Supermarket Customer Observation and Electronic … Customer Observation and Electronic Data Analysis With Implications for the Marketing Plan by Harry F. Krueckeberg Professor of

Supermarket Customer Observation and

Electronic Data Analysis

With Implications for the Marketing Plan

by

Harry F. KrueckebergProfessor of Marketing

College of BusinessColorado State University

Preface

This report is based on research initiatedto determine the potential role of observationmethodology in market plan development. Thisreport is based on 100 observations conducted ina supermarket dairy department.

The findings have implications for marketplanning, merchandising and communicationstrategies.

As a pilot project, the results are notbased entirely on randomly selected customers.This report represents efforts to apply andrefine the observation technique and notnecessarily to convey results that can be pro-jected to all shoppers in a supermarket. Furtherwork is being done to accomplish randomobservations.

Introduction

Role of Planning Data

Planning for efficient supermarket salesmust be based on valid data. Economic datasuch as gross sales, gross profits, and otherindicators of store productivity are gatheredinternally by management, The computer ageallows management to generate numerous datacharacteristics and statistics.

Mail and telephone surveys of consumers,both among a store’s customers and its non-customers, are another source of informationuseful in creating and monitoring the marketingplan. Survey data facilitate the determination

February 89/page 74

of consumer awareness, attitudes and shoppingactivities.

Grocer associations, trade magazines,consultants and others conduct projects whichgenerate industry data. This secondary datamay or may not be applicable to a specificsupermarket in its planning process. Both pri-mary and secondary data are needed for effec-tive market planning.

This report describes a pilot project whichimplemented customer observation methodologyand analyzed data using a mini computer and astatistical software package.

Observation in Context

The process of conducting research oncustomer behavior includes numerous alterna-tive methods and procedures. Observation isone alternative methodological tool useful fordata collection. The characteristics of theobservational method do not automatically limitanalytical methods and procedures.

The procedural alternatives involved inimplementing observation permit statisticalanalysis of a high degree (ANOV, Regression,etc.) or simply descriptive analysis consisting offrequencies, percentages, means and others.The alternative random procedure or non-ran-dom procedure dictates the extent to whichobservation data can be analyzed.

Journal of Food Distribution Research

.. --— ---

Project Justification Research Project Objectives

There are advantages to simplifying thecollection of research data. Yet there is also aneed to maintain a level of validity in theresearch resulta. Many times there is a need ordesire for supermarket store managementand/or employees to conduct research in orderto obtain information in a timely and effectivefashion.

Observation research can be justified interms of its time and cost. Store employees canbe trained to conduct observation research.Sophisticated analysis or simple tallies of thedata can be made. Each result can be a valuablepiece of information which management can useto determine shopping patterns, describe shop-pers, profile market segments, and gain insightinto store layout, advertising and promotionalstrategies.

Customer observations should be consid-ered a means of establishing communicationwith the consumer. Communication uses andtransports information from one decision makerto another. Communication is a critical tool inmanagement decision making. Direct observa-tion of overt behavior can be interpreted and“encoded” to facilitate communication betweenconsumer and observer.

Communication with consumers is essen-tial to profitable business management. Two-way communication is extremely important.But, does this communication have to be direct?Could it be observed, interpreted and reacted toin order to complete the communication cycle?In supermarkets today, consumers react to manyfactors high prices, poor quality and poor ser-vice. Management needs to determine problemsin some manner and then to take steps to reducethe number of complaints or possibly eliminatethem. This process involves the presence oftwo- way communication.

This two- way communication can beaccomplished indirectly by both parties. Thecustomer can communicate through actionswhich are observable. Management, in turn,can communicate through its reaction to theobserved actions of the consumer.

Research Project Goal

The goal of this research project was “todetermine the potential role of the observationmethod in developing market planning informa-tion.”

