Direct Marketing Magazine - March 1997
Only the Shadow Knows
Shining a Light on Shadow
Demand
Paul V. Teplitz
When you ask DM inventory managers what harm occurs when theyre out
of stock, nine out of ten cant give you an answer. They can tell you
initial fill and final fill rates, but they can only guess at what lies beyond,
in the area sometimes called shadow demand, or lost phone
demand, or skip-its.
The unfortunate fact is that the vast majority of order entry systems in
use today simply dont gather information about what happens when customers
learn the products theyve requested are out of stock. Most systems
just record actual orders or backorders. For example, if a customer asks
for an item, learns it is unavailable, and orders a substitute item, most
order entry systems merely record the order, with none of its history.
Re-cently-released systems have begun to capture partial information, but
a complete picture appears to be years away.
Even the initial fill and final fill numbers reported by such systems leave
important gaps. The next time someone quotes you such a number, probe to
see whether it meas-ured early in the mailings life or if it is a true
overall average. Does it include soldouts? What about substitutions? Does
it include items the company canceled, for example, due to vendor failures
or unacceptable quality? You will probably find that a true
clinical measure of stockouts is worse than what your systems
report.
Stockouts are surprisingly common. When you consider that the average cus-tomer
order often includes 2 or 3 items, the likelihood of encountering a stockout
mounts quickly, as the following table shows.
Chances that a customer will encounter a stockout
| |
Average
|
Items
|
Per
|
Order
|
|
| Initial
Fill Rate |
1
|
1.5
|
2
|
2.5
|
3
|
| 100%
|
0%
|
0%
|
0%
|
0%
|
0%
|
| 95%
|
5%
|
7%
|
10%
|
12%
|
14%
|
| 90%
|
10%
|
15%
|
19%
|
23%
|
27%
|
| 85%
|
15%
|
22%
|
28%
|
33%
|
39%
|
| 80%
|
20%
|
28%
|
36%
|
43%
|
49%
|
| 75%
|
25%
|
35%
|
44%
|
51%
|
58%
|
| 70%
|
30%
|
41%
|
51%
|
59%
|
66%
|
| 65%
|
35%
|
48%
|
58%
|
66%
|
73%
|
| 60%
|
40%
|
54%
|
64%
|
72%
|
78%
|
Many direct marketers consider an initial fill rate of 80 percent acceptable
and a rate in the mid-80s as pretty good. Yet, when soldouts, canceled
items, and substitutes are considered, an 85 percent initial fill rate probably
translates into an 80 percent in-stock rate, or worse. As the shaded area
in the table shows, this pretty good performance still leaves
30 to 40 percent of customers hearing the dreaded words, Im sorry,
but were out of that item...
Lets look closer into the shadowy world of shadow demand. From occasional
surveys and from the few companies whose systems capture relevant information,
we can piece together some educated guesses of what happens when customers
hear their items are unavailable. Typically they make one of three choices:
1) accept a backorder, 2) substitute another item, or 3) give up and skip
the item altogether.
What happens next? Surprisingly little seems to be known, and the possibilities
multiply quickly as you involve more stages in the process. Figure 1 below
outlines some of the more common outcomes, depending on which path is chosen.
The bolded entries in the right column indicate the least desirable outcomes.
This is a simplified picture, for example, it makes no adjustments for returned
mer-chandise. Nor do the example percentages capture the escalation that
occurs as initial fill rates get worse. Typically, as initial fill rates
sink below 70 percent, the portion of cus-tomers who cancel or say skip
it rises rapidly.
The path that people take through this range of options will depend on many
fac-tors such as:
· Whether they need the item by a certain day, for example, for a trip.
· Whether it is an impulse purchase.
· Whether they are a first-time or repeat customer.
· How often theyve encountered backorders with you before.
· How competitive the market is for the item they selected.
The good news in this picture is that roughly two-thirds of customers
encountering a stockout will get what they wanted, with only a delay. The
bad news is that about a quarter of them end up with a distinctly bad outcome,
such as going to a competitor or owning a product theyre unhappy with.
