Trade Area Delimitation

8/9/2019 Trade Area Delimitation

1/35

Trade Area Delimitation and AnalysisTrade Area Delimitation and Analysis

Trade Area Conceptualization (1):

y Refers to the spatial extent (or distribution) of

customers around an individual stores or anetwork of stores.

y can be viewed as a contiguous area (or polygon)around a store (supply point) that contains themajority of the customers or potential customers(demand points).

y also known as market area orcustomercatchment area.

y


2/35

Trade Area Conceptualization (2):Trade Area Conceptualization (2):

y

Also viewed as the way of mapping theconfines of interaction between a set of storelocations and the customers that patronizethem.

y Interaction can be measured in different ways:number of customers,

number of transactions

rupee value of transactions

y It has a spatial dimension and geographicalboundaries, though boundaries are not alwaysclear


3/35

Trade Area Conceptualization (3):Trade Area Conceptualization (3):

y Trade Areas vary in size and shape.

y Factors that affect trade area size and shape are:x Store size (attractiveness)

x Settlement patterns (residential density)

x Transportation network

x Barriers to movement

x Presence of competitors (which provide alternativelocations and intervening opportunities)

y Can be used to provide information for trade areaanalysis

x characteristics of consumers/customers

x

screen development potentialx assess existing stores performance,

y Can be conceptualized and defined in different ways.


4/35

Who are concerned with trade areaWho are concerned with trade areadelimitation/analysis?delimitation/analysis?

y Retailers/ commercial service providers

y Commercial property developers

y Real estate department of retail chains

y Leasing companies

y Location analysts working for the above

y Marketing firms who do advertisement for businesses

y Educators who train students in the profession of marketinggeography, retail geography, and business geography


5/35

Three approaches to trade areaThree approaches to trade area

delimitation:delimitation:

y Spatial Monopoly (Deterministic)

y Market Penetration (Probabilistic)

y Dispersed Market (Customer profiling)


6/35

Deterministic approach has theDeterministic approach has the

following characteristics:following characteristics:

y Makes a clear-cut assumption about the spatial

dimension of the trade area

y

Trade areas are polygons, each has definiteboundaries; they do not overlap

y Assumes all customers come from this area;

(those living outside are excluded from

consideration)


7/35

Probabilistic approach has the followingProbabilistic approach has the following

characteristics:characteristics:

y Makes no clear-cut assumption about the spatialdimension of the traded areas

y Trade areas are not polygons, with no definite

boundaries; they overlap

y Assign persons (households) to stores partially,with the assumption that people do not alwaysgo to the closer store

y Treat trade areas as the surface of probabilities:primary (60%) and secondary (60-80%) etc.


8/35

Dispersed Market (also known asDispersed Market (also known asCustomer Profiling) has the followingCustomer Profiling) has the followingcharacteristics:characteristics:

y The supplier is often highly specialized. (e.g.,specializing one or two lines of importedfurniture, selling a narrow selection of books, orserving a widely scattered ethnic group.)

y There is no obvious spatial concentration ofcustomers; customers are widely dispersed.

y Distance decay relationship is weak

y Trade area is defined through customer profiling(i.e., age, income, ethnicity and life style.)


9/35

Two types of data for trade area analysis:Two types of data for trade area analysis:

Secondary data : the most commonly used are census data less expensive; and need less effort to acquire

can be used to identify potential customers, but many of thesepotential customers do not necessarily patronize the store. So, thedemographic profiles produced are not real customer profile report.

Primary data: compiled by retailers. collected at POS (either based on credit card transactions or by sales

associates asking postal codes and phone numbers)

Through customer data analysis, retailers develop a customer profileconsisting of demographic, social and economic attributes.

They can also use this profile to search for suitable sites in newmarkets.


10/35

User defined trade areaUser defined trade area

y Also called rules of thumb. It is hand-drawn around a given

store, from which the analyst believes the majority of customersare attracted.

y Relies on the level of experience and expertise of the person whodefines the trade area. It assumes that the person has knowledgeof customer base and how far they travel.

y It is highly subjective, not scientific.

y Quality can be improved, if limited customer spotting data areavailable and used as reference.

y Usually used to define trade area for a single store

y There are two types of such trade areas: Unconstrained trade area that do not follow census

geographies (but may follow physical barriers) Boundary constrained


11/35

User Defined Trade AreaUser Defined Trade Area

Free-hand Census tract confined

DA confinedFSA confined


12/35


13/35

Circular Trade AreaCircular Trade Area


14/35

Percentage of CustomersPercentage of Customers

y Percentage of customers uses customerdata.

y The analytical tools is the customer spottingmap.

y This simply is a map of distribution ofcustomers around a given store.

y Boundaries are drawn to include a givenpercentage of customers.

y Usually, distance is used to select the closest60% and 80% of the customers to the storelocation.


