March/April 2015 // Public Gaming International //
55
Now Showing: Today’s most
popular games
by Dr. Stephen Wade
Lottery managers want to be nimble
and responsive to the market, but they
are not getting the information they
need to do this well.
What happened last month may be pretty well reflected in the
record of pack activations and settlements. These are accounting
transactions that tell about what our retailers did. However, if
we want to know what our players are doing right now, we need
to look closer to the player and we need to look sooner. To be
nimble we need to know where the retailers are going to end up
at the end of this month, not last month. The retailers are going
to end up where the players are taking them.
In the old days, the amount of IT work required to get real-
time insight into what players are doing may have been pro-
hibitive. That is not much of a challenge now. The challenge is
to adopt and use more meaningful measures. And by “use,” I
mean applying the information to do something different than
we would otherwise have done. Better information gives us con-
fidence to put tickets where they are needed. For example, know-
ing which instant games are being consumed faster by players,
and then being able to take real time actions to adjust inventories
and facings, can translate into higher sales and more efficient
inventory flow; this is what the Power of Now means to me.
Using current business intelligence on what players are doing
helps lotteries avoid two obvious problems: being “out of stock”
on games that are moving quickly, and being overstocked on
games that are moving slowly. Empty bins that should be serv-
ing the most popular game are a glaring lost opportunity. At the
same time, too many packs of less popular games will eventually
be returned for credit and may end up being shredded.
In my work, I have defined a quantitative way of measuring
popularity that has turned out to be very useful. I have called this
metric the “Popularity Index.”
The “Popularity Index” metric is based on the familiar retail
concept of “turn rate.” Retail turn rate compares the rate of sales
to the value of inventory. People who run stores buy inventory
they sell to customers. At any given point, retailers have a mea-
surable amount of money tied to the inventory that they carry;
or the value of inventory. When the inventory is sold, retailers
often talk about the rate at which they turn inventory into cash,
or “Turn rate.”
The turn rate is much more meaningful than a simple rate of
sales per week. Consider how it can be used to compare retail-
ers: $5,000 of sales per week in a store where the average value
of inventory is $2,500 is a turn rate of 2.0, and that is stellar.
Five thousand dollars per week from a chain that averages $20
thousand of inventory is a turn rate of 0.25, and that may be
a concern. High turn rates occur when inventory quantity and
product selection most closely match consumer desire. Low turn
rates often occur when we have too much inventory of products
that are less popular.
Just as we recognize that the proprietor of the one-cooler shop
with a turn rate of 2.0 (in this case with units “per week”) is doing
a better job of managing lottery inventory than the big chain, we
can recognize that a game with a turn rate of 2.0 is doing a better
job of moving through the retail pipeline than a game with a turn
rate of 0.25 per week. When we use turn rates to compare games,
we are talking about the rate at which players are converting the
inventory held by all retailers into cash. Players may be buying
$250,000 per week of two games. However, selling this volume
is a bigger accomplishment if the value of the game’s inventory
across all retailers is $125,000, than if it is $1 million.
When we compare games, we usually compare within a mental
category; “Which of these $5 games are we going to need more
of?” To answer this kind of question, and to make the whole
business of measuring popularity more intuitive, I defined the
Popularity Index as the turn rate of the particular game, divided
by the turn rate of the category to which it belongs. A game of
average popularity has, intuitively, a popularity of 1.0 and the
metric has no dimensions.
How we get to this number
When retailers talk about sales, they are usually talking about
sales to their customers. When lotteries discuss instant ticket
sales, the language may have variations. When some lotteries say
“we sold” they mean “the retailer agreed to pay” (this is activa-
tions called sales). Other lotteries say “we sold” when they mean
“the retailer paid” (this is settlements called sales). Activations
and settlements define an accounting view, but they do not di-
rectly reflect what the retailer’s customers are doing.
By knowing that 10,000 winning tickets of a game called “Lucky
Diamonds” have been validated for prize payment this week, and
that the game has odds of 1 in 3.0, we can reliably calculate that
about 30,000 tickets of that game have likely been played this
week. We may estimate more accurately if we take a small rate of
unclaimed prizes into account. Some lotteries monitor this and
call it “likely sales” or “validation-based sales.” By expressing likely
sales over a period of time, we get very close to what a retailer
would call rate of sales. We might define our units as “tickets per
week” (30,000 per week) or “dollars per week” ($300,000 per