The research objectives of this projectwere

1. Determine shopping and purchase actionsof food store shoppers using direct obser-vation during normal shopping hours.

2. Determine procedural strengths andweaknesses of the observation method ingenerating planning data.

3. Determine specific items purchased bydemographic market segments, i.e., age,sex, etc.

4. Determine alternative types of recommen-dations for marketing strategies and tac-tics that can result from observationmethodology.

Project Procedures

Data Collection

Permanent records were generated duringthe observation of each shopper. The generat-ing of permanent records did not detract fromthe unobtrusive nature of the procedures.Observers had no discretion about what was tobe observed. All actions and characteristics tobe observed were included on the record page.

Consumers were “picked up” at either endof the dairy department or in mid-department.Consumers were not second-guessed; that is,consumers were observed regardless of whetheror not interest was apparent or whether theyseemed to be only passing through. All persons,once picked up, were documented and con-sidered in the analyses even if no purchaseswere made.

This project was conducted in one super-market during a four-week period. The 100observations were conducted in onedepartment--the dairy department of the store.

One person completed all observations inorder to maintain a consistency of procedure.One department was involved in order for theobserver to remain very unobtrusive. Themethod of observation in this project can bedescribed as

1. being in a natural setting and situation,2. unobtrusive,3. structured,4. resulting in direct observations, and

Journal of Food Distribution Research February 89/page 75

5. involving a human observer.

The natural store setting was used. Therewas no obvious indication of any observationsbeing made. The observation was structured sothat specific pieces of data were observed. Theobserver was not instructed to “see what you cansee.” Rather, a consistent instrument (floorplan) was used. The observations were madedirectly; no mechanical equipment was involved.

The number of persons per time periodwas not proportional to the actual numbersshopping at various times during the day ornight. The procedures were used in order todemonstrate what can be done, and how,regardless of store, department, time of day orany other situation.

Data Interpretation

The findings should be interpreted andthe interpretation influenced by several situa-tional factors, Such things as the followingshould be noted

1. Weather (rain, snow, etc.)2, Location of store3. Street or construction obstructions4. Time of week (end of week)5. Time of month (payday, etc.)6. Time of year (holidays, etc.)

Strengths of the Observation Method

The strength of observation data lies in“what it is.” It is detailed action applicable tothe setting in which the observation is made.It can be a complete record of actions takenduring a period of time, within a prescribedspatial dimension.

Observational data do not result frominterpretation. The person observed is notrequired to recall actions, to answer a question-naire or to complete a personal interview.Likewise, the observer is not in a position tointerpret an answer by an interviewee. Actionis recorded, not interpretation.

The observation itself, the recording ofthe observation, the data processing, and theanalysis have each individually and all collec-tively created greater potential for managementinformation development. In summary, thebenefits of observation include:

● Situation description development formarket plan development.

February 89/page 76

●

●

●

●

●

Department, sectiontivity improvement.

Customer serviceimprovements,

or location produc-

and convenience

Restocking procedures and policies analy-sis and improvements.

Complementing the direct product profitanalysiso

Target market profile analysis and defini-tion.

Weaknesses of the Method

There are those who express the opinionthat a method that generates action dataexclusively is methodologically weak. Thisreflects excessive expectations of the methodand an attempt to stretch the capabilities of aresearch method which has limited purpose andlimited potential.

One of the weaknesses often referred to isthe inability to explain the cause or reason forthe action taken, This is not one of theintended purposes of observation. The alterna-tive method which a researcher can implementin order to accomplish interpretation is the per-sonal interview. This method can be imple-mented upon completion of the observationwithout compromising the observed data.Unanswered questions can be resolved directlyand obtrusively.

Observational data can be biased, just asany other data can be biased if non-acceptableprocedures are used to carry out the observationmethod. Bias”is minimized by providing everymoment of a time frame and every unit of theobserved population equal access to theobserver. As applied to a supermarket, if adepartment is being observed, all shoppers dur-ing shopping hours should have an equal chanceof being observed.

