Table of Contents
!
A Strategy Simulator for Newspapers:
Overcoming the "Silos" to Do Whole-Enterprise Planning
Gary B. Hirsch
Creator of Learning Environments
7 Highgate Road
Wayland, Massachusetts 01778 USA
1 (508) 653-0161
GBHirsch@ attbi.com
Vincent Giuliano
Electronic Publishing Group
241 Glezen Lane
Wayland, Massachusetts 01778 USA
1 (508) 358-2207
vgiuliano@ attbi.com
Melody Winnig
Electronic Publishing Group
241 Glezen Lane
Wayland, Massachusetts 01778 USA
1 (508) 358-2207
mwinnig@attbi.com
Abstract
Newspapers have faced a series of challenges for many years, not always successfully.
Declining readership has led to consolidation and downsizing, especially in the US. The
ability of newspapers to respond to these challenges has been compromised by having
their departments function as independent "silos" without much integration. Newspapers
began facing additional challenges in the 1990's as online delivery of news began to
compete with print. This paper describes a simulator designed to help newspapers
develop coherent strategies that integrate across the departmental silos and provide
direction at the enterprise level. The simulator also enables newspapers to examine
strategies for developing online businesses and making the transition to media
enterprises with newspaper and online components. The paper presents screens from the
simulator with a unique design that enables users to closely associate system behavior
with its underlying causal structure.
Key Words: newspaper, publishing, online services, simulators
Background: Transitions Facing Newspapers and Other Management C hallenges
At the time we created the Newspaper Company Simulator in the late 90s, the newspaper
industry in the US and Western Europe had been facing a series of challenges, not always
successfully. Readership had been steadily dropping - for almost 20 years in the US,
with readers migrating to other media, namely TV and Intemet. Demographics of the
readership had also been steadily shifting, with older people and retirees outweighing
younger more affluent readers to whom advertising is mainly directed. With these
declines, there had also been a long-term decline of advertising market share.
Advertising is the life-blood of big US newspaper companies, typically representing 80%
or more of revenue.
US newspapers had already undergone several rounds of downsizing: of staff, editorial
and news coverage, and even the page dimensions of the paper itself. Because of high
fixed costs, however, the limits of downsizing were being approached. Big-city daily
papers were hardest hit, while some small-town newspapers and weekly papers continued
to do well. Having no other alternatives to become profitable, a few prominent
newspapers had already closed their doors, like New Y ork Newsday.
For over 75 years, the history of the newspaper survival in big US cities has mainly been
one of cannibalism. In the 1930s the typical large US city had 3-7 daily newspapers
competing with each other. This is still true in much of Europe and Latin America. But
as newspapers have died off, the remaining stronger ones “ate” their businesses and
managed to survive. Today, with very few exemptions, big US cities either have one
newspaper or two newspapers run by the same management under a “joint operating
agreement.” The opportunities for survival-by-cannibalism mostly don’t exist any more.
Newspapers were among the first to start web services, and most newspapers had already
built online editions and web services by the late 90s. During those wildest days of web
expansion when we planned our model, some progressive newspaper executives
anticipated that newspaper companies would evolve into primarily being web companies.
Indeed, there is a strong parallelism of the traditional services of a newspaper and those
that can be provided dynamically online.
Accordingly, we built our model in part to allow modeling of such a transition. We set
out to model the typical online service of a newspaper company, but what online services
consisted of continued to evolve rapidly at the same time we were building our model.
As we added more features to represent what was going on in the online marketplace, we
soon found that the online part of our model was significantly more complex than the
newspaper part. We added variables to our model like Relevance of Web A ppropriate
Content, Database A pplication Effect on Personalization of Design, and Staff Effort
Required Per Unit of New E Commerce Applications, but there was no end to the
seeming need for new complexity and variables.. The newspaper part was stable,
newspapers themselves exhibiting virtually no changes in their business or operational
patterns. For the online part, however, we found ourselves confronted with a steady
series of new online trends, all heralded as being the wave of the future. We could not
tell which trends were fads, which permanent, and which ones were appropriate to
newspaper online services. These innovations included online city super-sites
independently trying to enroll every small business with its own web, online auctions,
online shopping malls, online auto, real-estate and employment super-sites, online music
exchanges, online money management and brokerage, etc. - even online gambling.
Although we introduced a few such elements into our model we stopped as the
complexity of the model increased and it became clear that the evolution was continuing.
The newspaper companies at the time were also confused by such online development.
