University of Washington
Geography 350
Determining and assessing the primary market area

 

Determine the primary market area, given competitors’ locations:  customer spotting

If we really want to recognize market interpenetration for an existing outlet, we should ask “Where are our customers actually coming from?”  How would you know that?
Old approaches:

  • Check the license plates in the parking lot, and look up the addresses.
  • Offer promotional prizes;  entrants have to submit their home addresses.

 New approaches:

  • Provide discounts for customers who present a loyalty card at the point of sale.  (How could you increase the likelihood that consumers provide their actual addresses on a consumer card application?
  • Use info from credit/debit cards.
  • Request ZIP codes at checkout.

 

First, with this info, you can map your customers’ locations:  that itself tells you something about your market area and customer characteristics (since “birds of a feather live together”).

 

Second, you can

  • identify small areas (census block groups, or 9-digit ZIP codes, or whatever small area for which you purchase geolifestyles data), and
  • note how many customers live in each area.
  • If you assume that each customer spends the same amount, you can divide your store’s sales by all customers, note how many customers are in each small area, and determine your sales by small area.  With consumer discount cards, you even know how much each consumer spends per week, so you can trace where you high-spenders live and note how much revenue you derive from each small area.

 

Third, you can aggregate small areas that are contiguous with each other and to the store location, and see what proportion of your sales come from customers within various zones around the store.  Jones and Simmons suggest that the “primary market area” is the set of contiguous small areas that make up 60% of store sales. 

  • This is where you should focus your marketing efforts to get new customers:  they would find your store equally accessible as most of your customers, and they probably share some key characteristics with most of your current customers.
  • You may also be able to improve sales by tailoring your product mix, price points, store “feel,” and level of service to the demographic and lifestyle characteristics of the small areas within your primary market area.

 

Fourth:  Where is your store most successful?

  1. You can easily learn the number of households and the median (or better, average) HH income in each small area. 
  2. The Bureau of Labor Statistics performs a Consumer Expenditure Survey (http://www.bls.gov/cex/) that tells you what proportion of household income is spent on each of many different types of goods and services. 
  3. For each small area, multiply the number of households (HH) by the average HH income (HHI), and multiply that by the proportion of HH income that you expect to be spent on the goods or services you provide (exp/I):  HH x PCI x exp/I = the amount that you guess is spent in each small area on your kinds of goods or services. 
  4. What proportion of that money do you get from that small area?  Store sales in area i / total expenditures from area i.
  5. In which small areas do you have the highest market share?  What are their demographic and lifestyle characteristics? 
  6. Should you move your store to be closer to the nexus of those areas?
  7. What other small areas have similar characteristics?  Should you target marketing toward them?  Should you open a store near them?

 

Fifth:  What seems to determine the small areas in which you have the highest market share?

    Jones & Simmons provide an example of multiple regression analysis in which the store’s market shares in small areas is related to specific characteristics of each small area:

o   competitors’ sales (hard to get;  you could use the distance from the center of each small area to the nearest competitor)

o   pct of females in the labor force

o   urban vs. suburban

o   pct HH headed by someone >65 years old

o   I would add distance to your current store.

    This allows you to characterize those areas, and even to predict what market share you should have in a given area.

    It would also allow you to predict what your market share would be in a set of small areas surrounding a proposed new store location.

 

 

Forget the data on individual customers.  The Huff model allows us to designate a primary and secondary market area based on probabilities of consumer behavior.

 

Five steps:

1.     Divide the area into small statistical units.

2.     Determine the square footage of retail selling space of all shopping centers included within the area of analysis.

3.     Compute travel times.

4.     Calculate the probability of consumers in each unit going to the particular shopping center.

5.     Map the trading area of the shopping center in question by drawing lines connecting all statistical units having like probabilities.

P(Cij) =

probability that a consumer in small area i will shop at your store j

= Rj dij -a / Sj (Rj dij -a)

where R is a measure of retail attractiveness, such as the retail floor space in each retail store or shopping center, and d is distance.

See Jones & Simmons pp. 307-313.

We can attempt to increase P(Cij) by increasing Rj relative to ∑j Rj -- by reducing price, or by increasing sales area, selection, or service.
 


Finally, you can use Methods 6-9 to develop analog models:  based on your current experience, what market areas would be the best for a new store?
copyright James W. Harrington, Jr.
revised 29 March 2011