Likewise, there is no control of variablesinvolved in comparable situations, i.e. two storesor two different time periods. Comparability islimited because of a lack of ability to determinecause and effect.

Others consider the data as qualitativedata rather than quantitative. Wells andLoScinto [11state “that the re~orts are narrativerather th&” quantitative.” This report stresses


. .—..— ----

that some of the data can be quantitative andcan be analyzed statistically.

The above weaknesses of the dataresulting from observation are for the most partthe result of excessive expectations of themethod. On the other hand, the weaknesses canbe offset by procedures which result in datarepresentative of locations and time periods.

Highlights of Observed Findings

Introduction

The highlights that follow describe anumber of pieces of information resulting froman implementation of the observation technique.These types of findings should provide storemanagement with opportunities to develop mar-keting plans and communicate more effectivelywith the customer.

Highlights

1. Ninety-nine (99) percent of the observedshopped at least two sections (“section”represents a four-foot linear section ofthe display case) of the dairy case on thefirst pass through the department.

All parties shopped an average of 12 ofthe 18 dairy case sections.

2. Eighty-five (85) percent of all partiespurchased at least one unit (not item)from the dairy case.

All parties observed averaged 2.45 unitspurchased on the first pass through thedepartment.

The 85 percent purchased an average of2.88 units.

3. Forty (40) percent shopped at least twosections of the dairy case on the secondor return pass through the dairydepartment.

All parties shopped an average of 4.07sections on a second or return passthrough.

The 40 percent of the parties making asecond trip through the departmentreshopped an average of 10.18 sections.

4.

5.

6.

7.

8.

9.

Nineteen (19) percent purchased at leastone unit of product on the second orreturn pass through.

AU parties observed averaged .27 unitspurchased. The 19 percent of partiespurchased an average of 1.42 units on thesecond trip through the dairy department.

Forty-eight (48) percent stopped in frontof at least one dairy department sectionwithout making a purchase.

All parties stopped in front of an averageof .77 (less than one) sections.

The 48 percent stopped at an average of1.6 sections.

Twelve (12) percent stopped in front of atleast one dairy department section on thesecond pass through the dairy departmentwithout making a purchase.

All parties stopped in front of an averageof. 16 sections on the second pass through.

The 12 percent stopped at an average of1.33 sections.

Ninety-nine (99) percent passed at leastone gondola end on the first pass throughthe dairy department. All parties passedan average of 4.03 gondola ends out of the6 possible.

Twenty-six (26) percent were observedstopping at a gondola but did not make apurchase. The average party shopped .33gondolas. The 26 percent shopped anaverage of 1.27 gondolas.

Fourteen (14) t)ercent of all ~arties r)ur-chased at ‘leak~one unit from a goridolaend. The average party purchased .15units.

The 14 percent purchased an average of1.07 units.

Findings and Results

Introduction

The following findings are presented inorder to demonstrate several alternative ways ofanalyzing observation data. This is not anexhaustive list and the data allow numerous


..— — -- —.. —-—..-.

additional analyses to be made. The groupingsof findings can be identified as

1.

2.

3.

4.

5.

6.

Description of individuals observed.

Locations of purchases among shelves.

Relationships of descriptions of theobserved and locations of the purchases.

Action profiles of market segments.

Ingress and egress characteristics,

Profile of the linear distribution of mr-chases and observations.

Description of Individuals Observed

Tables 1 through 5 represent descriptionsof the shoppers and how they shopped the dairydepartment. This analysis provides theresearcher with information that will probablyraise questions to be answered through furtheranalysis.

Table 1

The description of the partiesobserved with regard to sex, numbersin the parties, and the composition ofthe shopping party, 100 Fort Collins,CO Shopper observations, Fall 1987.