Despite being early online entrants, they tended to pursue conservative policies by the
late 90s. There generally was a culture clash between the traditional management and
editorial staff of the newspaper and the younger, more innovative and ambitious people
hired to operate the newspaper website. Newspaper company executives mainly saw the
online service as an extension of their printed newspapers, and selling advertising as the
best revenue model for online. This revenue model did not work for the most part.
Charging for access was also unsuccessful. The online Wall Street Jounal was probably
the only main exception. Local newspaper online services were mostly money losers.
They enjoyed very little online market share compared to popular national services like
Yahoo!, e-bay, Quicken, AOL and MSN. There was a serious open question that still
exists today - how much to invest in traditional newspaper approaches, and how much to
invest in the new online technology. Since the online services were money losers, they
tended not to get the resources they wanted.
Very few of the newspaper online services were innovative or noteworthy. Therefore our
model was in fact not bad at representing what newspaper companies actually did online
- even though it did not undertake to represent what they could have done if they were
more innovative.
Finally, there was a serious resistance in newspaper companies to address their critical
issues of survival and needs to evolve. Newspaper companies had evolved over 100
years into organisms with highly-specialized, but compartmentalized departments (like
Editorial, Advertising, Circulation and Production), but with little change- oriented
brainpower that allowed them to balance priorities, or to sense the changes in
environment and evolve accordingly. The online service was yet-another such
department. Newspaper companies mainly watched what other leading newspaper
companies did, and imitated that. Look at a newspaper company in the US, Columbia,
Norway, China, India or Nigeria - they were all organized more or less the same, they all
went to the same conferences where they listened to one another, and they all tended to
think along more or less the same lines.
Background on the Simulator and E] Universal, Why It Was Developed
Our consulting experience with newspaper companies indicated to us that, if newspapers
were to survive in the longer term, they confronted a strong need to break out of their
boxes. There was a strong need for developing strategic perspectives for survival, for
systems thinking among management, and for communication among departments that
operated as isolated silos of responsibility. The circulation department could launch a
promotional drive that resulted in higher circulation which resulted in higher cost for
paper, production, and distribution, but without significant increase in advertising.
Readers might be happy, but the bottom line could suffer severely. The production
department could justify acquiring new presses in terms of needing fewer production
staff, daily production-cost savings, earlier closing deadline, and a higher quality product.
The advertising and circulation departments could also see an advantage to more and
better color. However, in the face of an overall declining business, the hundreds of
millions of dollars of debt burden required to acquire the presses could be ruinous.
Especially if the impact on increased advertising was minimal. The newspaper could
beef up its editorial department and create a higher-quality editorial product. There
would be more costs not only for the editorial department but for the greater circulation
and larger number of pages in the paper and more newsprint. It would seem that the
paper is prospering. But, if the local market for advertising is near-saturated, the
financial result could be disastrous.
Then, there were the very difficult issues of how and where a newspaper company should
invest:
« In presses and other production technologies, for the reasons indicated?
¢ In editorial staff to make a better product?
* In possibly making less and buying more editorial content?
* In better editorial systems?
¢ In promotions for the newspaper?
« In more extensive online services?
Also, each of these are areas where current levels of investment could be cut back.
Further, there was an issue of how to assess the interrelationships of investment decisions
and various financial factors, such as cash flow and impact of debt burden. The online
service itself posed additional issues that were difficult for newspaper people to fathom,
such as:
¢ The extent to which the editorial and online newsrooms should be coordinated or
allowed to operated independently
« Whether the free online edition cannibalized circulation from the newspaper?
And what to do about this, such as withhold articles from online until they appear
in print?
* Should selling online advertising be done by newspaper ad sales staff, or by
people with main allegiance to the on line service?
« How much to invest in online technology, operating staff, advertising, acquiring
information from third parties, or in brand new services such as online auctions?
Finally, and perhaps most central to all, there was the question of how to get managers in
the various newspaper and online fiefdoms to talk with each other so they could mutually
appreciate each other's viewpoints and come to overall strategies in the best interest of
the company as a whole.
Based on these and similar situations, we elected to build a simulator that simulates the
detailed operation of a newspaper company, including some peripheral aspects of
newspaper company operation, such as doing contract printing. We also elected to
simulate the operation of a typical basic newspaper online service. We needed to be able
to simulate the newspaper proper and the online service independently, and as they are
combined on the enterprise level
All of these matters had to be looked at over a period of years, and also anticipating
contingencies of the marketplace. What would happen if the price of newsprint goes up
significantly, as it does from time to time? What if the cost of maintaining debt goes
significantly up, or down? What if the migration of readers from newspapers to online
goes up or down? What if, for Intemational newspaper, the country’s currency is
devaluated thus greatly increasing the cost of newsprint?
