consumers to spend money on a product. A positive coefficient in front of a variable means the parameter will influence consumers to spend more. For example, if the variable for income has

a coefficient of 1 and the variable for gender has a coefficient of 2, it means that gender is twice as likely to influence consumer spending as income. Select which market segment to target based on the regression data. If the variable for income has a large, positive coefficient, it probably makes sense to zero in on consumers in the high-income market segment.

Decision tree This article is about decision trees in decision analysis. For the use of the term in machine learning, see Decision tree learning. A decision tree is a decision support tool that uses a tree-like graph or model of decisions and their possible consequences, including chance event outcomes, resource costs, and utility. It is one way to display an algorithm. Decision trees are commonly used in operations research, specifically in decision analysis, to help identify a strategy most likely to reach a goal. Another use of decision trees is as a descriptive means for calculating conditional probabilities.

In decision analysis, a "decision tree" and the closely related influence diagram is used as a visual and analytical decision support tool, where the expected values (or expected utility) of competing alternatives are calculated.

A decision tree consists of 3 types of nodes:1. Decision nodes - commonly represented by squares 2. Chance nodes - represented by circles 3. End nodes - represented by triangles

Drawn from left to right, a decision tree has only burst nodes (splitting paths) but no sink nodes (converging paths). Therefore, used manually, they can grow very big and are then often hard to draw fully by hand. Traditionally, decision trees have been created manually - as the aside example shows although increasingly, specialized software is employed. Analysis can take into account the decision maker's (e.g., the company's) preference or utility function, for example:

The basic interpretation in this situation is that the company prefers B's risk and payoffs under realistic risk preference coefficients (greater than $400Kin that range of risk aversion, the company would need to model a third strategy, "Neither A nor B"). Influence diagram

A decision tree can be represented more compactly as an influence diagram, focusing attention on the issues and relationships between events.

The squares represent decisions, the ovals represent action, and the diamond represents results. Uses in teaching This section requires expansion. Decision trees, influence diagrams, utility functions, and other decision analysis tools and methods are taught to undergraduate students in schools of business, health economics, and public health, and are examples of operations research or management science methods. Advantages Amongst decision support tools, decision trees (and influence diagrams) have several advantages: Decision trees:

Are simple to understand and interpret. People are able to understand decision tree models after a brief explanation. Have value even with little hard data. Important insights can be generated based on experts describing a situation (its alternatives, probabilities, and costs) and their preferences for outcomes. Use a white box model. If a given result is provided by a model, the explanation for the result is easily replicated by simple math.

Can be combined with other decision techniques. The following example uses Net Present Value calculations, PERT 3-point estimations (decision #1) and a linear distribution of expected outcomes (decision #2):

Disadvantages Decision trees:

For data including categorical variables with different number of levels, information gain in decision trees are biased in favor of those attributes with more levels.

Example Decision trees can be used to optimize an investment portfolio. The following example shows a portfolio of 7 investment options (projects). The organization has $10,000,000 available for the total investment. Bold lines mark the best selection 1, 3, 5, 6, and 7, which will cost $9,750,000 and create a payoff of 16,175,000. All other combinations would either exceed the budget or yield a lower payoff.

Example In the game of "20 Questions", the querent tries to construct a short binary decision tree that isolates a specific item. The item's identity question is asked when the current decision tree node is considered reliable by the querent. Definition Decision trees are predictive models, used to graphically organize information about possible options, consequences and end value. They are used in computing for calculating probabilities and data mining, and the decision trees examples below relate to 'simpler' decision making, so to speak.

Decision trees are used for deciding between several courses of action. They create a visual representation of the various risks, rewards and potential values of each option.

Decision trees examples drawing your own Start with your decision and represent this on the left side of a sheet of paper with a small square. Then, for each possible option, draw one line out from the square towards the right. Leave plenty of space between these lines. Write each option on it's line. Then take the lines one at a time. At the end of the line, do you get a particular result, or is it uncertain or is there another decision to be made? If it is another decision, draw a square. If uncertain, a circle, and if a result, draw nothing. Review each square and circle. For the squares (decisions), draw lines for the choices, marking them in as you go. For the circles (uncertainties) draw further lines for the possible outcomes. Keep going until you have filled out the possibilities leading from your original decision. You will have something similar to the first of the decision tree diagrams.

Cluster analysis is a class of statistical techniques that can be applied to data that exhibit natural
groupings. Cluster analysis sorts through the raw data and groups them into clusters. A cluster is a group of relatively homogeneous cases or observations. Objects in a cluster are similar to each other. They are also dissimilar to objects outside the cluster, particularly objects in other clusters. The diagram below illustrates the results of a survey that studied drinkers perceptions of spirits (alcohol). Each point represents the results from one respondent. The research indicates there are four clusters in this market.