Percent Partv Com~os ition

213014

2151

43

100

Individual femaleIndividual maleMixed adult couple

Female with childrenMale with childrenCouple with children

Female under 18 yearsMale under 18

Total Observed Parties

old

62 Total adult females observed53 Total adult males observed

115 Total adult persons observed

February 89/page 78

Table 2

The use of shopping carts andshopping lists by 100 shopping partiesobserved in Fort Collins, CO, Fall1987.

Percent Information

Used a shopping cart;; Did not use a cart

40 Used a shopping list60 Did not use a list

100 Total observation

Table 3

The time of day that grocery shoppersobservations were made, 100 shopperobservations, Fort Collins, CO, Fall1987.

Pe cent.

r eo f Dav

6 AM to 1059 AM2: 11 AM to 1:59 PM

2 PM to 459 PM:: 5 PM to 8:59 PM

4 9 PM to 11:59 PM

100 Total

Table 4

The number of shopping partiesobserved with accompanyingchildren, 100 parties observed,Fort Collins$ CO, Fall 1987.

No children observed;; One or more

100 Total


----

Table 5 Table 6

The estimated age classifications of the100 shopping parties observed in FortCollins, Co, Fall, 1987.

Percent Age Estimate

9 Under 1819 18-2443 25-34

17 35-497 50-655 66 and over

100 Total

Weighted average age 33 yearsMedian 25-34Mode 25-34

Location of Purchases Among Shelves

Where in the department were purchasesmade? Individual males made fewer purchasesfrom lower shelves (Table 6) while females withchildren purchased from all shelves in the dairycase. Further analysis provides more details(Table 7) about the shelf location from whichvarious parties (by sex classification) made pur-chases. Additional analysis involves anothercustomer characteristic, age (Table 8).

Relationships of Descriptors andLocations of Purchases

What did the observed person buy in thedairy department? Tables 9, 10 and 11 presenta description of some of the purchase behavior.For example, individual males were notobserved purchasing sour cream. Milk pur-chases represented 21 percent of the purchasesobserved. Sour cream was not purchased bypersons between the ages of 35 and 65 yearsold.

The type of party observed andthe shelf location from whichpurchases were observed, 100Fort Collins, CO supermarketobservations, Fall 1987.

Partv Obse rved Obse rved Shelf Purchases

Individual female Fewer purchases from mid-dle shelves

Individual male Fewer purchases fromlower shelves

Mixed Adults Fewer purchases fromlower shelves

Females with All shelveschild/children

Others Fewer purchases fromlower shelf

Table 11

The product groups most predominantlypurchased by parties of various charac-teristics, 100 Fort Collins, CO, super-market observations, Fall 1987.

Observed CharacteristicsProduct Grou~ of Purchases

Eggs Purchased by all parties

Dough Purchased by all parties

Spreads Less frequently purchasedby individual females

Juices Fewer purchases made bymixed adult parties

Yogurt Purchased more frequentlyby mixed adult parties

Cottage cheese More frequently purchasedby females with a child

Sour cream No purchases observed byindividual males

Milk Fewer purchases by ‘individual females

End gondolas Fewer units purchased bymixed adults and others


-.. —-. -— .-. --..-----

Table 7

The shelf location of dairy product purchases, by sex characteristics of party shopping, 100Fort Collins, CO, supermarket observations, Fall 1987.

Location of Purchase------- ------- ---------- ------------- --------

Upper Middle Lower Characteristic ofShelves helves Shelves artv Obser ved

--------------- percent ---------------

16 10 Individual Female14 ;: 15 Individual Male25 26 19 Mixed Adult Pair

25 20 27 Female with Child(ren)20 18 19 All Others

100 100 100 Total

35 38 27 Total of all purchases

Table 8

The shelf location of dairy product purchases, by age of party observed, 100 Fort Collins,CO, supermarket shopper observations, Fall 1987.