El Universal of Caracas was and still is the leading newspaper in Venezuela. It competed
with six other daily papers in Caracas, and its market share of advertising was equal to
that of several of its largest competitors put together. The paper dates back to the
nineteenth century. In the early 90s the newspaper came under a new progressive
management that launched a series of studies and activities to modernize the paper and
make sure it remained in the forefront in its market. It also created one of V enezuela’s
leading web services. El Universal faced most of the kinds of issues identified above,
and became one of three partners in the SimV enture undertaking, along with , an
experienced international consultant in the newspaper industry, and who built the actual
SimV enture model. The newspaper was a sponsor of the SimV enture model and also
provided much of the operational insight and data needed to develop and test it.
The Miami Herald was also very helpful to us in the development of the model,
particularly its online aspects. The model was based in the first instance on what we
learned at El Universal, but was modified and upgraded also to encompass insight and
operational data provided us by the Herald.
Structure of the Newspaper Model and Simulator
There is a long history of using management simulators to teach System Dynamics and
Systems Thinking. Senge and Lannon describe the value of simulators or "microworlds"
as tools for organizational learning. Certain simulators such as People Express have had
extensive use for general management training and have been studied by Sterman for
their effect on learning. Simulators have been developed in specific fields such as health
care to help those in the field understand the implications of major changes (see Hirsch
and Immediato, Hirsch and Kemeny). Particular design issues such as the desired degree
of transparency in simulators have been studied by Grossler and by Machuca et al.
There have also been applications of System Dynamics to the newspaper and publishing
industries by Hall.
The model ties together the various functional areas of a newspaper company, which we
referred to above as separate silos, into an inter-related entity. Figure 1 shows an
overview diagram of the newspaper module and how the various functions such as
editorial, circulation, advertising, and production relate to each other. Editorial Content
is the principal determinant of the Perceived Value of the Newspaper for readers that, in
turn, determines the Weekly Circulation. Circulation drives revenue to a limited extent,
but, more importantly, drives A dvertising Content that is typically responsible for 80% of
a newspaper's revenue. Circulation also determines Production Costs for labor to run the
presses and distribute the paper and for the paper and ink that are used. The newspaper's
Net Income determines the levels of staffing and investments it can afford.
Media Company Simulation LU
Year: 2000 Newspaper Control
Strategy: news1
Perceived Value of Nempong
Figure 1: Overview of the Newspaper Model
The simulator’s interface makes it possible to see the behavior of key variables directly
on the overview diagram. In the simulation displayed in Figure 1, the newspaper
operates in a benign environment. Little is assumed to change, but there is a small
decline at end of the ten-year period, though the newspaper remains profitable. Why
does this happen in a stable scenario? +The answer lies in the aging printing plant
(reflected in A ge of Capacity) that becomes less flexible and less able to handle changing
formats, color, etc. This reduces the Editorial Quality and the Perceived Value of the
Newspaper to readers who buy fewer copies. Fewer lines of Advertising Content are sold
as Weekly Circulation decreases and Revenue and Net Income decline as a result.
Clicking on buttons on this overview screen also lets users "drill down" and understand
what's going on in the model in greater detail. Figure 2 shows a more detailed view of
the factors affecting Editorial Content and the Perceived Value of the Newspaper that is
reached by clicking on the Editorial button on the overview. This is also the way that
users can get to detailed results in each area and compare performance across several
simulations.
Meda Company Simulation
Year: 2000 Compare Strategies for Newspaper
Strategy: news]
Fraction a in Color
Product Flexiilty’ —#Perecived Value / egal
[oem SKS to Advertising Ratio’
Amerage Age of Presses!
Display Advertiang Content’
weal IN
Maxum Saxe of Paper Desired Fraction Edteeial
Investment in New Press Hiring Editorial Staf?
Retarement of Presses
Number of Extra Sections’
Investment n Eattonad Design and Systems
Figure 2: Detailed View of the Factors Affecting Editorial Content and Quality
The screen in Figure 2 shows that the Perceived Value of the Newspaper is partially
determined by Product Flexibility, the ability to change formats quickly (which requires
reasonably modem presses). The amount of Editorial Content and the Editorial to
Advertising Ratio are the other principal determinants. Editorial Content depends on the
size of the Editorial Staff and Investments in Editorial Design and Systems that affect
Editorial Productivity. The items shown in blue on the diagram in Figure 2 are all
decisions that will affect editorial performance and the Perceived Value of the Newspaper
to readers. Some are financial decisions such as Investment in New Presses while others
are policies such as the Desired Fraction Editorial.
Users can get access to these various decisions by clicking on the italicized buttons. The
one marked "Promotion, Pricing, and Content" takes them to the decision screen shown
in Figure 3.