Illustration of clusters

Another example is the vacation travel market. Recent research has identified three clusters or market segments. They are the: 1) The demanders - they want exceptional service and expect to be pampered; 2) The escapists - they want to get away and just relax; 3) The educationalist - they want to see new things, go to museums, go on a safari, or experience new cultures. Cluster analysis, like factor analysis and multi-dimensional scaling, is an interdependence technique: it makes no distinction between dependent and independent variables. The entire set of interdependent relationships is examined. It is similar to multi-dimensional scaling in that both examine inter-object similarity by examining the complete set of interdependent relationships. The difference is that multidimensional scaling identifies underlying dimensions, while cluster analysis identifies clusters. Cluster analysis is the obverse of factor analysis. Whereas factor analysis reduces the number of variables by grouping them into a smaller set of factors, cluster analysis reduces the number of observations or cases by grouping them into a smaller set of clusters. In marketing, cluster analysis is used for

Segmenting the market and determining target markets Product positioning and New Product Development Selecting test markets (see : experimental techniques)

Basic procedure Main article: Cluster analysis 1. Formulate the problem - select the variables to which you wish to apply the clustering technique 2. Select a distance measure - various ways of computing distance:

Squared Euclidean distance - the sum of the squared differences in value for each variable Manhattan distance - the sum of the absolute differences in value for any variable Chebyshev distance - the maximum absolute difference in values for any variable Mahalanobis (or correlation) distance - this measure uses the correlation coefficients between the observations and uses that as a measure to cluster them. This is an important measure since it is unit invariant (can figuratively compare apples to oranges)

3. Select a clustering procedure (see below) 4. Decide on the number of clusters

5. Map and interpret clusters - draw conclusions - illustrative techniques like perceptual maps, icicle plots, and dendrograms are useful 6. Assess reliability and validity - various methods:

repeat analysis but use different distance measure repeat analysis but use different clustering technique split the data randomly into two halves and analyze each part separately repeat analysis several times, deleting one variable each time repeat analysis several times, using a different order each time

Clustering procedures There are several types of clustering methods:

Non-Hierarchical clustering (also called k-means clustering)

first determine a cluster center, then group all objects that are within a certain distance examples:

Sequential Threshold method - first determine a cluster center, then group all objects that are within a predetermined threshold from the center - one cluster is created at a time Parallel Threshold method - simultaneously several cluster centers are determined, then objects that are within a predetermined threshold from the centers are grouped Optimizing Partitioning method - first a non-hierarchical procedure is run, then objects are reassigned so as to optimize an overall criterion.

Hierarchical clustering

objects are organized into an hierarchical structure as part of the procedure examples:

Divisive clustering - start by treating all objects as if they are part of a single large cluster, then divide the cluster into smaller and smaller clusters Agglomerative clustering - start by treating each object as a separate cluster, then group them into bigger and bigger clusters

examples:

Centroid methods - clusters are generated that maximize the distance between the centers of clusters (a centroid is the mean value for all the objects in the cluster) Variance methods - clusters are generated that minimize the within-cluster variance

example:

Wards Procedure - clusters are generated that minimize the squared Euclidean distance to the center mean

Linkage methods - cluster objects based on the distance between them

examples:

Single Linkage method - cluster objects based on the minimum distance between them (also called the nearest neighbour rule) Complete Linkage method - cluster objects based on the maximum distance between them (also called the furthest neighbour rule) Average Linkage method - cluster objects based on the average distance between all pairs of objects (one member of the pair must be from a different cluster)

Review the diagram and if there are any more options, fill them in, too.

Evaluation Now it's time to work out which option is most valuable to you. First, estimate how much each option would be worth to you. Then review each circle/point of uncertainty. Here you determine the probability of each outcome. Make sure percentages add up to 100, or fractions amount to a total of 1. Your decision tree diagrams will now look something like this.

Calculate Start at the left hand values and work to the right. At any circles, multiply the end value by the probability of it occurring. So in our example, the probability of promotion is 0.7. Multiply this by 80,000 to get 56,000. Each of the end values is recalculated in this way. It's also useful to add in any costs that will be incurred along the way. This gives a more accurate picture of the net value.

Other decision tree examples Decision trees are frequently used for major, complex decisions such as times of major rebranding in organizations, or deciding whether to expand or not and in which directions.

Ups and downs

Decision trees clearly lay out the situation so that each and every 'node' can be scrutinized. Many possible consequences can be worked through. Information available can be used to give clearer and more accurate assessments. For those who like software and pretty computer printouts, there are addons for excel that create excel decision trees quickly and easily. One of the dangers of course, is that the decision never gets made because so much time is put into creating pretty pictures. The decision trees examples here can just as easily be drawn by hand. And remember the computer is not making the decision. The information put into the tree will determine the results. And, of course, this is influenced by the required outcome and the questions asked about it.