Location of Purchase*------- ------- ------- -------- ------- ------- --

Upper Middle LowerShelves Shelves Shelves Age of Partv

--------------- percent ---------------

22 13 Under 18 Years Old14 ;: 17 18-2448 40 42 25-34

14 9 20 35-490 6 5 50-652 8 3 66 Plus

100 100 100 Total

February 89/page 80 Journal of Food Distribution Research

.- —.—.

Table 9

The product groups purchased in the dairy department by observed parties by sexclassification, 100 Fort Collins, CO, supermarket shopper observations, Fall 1987.

-------------------- Characteristic --------------------

Individual Individual Mixed Female & Product % ofAdults

.e ren) Others rourx Total

----------------------- percent -----------------------

23331414

24 18 20 Spreads27 ;: 23 Milk

;; YogurtK 21 :; ;: Eggs

12 23 23 23 Juices11 22 23 11 Dough14 44 Cottage cheeseo U 14 # Sour cream

;;1815

12433

Table 10

The product groups purchased in the dairy department by observed parties by agecharacteristic, 100 Fort Collins, CO, supermarket shopper observations, Fall, 1987.

---. ------.-:---- shopping p~ty -----------------

Under 18- 25- 35- 50- 66 Product % ofus Total rouus rc~es

------------------------- percent -------------------------

45 20 2 100 Spreads 24:; :: 40 10 ; 4 100 Milk 2122 20 12 2 100 Yogurt 1818 15 :: 6 : 13 100 Eggs 15

15 50 4 100 Juices 120 :: ;: o 2! 100 Dough 4

29 %/ 14 0 0 100 Cottage cheese 3x 14 43 0 0 14 100 Sour cream 3


. . ..— — .—-. —.. . . .

February 89/page 81

Action Profiles of Market Segments

Marketing plans frequently include adescription of the target markets. How can atarget market be profiled? Given a profile, howmany customers are represented or included inthe description?

A number of profiles of several marketsegments observed are presented below. Eachof these profiles permits managers to determinethe relationships of characteristics of thoseobserved and the actions they have taken in thedepartment.

1.

2.

3.

4.

5.

6.

1.

2.

3.

4.

5.

6.

“Individual Female”

Accounted for 21 percent of the partiesobserved.

Seventy-one (71) percent used a shoppingcart.

Thirty-three (33) percent used a shoppinglist.

Thirty-eight (38) percent shopped in themorning and early afternoon and 38 per-cent shopped after 5 p.m.

Estimated average age, 35.7 years.

Purchased an average of 1.68 units ofproducts.

“Individual Male”

Accounted for 30 percent of the partiesobserved.

Fifty-seven (57) percent used a shoppingcart.

Thirteen (13) percent were observed usinga shopping list.

Fifty (50) percent were observed shoppingafter 5 p.m. in the afternoon.

Estimated average age, 35.4 years.

Purchased an average of 1.78 units.

1.

2.

3.

4.

5.

6.

7.

1.

2.

3.

4.

5.

6.

7.

“Observations with Children”

Twenty-seven (27) percent of observedparties included children.

Sixty-seven (67) percent of observed par-ties with children included a sole femaleand 22 percent included a sole male.

Ninety-six (%) percent of parties withchildren used a shopping cart.

Sixty-four (64) percent of parties withchildren used a shopping list.

Sixty-one (61) percent of parties withchildren shopped after 5 p.m.

Sixty-four (64) percent of parties withchildren included adult shoppers age 25-34.


t PI’*

“24 Years and Under Age Group”

This age group was equally distributedamong the 5 major descriptive groupsstudied, i.e. individual female and male,etc.

Twenty-eight (28) percent of allobservations were age group 24 years andunder.

Seventy-one (71) percent used a shoppingcart.

Thirty-six (36) peroent used a shoppinglist.

Forty-three (43) percent shopped from 2p.m. to 5 p.m.

Fourteen (14) percent were accompaniedby at least one child.



.—

1.

2.

3.

4.

5.

6.

