M303 Secondary Data

FALL 2026 · LECTURE 04
M303
Secondary Data
Use existing evidence to sharpen the question before collecting anything new.

Where we are

FROM RESEARCH DESIGN TO DATA COLLECTIONBefore collecting anything new, check whether existing evidence can answer—or sharpen—the research question.
02
01

Use your name tag.
It helps everyone learn names and participate.

02

Please put away phones, tablets, and laptops
unless I ask you to use them.

Learning Objectives

01Distinguish

Secondary data and research from primary data and research.

02Evaluate

The benefits, limitations, and fit of secondary data.

03Locate

Internal and external sources of secondary data.

04Compare

Retail audits, consumer panels, scanner data, and single-source systems.

Problem formulationResearch designData collectionAnalysis & recommendation

Prior-class review

01

When you have little prior knowledge of the problem, use ________ research.

02

Methods that are unstructured and mostly qualitative are typical of ________ research.

03

When hypotheses are well defined and specific, the research design is usually ________.

04

Research used to describe an existing situation is ________ research.

05

“Half of Indianapolis residents drink beer weekly” is an example of ________ research.

06

Questions such as what, who, where, when, and how are associated with ________ research.

07

Research that can establish whether one variable causes another is ________ research.

08

Experiments are the defining method for ________ research.

09

Match the problem stage: ambiguous → ________; understood → ________; clearly defined causes → ________.

Primary vs. secondary research

Primary research
Secondary research
Original data collected…
By you—or a company you hire—for the current research question
By someone else, usually for another purpose
Examples
Surveys · focus groups · interviews · observation · experiments
Using existing internal records, published data, or standardized services
Qualitative or quantitative?
Can be either
Can be either

Why use secondary data?

Start
smarter
01

May be sufficient to solve the problem.

02

Clarifies the problem and supports hypotheses.

03

Improves the design of primary data collection.

04

Provides a benchmark for validating primary results.

Examples: using secondary data

01Understand an industry

Establishments · revenue · employment · payroll per employee

02Estimate demand

Use sources such as the U.S. Census Bureau.

03Select target markets

Compare markets before investing in primary research.

04Geocode opportunity

Combine geography with demographic and psychographic information.

Example source · U.S. Census Bureau

Pros and cons of secondary data

PROS
  • Lower cost
  • Less effort
  • Faster access
  • Sometimes more accurate
  • Sometimes the only feasible source
CONS
  • Collected for another purpose
  • No control over collection
  • May be inaccurate or outdated
  • May use the wrong units or format
  • May not meet the current requirements
  • Often requires assumptions
Check: who collected it · why · how · what · when · whether sources are consistent

Sources of secondary data

Secondary
data
Internal sources

Customer databases
Sales and patronage
Marketing activity and cost
Distributor and customer feedback

External sources

Published data
Government · trade reports · books · news · annual reports

Standardized sources
Audits · panels · scanners · single-source systems

What is a customer database?

Raw customer information
enhanced to create value
TransactionsInventoryFeedbackDemographicsPsychographics
Historical source-deck claims

94% of large U.S. companies had customer databases; Ford held 50 million records and Kraft Foods held 30 million names.

Illustrative legacy figures—not current estimates; the source deck did not supply publication dates.

How to use customer databases

TRANSACTIONS
01

Understand customer preferences

02

Identify payment patterns

03

Analyze purchase behavior

04

Segment loyal, profitable, and loss-making customers

Example: segmenting by CLV

Rank customers into deciles based on customer lifetime value (CLV).
1
2
3
4
5
6
7
8
9
10
High CLVMedium CLVLow CLV

Example: customer profiles

Fictional mature retail customer illustrating the high-CLV profile
TYPICAL HIGH CLV CUSTOMER

Gender: female
Age: 35–54
Marital status: married
Children: present
Estimated household income: $125,000+
Distance: lives closer to the retailer
Loyalty card: member
Channel: mail-order shopper
Shopping pattern: shops frequently in upscale stores

Fictional younger retail customer illustrating the low-CLV profile
TYPICAL LOW CLV CUSTOMER

Gender: male
Age: 25–34
Marital status: single
Children: none
Estimated household income: below $50,000
Distance: lives farther from the retailer
Loyalty card: not necessarily a member
Channel: single-channel shopper

Example: holiday catalog targeting

Fruit gift basket from the source deck
PROBLEM

Fruit-catalog company XYZ wants to know which customers should receive holiday gift cards.

Whom should the company target?

External sources of secondary data

01Published data

Sources created by governments, firms, associations, journalists, and scholars.

02Standardized sources

Data collected repeatedly through consistent procedures for multiple users.

Which source best fits the decision, time horizon, unit of analysis, and required detail?

What is published data?

The most widely used source of secondary data
01

Readily available

02

Can answer part—or all—of the research question

03

Available in print or searchable digital databases

Sources of published data

Federal governmentState and local governmentsStatistics agenciesBusiness publicationsNews articlesAnnual reportsAcademic publicationsLibrary databasesComputerized bibliographiesSyndicated services
Business publication shown in the source deckNewspaper shown in the source deck

Standardized sources of secondary data

Data collected repeatedly using standardized procedures, usually for multiple subscribing users.
01

Retail store audits

02

Consumer panels

03

Scanner services

04

Single-source systems

1. Retail store audits

DATA COLLECTION

At regular intervals, auditors visit a sample of stores to count inventory and record deliveries since the previous visit.