Media Company Simulation i
‘Year: 2000 Newspaper Promotion, Pricing & Content Decisions
Strategy: news]
Metro Newspaper ‘Metro Subscription
Pies —
National Newspaper National Subscription
Pie a
Display Advertising Metto Subscriber
Retest panel) es |?) —
a b_
araen Ed $10M
Metto Promotions National Subscriber
@ perYea) Promotion
‘Ye
30 bw] 10M Garren) 0 be] 10M
National Promotions Hapa) Number of Extra
( pervean Seations
Matketing Expenditures
@ per Year)
mf omrene |
Desired Fraction
Editorial
50 pu] 10M (Fraction of Content) 0.37
[dette Resuts | [_ Avance One Year | Enterprise Allocation to Newspaper: $ 5.420 M
[__imestmen Decisions | [_canratconct | Newspaper Budget Available: $ 135.51 M
Figure 3: Decision Screen for Pricing, Promotion and Content
The "sliders" on the screen shown in Figure 3 let them set the (Metro and National)
Newspaper Price and Display Advertising Rates per column-inch, control (Metro and
National) promotions budgets, and decide whether to add Extra Sections among other
things. Other decision screens focus on investments and staffing. Users typically move
though simulations a year at a time, making decisions, moving forward, observing the
results, and making a new set of decisions based on those results.
Strategy for the Newspaper as a Whole: Overcoming Silo-L evel Thinking
With all of these decisions at our disposal, how can we craft a strategy that improves on
the lackluster performance we saw in Figure 1? In that simulation, obsolescing plant kept
us from competing effectively and Total Weekly Circulation, Total Ad Content, Revenue,
and Net Income all declined. One thing to try is revitalizing the paper by investing in
editorial design and systems and hiring editorial staff. Figure 4 provides an overview of
the results when this strategy is implemented.
This simulation assumes the same environment as the previous one. We try to make
things better by investing in Editorial Design and Systems ($1 million per year vs.
$240,000 needed to simply maintain initial level) and hiring 25 new Editorial Staff per
year. Most of the variables on the graphs go up as we might have hoped. The Perceived
Value of the Newspaper, Weekly Circulation, Total Ad Content, and Revenue all
increase substantially. The big surprise is that, despite these otherwise positive results,
Net Income falls and becomes a loss rather than just a weaker profit.
The graphs give us a clue to what is happening. Focusing on the Weekly Circulation and
Revenue side alone would suggest a favorable outcome, but why is Net Income going
down when everything else seems to be going so well? Total Cost is going up! Let's
drill down and look at what's going on in more detail. Figure 5 shows that Circulation
Revenue has gone up substantially in this simulation (News3-blue line), compared to the
previous simulation (News1-red line) as we would have expected it to. It is about $2
million higher than in the previous simulation. Figure 6, shows that Display Advertising
Revenue is also up over the previous simulation as the newspaper becomes more
attractive to advertisers as a result of the higher circulation. Advertising revenue is about
$3 million higher after the strategy has its principal effect. However Total Cost, as
shown in Figure 7 has doubled and increased by about $8 million. Circulation and
production costs are much higher. The result is lower Net Income. Notice the dramatic
drop in Net Income between years 4 and 5 as other variables are starting to increase. By
the end of the simulation, the newspaper is losing money.
Why do costs rise so dramatically? Circulation costs increase simply because there are
more papers to distribute. Higher production costs are due to a combination of more
papers being printed and an increase in the average size of the newspaper. The size of the
paper expands to accommodate the increase in Editorial Content and increased
Advertising Content attracted by the higher circulation. Display Advertising Revenue
increases somewhat as Weekly Circulation goes up, but is limited by the total size of the
newspaper advertising market. Display Advertising Revenue does not increase by
10
enough to offset the higher costs and Net Income drops. We've created a big problem by
focusing on one "silo", editorial, and ignoring the potential adverse effects on other areas.