7.

8.

1.

2.

3.

4.

5.

6.

7.

8.

“Shopping Cart Users”

Seventy-seven (77) percent ofobserved used a shopping cart.

all those

The most frequent cart users were femaleswith at least one child, 27 percent of allusers.

Fifty-one (51) percent observed using ashopping cart also used a shopping list.

A greater percent of persons (from 67percent to 83 percent) used shopping cartsas the time of day became later.

Thirty-five (35) percent of those whoused a shopping cart involved at least onechild.

Forty-three (43) percent of cart userswere age 25-34 years old.

The most infrequent cart users wereindividual males.


et Ma ket Prof IQr i

“Shopping List Users”

Forty (40) percent of those observed useda shopping list.

Thirty-five (35) percent of the list userswere females with at least one child.

The most infrequent users of a shoppinglist were individual males; 43 percent ofall users.

Ninety-eight (98) percent observed usinga shopping list also used a shopping cart.

Fifty-three (53) percent of shopping listusers shopped after 5 p.m.

Forty-five (45) percent of parties using ashopping list included children.

Forty (40) percent of list users were age25-34, with another 35 percent age 35 ormore.



The objective of store management aswell as departnwnt managers is to optimize thenumber of urtJt$ sold in the display space. Howcan observation data reveal PIogress foward thisobjective?

Analysis revealed the number of unitspurchased and, with this number, a number ofcharacteristics were associated. In other words,the average number of units purchased can becalculated for various market segments of theshoppers observed.

The greatest average number of unitspurchased was by mixed sex adult couples--4.5units (Table 12). Shopping list users purchased1.5 more units on average than did non-listusers. The table reveals a number of othercontrasts and comparisons.

Table 12

The number of units purchased byvarious segments of the customers whoshopped a supermarket dairy depart-ment, Fort Collins, CO, Fall 1987.

. .Umts Purchased

.

Entry from meat departmentEntry from HABA aisle

Mixed adult couplesIndividual femalesIndividual males

Cart usersNon-cart users

Shopping list usersNo list

Evening shoppersMorning shoppers

Children presentChildren not present

Under 24 years oldOver 34 years old

Exited to meat department

Overall

Ingress

average

3.101.50

4.001.681.78’

3,201.10

3.602.10

4.302.10

2.112.35

3.202.00

2.80

2.45

and Egress Characteristics

The placement of products in a depart-ment setting supposedly influences the route toand through the department. The route will be

February 89/page 83

--— ——-. . .

influenced by the customer’s planned next pur-chase, a knowledge of what items are next inthe shopping sequence, and other factors.

Observation of the dairy department,located in a corner of the store, revealed theingress and egress of the 100 customers (Tables13 and 14).

No one predominant pattern wasestablished. Forty (40) percent came from themeat department, following the periphery of thestore, but 7 percent went out along the peri-phery. Forty-one (41) percent entered the dairydepartment from grocery aisles (Figure 1).

Persons entering the department from aperpendicular aisle shopped an average of ninesections of the dairy case while people comingfrom other entry locations shopped an averageof 14 sections.

Persons entering from the meat depart-ment purchased an average of 3,1 units of prod-uct while those coming from the HABA aislepurchased an average of 1,5 units.

Additional graphic representation inFigure 2 shows both the ingress and egress pat-terns.

Distribution of Purchases

Figures 3 and 4 illustrate the observedlocation of purchases in the department by the100 customers.

The evaluation of these observationsreveals that there was not a significant variationin the number of purchases among the sectionsof the dairy department. There was a sig-nificant variation in the location of stops orobservations by customers.

Observations versus purchases occurredin the mid-section of the department. Theseobservations occurred in sections containingpredominantly impulse items vs. staple or regu-lar demand items.

