Increasing
Profits From Existing Customers
|
| Everyone wants to increase
profits, but obtaining new customers in a competitive environment can be
difficult. It is much easier to obtain more profit through increased sales
to existing customers than it is to search out new customers. Getting more
sales from existing customers also reduces your incremental costs.
Assuming you are already incurring the cost of delivery, if you can
increase the delivery size, your cost as a percentage of sales goes down.
Pulling two additional lines for an existing customer costs less that
pulling two lines for a new customer. |
| This message identifies some techniques for
increasing sales to existing customers through the use of basic data mining. |
Overview:
|
|
The information contained in this message was
extracted from an actual commercial office products dealer data warehouse.
Although we used the OPSoftware data mining tools developed for the DDMS
NT and UNIX computer systems, this is not a requirement. Anyone with basic
access to company data and commercially available data manipulation tools
can achieve the same benefits from the methods detailed in this message.
|
|
If you own a DDMS NT or UNIX computer system,
and would like more information regarding the data warehousing and data
mining tools available from OPSoftware, please contact Jack
Duncan by email or voice at 800-722-3615.
|
|
The methods detailed here represent only a few
of many possible scenarios. All queries
(queries are questions you ask your data warehouse) are basic and beginner
level complexity. Your individual situation may be different.
|
Where to begin?
|
|
Ideally, you would have your inventory and
customers well classified and could easily pull reports by item group and
type of customer. Unfortunately, this is not the usual case and most dealers can't
afford to spend the time categorizing their items and customers. There is
another way to quickly identify these missed opportunities through a few
simple queries.
|
Broader approach:
|
|
Although more precise comparisons can be
generated when customers and items are classified, you can still make
broader assumptions based on existing information.
|
|
If you can identify your top 100 requested
items, you can then make the broader assumption that most of your customers should
be buying some, if not many, of these top items. Customers that are buying
few, or none, of these top items are potentials for increased sales.
Because our assumptions are broad, accuracy will go down, but you will
still uncover many opportunities.
|
Step 1: Identify your top items:
|
|
You can easily identify your top 100 items
using data mining. The best way to identify your top items is by
the number of hits. Hits are not sales, but represent the number of times
the item was requested. Using your item database, create a query based on
year to date hits, sort the number of hits in descending order, and limit
the query to the top 100 rows. Be sure to include the item number in your
query, you'll need it later.
|
|
Once you are satisfied with your query
results, change the query type to a 'Make Table' query to create a new
table containing your top items. You can then open this new table and make
any manual adjustments such as deleting miscellaneous items. The separate
table is a good idea because items such as 'Photo Copy', 'Delivery Charge'
and other single customer unique items may creep into your top 100. The
separate table allows for modifications to your top 100 list without
affecting your live data warehouse.
|
Step 2: Identify your active customers.
|
|
For our purposes here, active customers are
defined as customers that
have made any purchase during a specified time period. This time period
should be broad enough to include seasonal activity, if any, and to
include those customers who order infrequently.
|
|
Create a query in your
sales database selecting all customers with item purchases between the
dates you want to compare. Be sure to set the query to unique values only
so that the customer is listed only once regardless of how many times they
have ordered.
|
Step 3: Match your top hits to customer sales:
|
|
Using your sales history database, create
another query linking the top items table you created in step 1 to your customer sales detail
matching on the item number field. Set the query to show unique values so
the customer is only listed once regardless of how many of the top 100
items they are buying. Be sure to limit this query to the same begin and
end dates if you also limited the customer count query to specific dates.
|
|
You can also adjust the top row count for top
50, 25, etc and compare this to the count of active customers to come up
with a chart like the one below:
|
|
|
In this example, and using the actual dealer
data, 38.11% of the total active customers were not buying ANY
of the top 100 items! These customers are prime targets for increased
sales.
|
Step 4: Produce order forms:
|
|
The customers not buying your top items can be further
sorted by sales rep. Then, and using your graphical
report writer, you could easily create a report in an order form format
showing only the items NOT ordered by each individual customer.
|
|
|
Sales Order Form
|
Customer Number: 12345
ABC Company
1234 Any Street
Atlanta, GA. 30024
|
| PREFIX |
ITEM
NUMBER |
DESCRIPTION |
UNIT |
PRICE |
QUANTITY |
| OD |
22010 |
Letter Size Multi-Purpose Paper |
CT |
$99.99 |
_______ |
| SAN |
65450 |
Black Medium Point Uni-Gel Grip Pen |
DZ |
$99.99 |
_______ |
| MMM |
6549 |
Yellow 3 by 3 Post-It-Notes 100 Per Pad |
DZ |
$99.99 |
_______ |
| And so on, and so on, listing all of your top x items
not ordered by this customer. |
|
|
Step 5: Plan your attack:
|
|
You should conduct a sales meeting explaining
your goals to your sales force. Prepare in advance individual sample boxes
for each sales rep containing as many products as possible that you want to
push. Pass out the order forms and sample boxes to your sales reps. Be sure
to explain that these order forms are based on general information and may
not apply to all customers.
|
|
You should also explain to your sales reps that
the reason some customers are not buying your top items is that they may be
ordering 'catalog' items and are unaware of your 'house brand' items. This
is where samples and good sales techniques can become important.
|
|
And finally, be sure to point out that there is
more commission involved when your top items are sold. If you are like most
dealers, you make better margins when customers order your top items. This
is proven out by other data mining exercises which will be the subject of
future messages.
|
Step 6: Send out the troops!
|
|
Send out your sales reps armed with the samples
and order forms and start collecting increased sales from existing
customers!
|
Bottom line:
|
|
You can spend a couple of hours creating queries
(they are included in the OPSoftware product), a couple of reams of paper
printing order forms, the cost of some sample product boxes, a morning sales
meeting, and dramatically increase your sales to existing customers.
|
Next steps:
|
|
You can continue to perform this exact same
process over and over on specified dates, or create variances such as
reversing the process by focusing on what customers are currently buying and
then searching for higher profit substitutes. But that is yet another useful
function of data mining and a topic for a future message!
|
|