15/35


16/35


17/35


18/35


19/35

Market PenetrationMarket Penetration

y Divides the area into grids (200x200m, 500x500m)

y Place the same grid over the customer spotting map

y Count the number of spotted customers in each cell

y Divide the number of customers in each cell by the cells totalpopulation

y The ratio or percentage is regarded as a measure of marketpenetration

y If sales are known from the customer data, the number ofcustomers can be translated into sales, and sales can be dividedby total disposable income in the cell to develop a ratio.

y Outward from the store location, the number of cells iscounted until 60% or 80% of the customers or sales arereached. These cells form the primary and secondary tradeareas.

y With this method, there may be some holes which have nodata or no customers; or some outliers which have a significantnumber of customers. It is the analysts decision to includethem or exclude them.


20/35


21/35


22/35


23/35

ThiessenThiessen PolygonPolygon

y a geometric procedure for delimitingtheoretical trade areas for a network ofstores

y assumes the stores are similar in size andsell similar products for similar price;consumers purchase products from theclosest store.

y most suitable for delimiting trade areas ofchain stores.


24/35

Thiessen polygonThiessen polygon


25/35

ThiessenThiessen polygonpolygon


26/35


27/35


28/35

ReilleysReilleys LawLaw


29/35

StatisticallyStatistically--Calculated Probabilistic Method;Calculated Probabilistic Method;

The Huff Model (1)The Huff Model (1)

y Huff model is useful in the following ways:

Generate customer volume estimate for existing stores

Generate customer volume estimate for proposed newstores

Answer such strategic questions:x What would happen to my trade area if my store expand by

50%?

x What would happen to my trade area if one of my stores isclose?

x What would happen to my trade area if an existingcompetitor were to leave the market?

x What if a competitor introduces a new store in the market?

Map the probability surface

Estimate sales potential


30/35



Huff model requires the following data:

y A list of stores (shopping centers), their locations and

attributes (attractiveness)

y A list of building block areas with demographic andsocial economic data (market size and purchasing

power)

y A matrix of distance, driving time, travel costs between

each building block and each store

y A sample data set (for calibrating parameters/weights) .


31/35



y The challenge is to estimate the parameters.

y There are two ways to estimate them:

1. To make an educated guess.This is used when sample data are not available. Itdepends on the experience and knowledge of the local market. Usually several

guesses are made for experiment to find out which one generates better results.

2. To statistically estimate or calibrate the model. Often, it includes using a numberof different non-linear models. This requires the use of sample data. Severalparameters are experimented, and a measure of goodness of fit is produced.Calculations are undertaken to estimate the direction and amounts each of theparameters should change to improve the fit. Each change is then entered intothe model, and the model is re-run until the best values that give rise to the best

fit to the sample data are found.


32/35

Huff s model calculationInformation required Store-A Store-B

Retail space in each store 36 sq m 57

Travel time 4 minutes 12 minutes

Lamda ( the type of product affects theway customer balance the attraction ofstore size against the inconvenience of

travel)

2 2

Pull factor (floor space) 36 57

Drag factor(travel time topower of lamda

16 144

Pulling power= floor

space/drag factor

36/16= 2.25 57/144= .396

Pulling power of all the

stores= weight of store-A+

store B

2.25+.396= 2.646


33/35

y Probability that consumer shop at store-

A= pulling power of A/ pulling power of

all stores put together

y 2.25/2.646= .85

y For B 1-.85 = .15


34/35

Sales potential estimateSales potential estimateS. P. = No. ofHH * average HH income

* % of income spent on consumer goods

* probability

Example:

S.P. in building 1 at supermarket A:=567 * $26,700 * 0.3 * 0.93

=$4.22 million


35/35

Comparison of Thiessen, Reileys and HuffComparison of Thiessen, Reileys and Huff

Factors Thiessen Reileys Huff

Quality of transportsystem

Yes(d-time; travel time

reflect quality oftransport system)

Yes(d=time)

Attractiveness Yes Yes

Types of goods Yes ()

Competition Yes Yes Yes

Transport barriers Yes (d=time) Yes

Accuracy of saleestimate

Low Low High

Comments Good for chain stores(similar size,identical goodsand price);

No major barriers;Simple to use

Good for differentsized stores;

Consider barriers;Relatively simple to

use

Good for differentsized stores;

Consider barriers;Complicated to use

Trade Area Delimitation

Documents