Media Company Simulation
Figure 4: Overview of Results with Strategy of Investing in Editorial Design and Systems
and Hiring Editorial Staff
11
Pic Compare Strategies for Newspaper ines Compare Strategies for Newspaper
hah Cede toma ‘eek Ss beer beta bore
is el
(2 4 8 @ # t we ON ff oo
Tey
Ortaca rome we} Dap Atrrang Frese es}
Crate ee 8) Tephr AirrngHree ee)
Ciraioen Ferenc tet aly Avreeng Peres tet
Figure 5: Circulation Revenue Figure 6: A dvertising Revenue
Pi ericol 7 Compare Strategies Sar Newspaper Perak Ty ‘Compare Strategies for Newspaper
Sag le Sevege Taio hagh ba Newpge: “pentng bese
aN a ee ee
ale = Bd — *
aM a “ ~
Figure 7: Total Cost Figure 8: Net Income
12
What can we do to avoid this problem? The problem is a combination of too much
circulation and too little revenue. Let's look to another functional area of the paper for an
expanded strategy. Raising the newsstand price of the paper from 50 cents to 80 cents
will have the combined effect of increasing Circulation Revenue per copy and lowering
Weekly Circulation. This is also equivalent to saying we're creating a better editorial
product and people should be willing to pay more for it. Figures 9-12 show the effect of
adding this element, initiated at the beginning of Y ear 4, to the Investment in Editorial
Design and Systems and increase in Editorial Staff that were the focus of the strategy
used in the previous simulation (News 3). The environment in which the newspaper
operates remains the same as well.
This strategy has the desired effect in trimming Metro Circulation, as shown in Figure 9.
(In these graphs, the new strategy News¢ is in blue, the previous strategy News3 is in red,
and the original simulation News! is in green.) With lower circulation, Total Cost is
also lower as shown in Figure 10. As a result of the lower Total Cost, Net Income
(shown in Figure 11) jumps with the increase in newsstand price and does not decline as
it did in News3. It actually remains higher than in News1. Circulation Revenue had
remained roughly the same with the higher price offsetting the lower volume. Display
Advertising Content (shown in Figure 12) and Revenue also did not suffer too much
since they were at saturation levels for the market in News3 and only went down a bit as
circulation dropped.
Tear Wl eres ‘en | rerwrnersr men
Swag ed Senta newt
Tayi Serpe Ted Cat
Figure 9: Metro Circulation Figure 10: Total Cost
13
Yomi ry Compare Strategies fr Newspaper Son = (Comnpore Strategies for Nemspuaper
Steg: eard Seritegy: eons
Trak be erga “pee bea Grek Lepr Coat
e 1
——
wv i —-
dont
r 3 2 oy a %
Toe Dat bite
Scheie etentedel
Sartre ae
ee a ——— =
oe rho poprenieay
Figure 11: Net Income Figure 12: Display Advertising Content
These results were much better than the disaster in News3, but produced only marginally
better Net Income than in the first simulation, News1, where technological obsolescence
caused profits to erode over time. What can we do to improve Net Income? If raising the
newsstand price helped a bit, why not try raising the Display Advertising Rate instead?
Much more of the newspaper's revenues come from advertising than circulation. An
increase in advertising rates should produce proportionally more Revenue. Figure 13
provides an overview of the results when we add an increase in Display Advertising
Rates from $50 to $70 per column-inch at the beginning of Y ear 4 to the Investment in
Editorial Design and Systems and the increases in Editorial Staff. Newsstand prices are
left unchanged in this simulation rather than being increased as they were in News4.
The results demonstrate that not all price increases are created equal. The increase in
advertising rates has a terrible impact where the increase in newsstand prices had a
beneficial effect. Why? Most of the key variables displayed in Figure 13 peak about in
the middle of the simulation after the rate increase for display advertising is introduced.
The Perceived V alue of the Newspaper to its readers drops despite the fact that
investments have been made in editorial systems and staff. This is followed by a decline
in Weekly Circulation and Total Revenues. Even though Total Costs go down, Revenues
drop faster and Net Income literally falls off the charts. Total Ad Content, rather than
dropping a bit in response to the higher rates, drops essentially to zero.
14
Year: 1999 Newspaper Control
Figure 13: Overview of Results with Investment in Editorial Design and Systems and
Hiring of Editorial Staff Together with Increase in Display Advertising Rates
What caused this sequence of events? The amount of Editorial Content in each day's
edition at most newspapers is determined by the amount of Display Advertising Content
purchased by advertisers. The decrease in Display Advertising Content caused by the
higher display advertising rates causes the amount of Editorial Content actually published
to be reduced, the Perceived Value of the Newspaper to readers to go down, and Weekly
Circulation to decrease. This causes Display Advertising Content to drop further
because the newspaper is less attractive to advertisers, especially at the higher rates.
These shifts set off a downward spiral in Display Advertising Content, Editorial Content,
and Weekly Circulation that overrides the impact of the editorial systems investment and
increased staff and produces a severe decline in Net Income.
Figure 14 shows how Display Advertising Content (now the blue line) in this simulation
(News 4b) declined compared to previous simulations. The decline began immediately
15
after the rate increase at the beginning of Y ear 4 and accelerated later in the simulation as
a result of the downward spiral. Figure 15 shows how the Perceived Value of the
Newspaper to readers (again, the blue line) declined in lockstep as declining Display Ad
Content drove Editorial Content downward as well.