Translated into overall terms, at an 85 percent initial fill
rate, about 9 percent of the demand a company generates will end up with
its competitors, or distinctly unhappy with its service, or both.
Dealing with Shadow Demand Measuring the Extent of the
Problem
In developing a strategy to reduce the effects of shadow demand, the top
priority is to make it more visible, so top management knows the true economics
of its business. As the old quality control saying goes, You cant
fix what you cant measure. The goal, therefore, should be to
develop estimates of the lost demand, then apply normal margins, lost shipping
revenue, backorder costs, and the like, to compute a cost of lost
demand number for reporting to management.
Some ideas for estimating the amount and costs of lost demand include:
· Ask your order entry software suppliers if they offer any ways to
capture sub-stitutions, soldouts, or lost telephone demand. Some do, now,
and your ques-tion might spur them to add more such capability in the future.
· Conduct a spot survey, for one week, say, among the telephone salespeople.
· Phone a sample of stockout customers and ask what they did upon learning
their item was unavailable. Did they do without it, get it elsewhere, or
buy a substitute (and if so, where)?
· Similarly, phone a sample of customers who bought substitute items
(logged manually by the telephone sales reps). Ask how happy they are with
their choice. If they had it to do over, would they make the same choice?
Dealing with Shadow Demand Damage Control
Even if you cannot fix the problem of shadow demand, there are ways to mitigate
the effects. Most of these approaches rely on two basic ideas: 1) without
increasing overall inventory levels, re-balance the inventory to focus stockouts
on products where they will do the least harm, and 2) induce customers to
accept substitutes and backorders, rather than going elsewhere.
Inventory-Oriented Approaches
· Shift stockouts toward less competitive items. In other words, carry
greater stocks on highly competitive items (oxford shirts, for example) and
lower stocks on items which are unique to your firm (hand-painted Italian
vases).
· Learn more about why your customers buy each item. If you can identify
ones whose timing is urgent, you can increase coverage for those items and
reduce coverage on less urgent ones.
· If you are using an inventory coverage model (a formula-based approach
for deciding how much to buy), increase the assumed backorder cost to allow
for replacing customers lost due to stockouts. For example, if your cost
of a new customer is $65, and there is a 10 percent chance that stocked-out
customers never come back, add another $6.50 to your backorder cost.
Approaches to Induce Customers Not to Go Elsewhere
· Increase your capability for suggesting substitute products. A few
companies have actually built this capability into their order entry systems.
When an item is out of stock, the system presents a list of alternatives
to the telephone sales-person. (Be careful that the substitutions are good
ones, or as the chart above shows, the customers may become doubly unhappy.)
· Use Federal Express, drop shipping, and other similar devices for
fast delivery once the item becomes available. In other words, look for ways
to minimize the customers inconvenience. Many companies do just the
opposite, using low-cost slow methods for shipping backorders.
· Adopt a strong Truth in Backordering policy. Over-optimism
on delivery dates just adds insult to the injury of a stockout. Think of
your own reactions, for example, when airlines post a new departure time
for a delayed flight. Dont you tend to add a cushion for over-optimism?
And, dont you resent their apparent lack of truthfulness?
· For long delays, take the initiative in shifting to a posture of,
Well call you as soon as it comes in.
· Some people have suggested compensating customers to induce them to
ac-cept backorders, such as by promising a bonus gift together with the delayed
shipment. How well this approach would work is an open question. (Any readers
with experience in this practice are invited to comment.)
· Consider making an arrangement with a non head-on competitor for backup
supplies when you are out of stock. Many industrial distributors have long
used this practice with good results.
Much more needs to be learned about what people actually do when they discover
the products they want are unavailable. Once this behavior is better understood
it will be-come easier to choose the right corrective actions. Enough is
known, however, to say one thing. By ignoring the problem, as most do, direct
marketers are costing themselves a great amount of needless customer frustration
and missed opportunities. Just shining a light on shadow demand would make
a considerable improvement.