The implications for communication stra-tegies are at least three-fold. First, specials andcoupons might reduce “looks” and increase pur-chases. Secondly, more advertising and promo-tional efforts may increase awareness of theitems and improve attitudes toward these prod-ucts. Thirdly, in-store demonstrations mayreduce looking and increase purchases.

Table 14

The number of units purchased byvarious segments of the customers whoshopped a supermarket dairy depart-ment, Fort Collins, CO, Fall 1987.

characte ristic Units Purchased

Entry from meat department 3.10Entry from HABA aisle 1.50

Mixed adult couples 4.00Individual females 1.68Individual males 1.78

Cart users 3.20Non-cart users 1.10

Shopping list users 3.60No list 2.10

Evening shoppers 4.30Morning shoppers 2.10

Children present 2.11Children not present 2.35

Under 24 years old 3.20Over 34 years old 2.00

Exited to meat department 2.80

Overall average 2.45

General Conclusions

Introduction

There are a number of conclusions as aresult of this project, conclusions regardingprocedures and results. Recommendations formanagement and for improving the proceduresfollows in subsequent sections.

Results

The dairy department observations indi-cated that consumers, regardless of characteris-tic, make a very quick trip through the depart-ment. The infrequent stops and the frequencyof units purchased reflect a hurriedly conductedshopping trip through the dairy department.

Observable differences in market seg-ments were obvious. Individuals purchasedfewer units while couples and evening shopperspurchased twice the number of units.Individual market segments were identifiable


—..

Table 13

The characteristics of the shoppers entering the dairy department from three differentdirections from the meat department, from adjacent aisles and from the HABA aisle,100 Fort Collins, CO shoppers.

Location of Ingress--------- ------ ------ ---”------ ------ ------ ------ ------ -

From Meat Adjacent HABAShotme r Characteristic De~artment Aisles Aisle Total

--------------- percent ---------------

Individual male 40 40 100Individual female 38 :! 19 100

Females in general 28 60 100Males in general 40 20 ;; 100

With shopping cartsWith shopping list

After 5 p.m. shoppersWith childrenAge 25-34

Ingress same as EgressSections shopped one-way (average)

Units purchased (average)End gondola unit purchases (average)


3640

382133

3514

3.1.11

29

2.9.37

14 1005 100

13 100100

2; 100

0 NA14

1.5.08

February 89/page 85

.—— -—.. . ..-.

In

February 89/page 86

-#&_

.-


. ..—. . . ..


ELanaW

+o

2

February 89/page 87

-.. .........

Figure 3

The number of items purchased from the various sections of the dairy department by100 supermarket customers, Fort Collins, CO, Fall 1987.

01 02 03 04 05 06 07” 08 09 10 11 12 13 14 15 16 17 18

‘ 3 4 6 Z/ 5 7 2 3 ~H 4 ~6 2 2 3 3 5 ~ [

2 4 .3 3“ 3 “x’ /3 (L (j 3 Y y K+ % 2 3 3“ 5 z

3 2 3 4 4 ‘% 7 3 -3 6 / _? / .3 3 .3 3 3 +— Eggs — Douah sutw FnM Jube YOmrt cottage sow — M&

SOUTHEND

1AISLE

200 11---

Pelt Suppms

Automotive

.,Mrrgmhe ‘ ,- Cheeie Crem

— I I — —

HAISLE

210 12---

DetergentsBleach

* SOaDHOuldmld

The number of observationssupermarket customers, Fort

01 02 03 04 05 00

Figure 4

in the various sectionsCollins, CO, Fall 1987.

of the dairy department by 100

07 08 09 10 11 12 13 14 15 16 17 18

hwgutrw

SOUTHEND

1

1AISLE

200 11---

Pet Sup@Da

Automottw

—

210----

AISLE12

etargentsbath

& SOqHouwhok

—

220---,

230----

AISLE

Hnl

AISLE AISLEJ4 240 15 250 16

--- ---

elw Shmlfma SemltyUDpln Owdermh Medlche

iqmmshmlpoo Pop

=Y Food ToothPwo BeerPmty Hose


-. .-

-——.,..= —. - -—

using numerous characteristics and, as a result,segment profiles could be developed.