This result suggests that the impact of changes on a system is extremely dependent on
where in the system those changes occur. While newsstand and advertising price
increases might be expected to have similar positive effects, higher advertising rates had
the unintended consequence of setting off a downward spiral because of an indirect effect
on editorial content. Having less space devoted to editorial content offset benefits from
improvements in editorial quality.
Sen 08 _ pererner sree Sen 08 a.
Sti apy Ay Coo Cag Pct Ta deg
Figure 14: Display Advertising Content Figure 15: Perceived Value of the Newspaper
The discussion up to this point is intended to be illustrative rather than to describe a
particular newspaper's experience. The examples are for an average-sized hypothetical
newspaper conforming to the industry-average parameters that we determined as a result
of our research. They also reflect default environmental assumptions. Our model can be
readily modified on two levels. Prior to and during a simulation, it is possible to modify
the environmental variables to account for such things as changes in the cost of
newsprint, currency impacts, and shifts in readership to other media. Ona more basic
level, over a hundred of the model’s parameters can be modified to reflect the conditions
of a specific newspaper by means of spread-sheet input. The challenge here is not in
modifying the model, but in getting a newspaper to look deeply into itself and its
16
information systems to come up with the appropriate parameters for the model. The next
section illustrates the value of being able to change environmental assumptions.
Operating in a Harsh Environment
These simulations so far have assumed a benign environment. But the discussion earlier
in the paper suggested that the newspaper industry is under pressure from many
directions and is likely to lose ground to other media. This harsh environment assumed in
the next simulation (Harsh1), reflects the gradual migration of readers and advertisers to
the Internet. It assumes that readership shifts to online news sources by 1.5% per year
and advertisers (both display and classified) shift to online sites by 4% per year.
Otherwise, all other assumptions are the same as in News 1 shown earlier including the
gradual aging of presses in News1? The results are shown in overview in Figure 16.
Media Company Simulation I
Year: 1999 Newspaper Control
Strategy: Harsh1
Total Cost
Figure 16: Newspaper Operating in a Harsh Environment (with Migration of Readers and
Advertisers to Other Media)
17
One might expect Weekly Circulation and Display Advertising Content and Revenues to
slowly degrade by roughly the same amounts as the migration in readership and
advertising. Instead, there is a severe crash in all numbers around the middle of the run
after several years of more gradual change. Why? This is a result of feedback loops
through Weekly Circulation, Display Advertising Content, Editorial Content, and
Perceived Value of the Newspaper and the downward spirals that they set off. Notice
how the slow declines gradually make the newspaper less attractive to both readers and
advertisers who get to a point where they suddenly start abandoning the newspaper in
greater numbers. The "bounce" in several of the curves such as Perceived Value of the
Newspaper is a result of the paper hitting its minimum size of 48 pages. At this point, the
Ratio of Editorial to Advertising increases (only because advertising continues to
disappear) and this helps to improve Perceived Value of the Newspaper and Weekly
Circulation and stops the slide created by the positive feedback loops. Unfortunately, this
occurs at a point where Net Income is negative and the paper is beyond saving.
These runs are a good illustration of the value of a causal model emphasizing feedback
such as this one vs. a spreadsheet that does straight-line projections. A spreadsheet
model might suggest only gradual changes over the entire ten-year period and result ina
complacent management attitude that leaves them unprepared for the sudden decline.
Perhaps the strategy we used earlier, labeled News4, could also provide some
improvement over this harsh scenario. This was the strategy in which Investment in
Editorial Design and Systems and hiring of Editorial Staff was supplemented by an
increase in newsstand price from 50 to 80 cents. Implementing this strategy (Harsh3) in
the harsh environment yields the results shown in Figure 17 on the next page. Rather
than the improvement that News4 showed over News1, Harsh3 shows no apparent
improvement over Harsh1. All key variables decline over the simulation. Figures 18 and
19 provide direct comparisons with of Harsh3 (blue line) with Harsh! (red line) for Metro
Circulation and Net Income. The effect of this strategy in a harsh environment is merely
to delay the inevitable rather than to improve performance. In fact, Harsh3 results in
even lower Metro Circulation and Net Income.
18
Year: 1999 Newspaper Control
Total Cort
Figure 17: Results with Investment in Editorial Design and Systems, Hiring of Editorial
Staff, and Increase in Newsstand Price in a Harsh Environment
19
Yeu 1999 == Compare Strategies far Newspaper Son = Coanpare Strategies for Newspaper
Strang: Barks: Sees: Hart
ebls Vad lrg, Coden igh e Pregeps “pay bene
cod toe es |
» 2 @ @ 2 ew & MW Oo tos § 42 @ 2 w
‘Votiy Mew: Conon Tart!