Beginning inventory + deliveries − ending inventory = movement through the store

They can also record shelf prices, displays, shelf space, and in-store promotions.

Count+Deliveries−Inventory=Movement

Historical example: Nielsen Retail Index

HISTORICAL SOURCE-DECK EXAMPLE

Nielsen’s auditing services were organized around four reporting groups:

Grocery productsDrugsOther merchandiseAlcoholic beverages
Nielsen Retail Index illustration from the source deck
SalesVolume · trend · shareDistribution% of stores · out-of-stocksSelling pricesVolume at each price or dealRetailer decisionsIn-store promotionsMedia advertisingTotal category · major brands

2. Consumer purchase panels

STORESKU → SALEaggregate movement
Store audit

How much moved through stores?

Missing link

Who bought it?
How often?
Did they switch?

HOUSEHOLDBUY → REBUYbehavior over time
Consumer panel

Which households purchased—and how behavior changed over time.

How are panel data collected?

01
Home audit

A panel member allows an auditor to check household stocks in selected categories at regular intervals.

Historically required households to retain packages and wrappers.
02
Purchase diary

A panel member records purchase details and submits the diary at regular intervals.

Traditionally returned by mail; the underlying method can now be digital.

Benefits of consumer panels

MARKET SNAPSHOT
Aggregate sales activityBrand shares
PURCHASE HISTORY
Heavy buyersBrand-switching ratesCumulative market penetrationTrial and repeat purchase

Limitations of consumer panels

01Selection bias

High refusal rates and incomplete population coverage.

02Panel mortality

Members drop out over time; the source deck cited annual dropout above 20% as a historical example.

03Testing effects

New members may report unusual purchasing because they know they are being observed.

04Missing purchases

One household reporter may overlook purchases made by other members.

3. Scanner data

Shopper at a scanner checkout from the source deck
Universal Product Code (UPC) systems record transactions at checkout.
They connect individual-household purchasing behavior and marketing activity to sales and revenue.
Product purchasing recordsProduct priceCoupon responseShelf space and displaysMarket and brand advertising

Benefits of scanner data

◎High accuracy

Transactions are captured at the point of sale.

◷Time savings

Data can be available faster than periodic audits.

↺Short-interval analysis

Study demand and marketing activity week by week—or even more frequently.

Historical example: P&G and Walmart

HISTORICAL RETAIL-ANALYTICS EXAMPLE
What should Walmart stock before a hurricane?

The source deck reports that strawberry Pop-Tarts sold at roughly seven times their normal rate before hurricanes—and that beer was the top pre-hurricane seller.

Knowledge is not only power. It can also be profit.
FlashlightStrawberry Pop-TartsRetail shelf

How to benefit from scanner data

STORE AUDIT · EVERY TWO MONTHS
Source-deck chart showing store-audit sales shares of 6.7, 15.7, and 16.5 percent at weeks 8, 16, and 24

A smooth trend may hide short-lived events.

WEEKLY SCANNER SERVICE
Source-deck chart showing weekly sales-share changes across 24 weeks

Promotions, shortages, and step changes become visible.

Measure promotion liftForecast new productsTest price and profitImprove assortment and placement

4. Single-source systems

01Recruit a panel

Identify participating households.

→
02Measure exposure

Monitor TV programming and commercial exposure.

→
03Link purchases

Use identification at scanner-equipped stores.

→
04Compare response

Relate exposure to household purchasing.

Historical source-deck example · Information Resources, Inc. (IRI)

Benefits and limitations of single-source systems

BENEFITS
  • Compare household purchases before and after exposure to different advertising messages
  • Obtain test results quickly
LIMITATIONS
  • Tracks purchases only at participating scanner-equipped stores
  • Cannot confirm whether viewers actively watched or merely left the TV on
  • Small test markets may not represent larger markets
  • Panel participation can affect behavior

In-class exercise

RETURN TO YOUR MARKETING PROBLEM OR OPPORTUNITY
Which secondary-data sources could answer useful questions about it?
01

Identify relevant internal, external published, and standardized sources.

02

For each source, specify the insight it could provide.

03

Note one limitation or assumption you would need to manage.

To sum up

01

Distinguish primary from secondary research.

02

Evaluate secondary data for fit, quality, timeliness, and bias.

03

Use internal and external sources.

04

Compare published data with standardized sources.

05

Know what audits, panels, scanners, and single-source systems can—and cannot—tell you.

Next class

NEXT
Qualitative and observational methods of primary data collection
ASSIGNMENT 1 · DUE BEFORE CLASS ON CANVAS
  1. Select a firm with a marketing problem you want to study. (1 point)
  2. Create a 2 × 2 SWOT table with three thoughtful bullets in each cell. (3 points)
  3. Use the SWOT to state a marketing problem in a few sentences. (1 point)