The end gondolas were not shopped byconsumers; the average person purchased .15units-- 15 units per 100 customers.

It can be concluded that the potential forimpulse shopping is relatively great. Sixty per-cent of those observed did not use a shoppinglist while 77 percent used a shopping cart.

Shelf locations of purchases reflectedsome differences among market segments.Several segments did not use lower shelves. Dothere need to be changes in product location inorder to achieve unit sales or greater departmentefficiency and productivity?

Product group purchases were influencedby market segments also. Again, this type ofdetermination leads to conclusions regardingadvertising strategies and product promotions.

Products were purchased throughout thedepartment. The “looks” at various productswere not distributed evenly throughout thedepartment. Lookers were more concentrated inseveral sections--sections involving impulseitems. It is concluded that these products areless familiar to consumers and would benefitfrom in-store product demonstrations, sales,special displays, advertising and promotion,

The entry into and exit from thedepartment leads to the conclusion that storetraffic does not consistently follow the peri-phery. The lack of a shopping list and theabsence of a dairy product on the list suggestsone reason for the traffic pattern.

Strategic Recommendations

The efficiency with which the averageobserved customer shopped the dairy depart-ment suggests a lack of a pull strategy to lureconsumers into the dairy department. Theresults instead suggest a push strategy; conse-quently, the consumer is not influenced bypoint of purchase display attractions nor by thedairy department in general.

The traffic pattern reinforces the needfor dairy department management to considera more concerted effort to pull customers intothe department.

The results suggest more advertising andpromotion focused on mixed adult couples andevening shoppers. These two market segments

were observed to purchase an average of 4.0 and4.3 units respectively.

On the other hand, several other low unitpurchasing segments might also receive thefocus of more advertising and promotion.These target markets would be individual males,individual females and persons over 34 yearsold.

Highlights of Procedural Findings

Introduction

This is the second of two highlights sec-tions. The earlier one dealt with selected find-ings of the observation project.

As a result of the observations in thesupermarket, several procedural findings weredetermined. These can be summarized as fol-lows.

Procedural Findings

1. Persons making the observations shouldbe better trained to recognize the age ofindividuals.

2. Distinct markings should be made on thefloor of the observed area so that a personobserved from a distance can be accur-ately traced in and out of the aisles.

3. Thorough familiarity with the displays inthe observation area needs to be achievedbefore beginning observations.

4. Determine if any drastic changes in theproduct placement or location in displaycases or on shelves is planned during theobservation time period. Plan orreschedule the observation period ifnecessary.

Procedural Recommendations

Based on this pilot project, the use ofseveral procedures was reinforced. The datarecording sheet used in the pilot project shouldnote

1. Time entering and time exiting, allowingfor the recording of elapsed time.

2. Ethnic classification of individual(s).

3. Any observable package inspection; out-side or contents.


---- .. —-I

4. Any observable and obvious shopping cart Referencesor basket contents of interest.

[1] Wells, William D., and Leonard A.5. Observation of any requests for assistance. LoScinto, ‘Direct Observation of

Purchasing Behavior,” Journal of6. Notations of obvious frustration or con- Marketing Research, vol. 3, August

sternation. 1966, pp. 227-33.

7. Notations regarding the observable use of [2] Lehmann, Donald R., Market Researcha newspaper and coupons. and Analysis, Richard D. Irwin, Inc.,

Second Edition, 1985, p. 84.

[3] Atkin, Charles K., “Observation ofParent Child Interaction in SupermarketDecision- Making,” Journal ofMarketing, October, 1978.


.

Documents

Supermarket Customer Observation and Electronic … Customer Observation and Electronic Data Analysis With Implications for the Marketing Plan by Harry F. Krueckeberg Professor of