‘Vediy Meme Cocke Tat
pass ___3 i}
tian | “tenatmat | “senteanin |
Figure 18: Metro Circulation Figure 19:Net Income
Again, the vicious cycle created by the decline in readership and advertising creates a
situation in which “old solutions" no longer work. It suggests that the high percentage of
fixed costs in newspapers that make them profitable in stable markets leaves them very
vulnerable to declines in circulation and advertising. It also suggests that viewing a
newspaper in isolation is dangerous because it limits options for dealing with changes
such as shifts to the Internet. Solutions that work probably need to consider the
newspaper as part of a system, a larger media enterprise.
The Media Enterprise and Newspaper Diversification: The Case of Online Services
What frameworks and tools can help newspapers create viable strategies when they face
such a difficult environment? Thinking of themselves as media enterprises and planning
a diverse set of services provides a much larger number of degrees of freedom for
planning. Figure 20 depicts an enterprise-level view. Moving up to the enterprise level
enables us to deal with an entirely new set of strategy questions involving diversification
toward other media such as radio and TV stations, other print media such as magazines,
and online services. It shows, for example, how a newspaper might interact with a family
of online services, the focus of our work in the late 1990's when it appeared that there
might be a major transition from print to online for metropolitan newspapers all over the
world. With some modification, any such diversification strategy can be examined in a
similar manner. While the online "boom" of the 1990's did not pan out as expected, this
capability can be helpful to newspapers for any diversification strategies. Some of the
20
principles we discovered relating to online services could, we suspect, be applied to any
other form of diversification.
Modes Company Simulation
Enterprise Control
= aw
Figure 20: Enterprise-Level View of a Newspaper Company
What would be the result if a newspaper made a series of investments in online services
with the intention of capturing the readership it loses under the set of simulations above
with assumptions about a harsh environment? There is a decision screen at the
enterprise level for creating diversification strategies. For example, it enables us to set up
an "investment fund" based on a fraction of profits and then allocate the proceeds of that
fund to the online activity, other activities such as local community newspapers, and the
metropolitan newspaper itself. That decision screen also deals with management hiring,
allocations of management time, and other issues alluded to earlier such as the degree of
integration between the editorial activities of the newspaper and online activity.
Figure 21 displays an overview of the results of a simulation in which there is an
investment fund that starts off at about $10 Million and is added to by 10% of profits
21
each year. In this strategy, we allocate only 10% of the fund to the online business, 10%
to other businesses such as community-level newspapers, and 80% reinvestment in the
newspaper itself. This could reflect the investment patterns of a conservative,
newspaper- dominated media enterprise. We fund the strategy as a newspaper with only a
half- hearted commitment to the Internet might. (The conversation in the Boardroom
might go something like this. "A consultant has suggested that we must grow an online
business, but we don't want to disappoint our stockholders and we must reinvest in the
newspaper to protect it from this online nonsense. Besides, if we invest less in online, it
only means that business will become profitable a bit later.")
Media Company Simulation
Year: 1999
Strategy: Onlinel
Figure 21: Overview of Online Module with Conservative Investment Strategy
The screen shown in Figure 21 focuses on the online business just as the newspaper
control screen shown earlier focused on the traditional print business. Figure 21 shows
how investments in staff, equipment, and purchased services enable the online business to
create Offerings for Users (content) and Quality of Experience which, in tum, attract
22
Frequent Visitors and Regular Users who generate Total Traffic. Traffic, in turn, attracts
and holds A dvertisers and Online Vendors who help the business generate Revenue and
Net Income.
The strategy involves goals for hiring what appears to be respectable numbers of staff
spread across different functional areas such as technical development and online
editorial and for developing lucrative revenue-producing services such as online vending
(e-commerce) and online classifieds. However, budget constraints built into the model
keep the online activity from meeting all of its goals.
The amount available to online under this low funding scenario (including its own
revenue stream which is low at the beginning) will be considerably less than the budget
calls for and the actual spending, hiring, activities developed, etc. will be reduced by the
budget constraint. The result is a simulation in which there is a brief peak in online
activity that dwindles later in the run. Offerings to Users top out and cause Total Traffic
to peak and then decline. While there are a number of Frequent Visitors, limited content
constrains the number of Regular Users who are the truly important generators of traffic.
Limited traffic and a failure to create revenue-producing activities limits Revenues from
the online business. Income produced by this business never makes it to break even
during the simulation.