Cross-selling
Cross-selling is the action or practice of selling among or between established clients, markets, traders, etc. or the action or practice of selling an additional product or service to an existing customer. This article deals exclusively with the latter meaning. In practice, businesses define cross-selling in many different ways. Elements that might influence the definition might include the size of the business, the industry sector it operates within and the financial motivations of those required to define the term. The objectives of cross-selling can be either to increase the income derived from the client or clients or to protect the relationship with the client or clients. The approach to the process of cross-selling can be varied. Unlike the acquiring of new business, cross-selling involves an element of risk that existing relationships with the client could be disrupted. For that reason, it is important to ensure that the additional product or service being sold to the client or clients enhances the value the client or clients get from the organization. In practice, large businesses usually combine cross-selling and up-selling techniques to enhance the value that the client or clients gets from the organization (and vice versa). Benefits that can accrue to the customer include the efficiency and leverage that result from using a single supplier for multiple products. When buying complex professional services, like consultingneeded to make and integrate an acquisition, the use of one firm reduces the fingerpointing that is common when a problem occurs in an area that straddles two or more services; if only one firm is responsible, fingerpointing is eliminated. For the vendor, the benefits are also substantial. The most obvious example is an increase in revenue. There are also efficiency benefits in servicing one account rather than several. Most importantly, vendors that sell more services to a client are less likely to be displaced by a competitor. The more a client buys from a vendor, the higher the switching cost. Though there are some ethical issues with most cross-selling, in some cases they can be huge. Arthur Andersen's dealings with Enron provide a highly visible example. It is commonly felt that the firm's objectivity,

being an auditor, was compromised by selling internal audit services and massive amounts of consulting work to the account. Though most companies want more cross-selling, there can be substantial barriers: 1. A customer policy requiring the use of multiple vendors. 2. Different purchasing points within an account, which reduce the ability to treat the customer like a single account. 3. The fear of the incumbent business unit that its colleagues would botch their work at the client, resulting with the loss of the account for all units of the firm. Broadly speaking, cross-selling takes three forms. First, while servicing an account, the product or service provider may hear of an additional need, unrelated to the first, that the client has and offer to meet it. Thus, for example, in conducting an audit, an accountant is likely to learn about a range of needs for tax services, for valuation services and others. To the degree that regulations allow, the accounts may be able to sell services that meet these needs. This kind of cross-selling helped major accounting firms to expand their businesses considerably. Because of the potential for abuse, this kind of selling by auditors has been greatly curtailed under the Sarbanes-Oxley Act. Selling add-on services is another form of cross-selling. That happens when a supplier shows a customer that it can enhance the value of its service by buying another from a different part of the supplier's company. When one buys an appliance, the salesperson will offer to sell insurance beyond the terms of the warranty. Though common, that kind of cross-selling can leave a customer feeling poorly used.[citation needed] The customer might ask the appliance salesperson why he needs insurance on a brand new refrigerator, "Is it really likely to break in just nine months?" The third kind of cross-selling can be called selling a solution. In this case, the customer buying air conditioners is sold a package of both the air conditioners and installation services. The customer can be considered buying relief from the heat, unlike just air conditioners. 1. A Life Insurance company suggesting its customer sign up for car or health insurance. 2. An wholesale mobile retailer suggesting a customer choose a network or carrier after one purchases a mobile. 3. A television brand suggesting its customers go for a [home theater] of its or another's brand. 4. A laptop seller offering a customer a mouse, pen-drive, and or accessories.

Upselling (sometimes "up-selling") is a sales technique whereby a seller induces the customer to
purchase more expensive items, upgrades, or other add-ons in an attempt to make a more profitable sale. Upselling usually involves marketing more profitable services or products but can also be simply exposing the customer to other options that were perhaps not considered previously. Upselling implies selling something that is more profitable or otherwise preferable for the seller instead of, or in addition to,[1] the original sale. A different technique is cross-selling in which a seller tries to sell something else. A recent study concluded that it is 70% easier to get an additional 3% in sales from an existing customer than it is to get more customers in the door to equal the same dollar volume in sales. In practice, large businesses usually combine upselling and cross-selling techniques to enhance the value that the client or clients get from the organization in addition to maximizing the profit that the business gets from the client. In doing so, the organization must ensure that the relationship with the client is not disrupted. In a restaurant and other similar settings, upselling is commonplace and an accepted form of business. In other businesses, such as car sales, the customers perception of the attempted upsell can be viewed negatively and thereby affect the desired result. Examples Some examples of upsales include:

suggesting a premium brand of alcohol when a brand is not specified by a customer (such as if a customer simply requests a "rum and Coke"). selling an extended service contract for an appliance suggesting a customer purchase more RAM or a larger hard drive when servicing his or her computer selling luxury finishing on a vehicle suggesting a brand of watch that the customer hasn't previously heard of as an alternative to the one being considered. suggesting a customer purchase a more extensive car wash package. Asking the customer to super size a meal or add cheese at a fast food restaurant.

Techniques Many companies teach their employees to upsell products and services and offer incentives and bonuses to the most successful personnel. Care must be taken in this type environment to thoroughly train employees. A poorly trained employee can let slip the incentive program and thus offend a regular and loyal customer. There is a level of trust between the customer and employee and once broken it may never be reestablished. A common technique for successful upsellers is becoming aware of a customer's background and budget, allowing the upsellers to understand better what that particular purchaser might need.

Another way of upselling is creating fear over the durability of the purchase, particularly effective on expensive items such as electronics, where an extended warranty can offer peace of mind. Upselling also works with things like expensive leather shoes, where the seller suggests to buy the waterproofing spray as well "to make the shoes last."