For the business to be successful, online staff must grow rapidly in order to produce
growing amounts of content and applications to attract users in an environment where
these offerings quickly become obsolete. Online staff does not grow rapidly enough, in
this simulation with too little funding, to overcome this rapid obsolescence and produce
content and quality of experience that can keep up with competitors' growth rates in
offerings and performance that are assumed to be 25%. This staffing shortfall results in
falling market share and traffic that stalls and then drops as our online business falls
behind its competitors. Revenues peak and then drop as well, depriving the online
business of an alternative source of funds for investment. These declines are
compounded and accelerated by staff becoming disenchanted by slow/no growth and
leaving and having their productivity reduced by turnover.
Rather than a partial success, the inability to achieve a critical mass results in decline and
essentially a complete failure. The feedback loops through staffing, content and
applications, traffic, commercial revenues, and further investments in staff that would
produce growth in a successful simulation help to hasten decline and failure if under
investment keeps a critical mass from being realized. In the next simulation, we try a
strategy in which 50% of the investment fund is put into online rather than the 10%
invested previously. Will this be enough to achieve critical mass and profitability?
As indicated, everything in this simulation (Online 2) is the same as the in previous one
(Online1), but 50% of the investment fund is devoted to online rather than 10% and
nothing is invested in the "other" businesses. The effect is to be able to fully fund the
ambitious staffing strategy and grow staff rapidly in all of the critical functions. As you
can see in Figure 22, the results are dramatically different with a successful online
23
business that reaches a level of profit comparable to that of the newspaper at the
beginning of the run. The newspaper is losing significant money by the end of the run, so
online profits are essential for survival. Online profits are coming on about half the
revenues that the newspaper required to produce the same level of profit since there is no
need for presses, paper, trucks, etc. By the end of this simulation, the total online staff is
around the same size as the newspaper's editorial operation alone.
Meda Company Simulation
Year: 2000 Online Control
Strategy: Online?
Figure 22: Overview of Online Business with More Aggressive Investment Strategy
Why is Online? is a profitable run where Online] is a disaster? The comparative graphs
in Figures 23-26 highlight the differences between the two strategies in more detail.
Depth and Breadth of Offerings is significantly greater for Online2 (blue line) than
Online 1 (red line). Asa result, more Regular Users are attracted and Total Traffic and
Revenues are higher and there is a considerable jump in Net Income. Under the detailed
graphs, examine the one under "Usage" that shows the relationship of Market Share and
Ratio of Offerings to Competitors to see how they compare for each of the runs (much
higher in Online2). The feedback loops through staff, offerings to users, traffic, revenue
producing applications, revenue, and the ability to fund more staff (that helped to speed
ae —— ers
24
Ten yy ‘Compare Strategies foe Online
Searg dee! Seong: ne!
Sr ea aig fg danke tach vie
a mu a
i] @aAeenaea xu
Tt) aE aes au
ve Vl ou po =~
Pod ail
1 L
' t s 2 a ® z a * a @ ‘ if * x a ° 2 s s a i»
ede
both wilt tiny (hee! —
brat mpeg (ie! —
Figure 23: Breadth and Depth of Offerings
a) = ( pes fr Oniee
‘santas ‘oapare Sraegies
‘ep agen
Figure 25: Regular Users
Figure 24: Total Traffic to Site
ry (Corupare Strategies far Online
‘Yor
‘Srey: (hed
Catia Ca Bee
a“
Figure 26: Online Net Income
25
up the decline in Online 1) produce spectacular growth in this simulation that helps to
assure the success of the online site.
This strategy also succeeds because it is well-balanced, hiring sufficient numbers of
people in each department and allocating their time well across different functions. The
result is that traffic develops at the rate it needs to and revenue-producing activities are in
place at the right time to harvest dollars from that traffic.
As we indicated earlier, the results achieved in Online2 were very rare for the online
businesses of newspaper companies. The purpose of this example was to show how the
simulator can be used as a tool by newspaper companies to examine diversification
strategies. It also illustrates a principle that applies to any type of diversification that
might be attempted by a newspaper. New businesses being established in competitive
markets must have a life of their own and the resources required for them to grow and
thrive. They cannot be viewed merely as appendages of the print business.
Newspapers’ Reactions to the Simulator
The most extensive use of the simulator was made at El Universal in Caracas that served
as the site of its development. A workshop using the simulator with management staff
had exactly the desired effect in helping the staff to see the limitations of silo-level
thinking and the importance of strategies conceived with the entire enterprise in mind.
The simulator also was used for more limited workshops and demonstrations at a number
of other newspapers and chains. At the Miami Herald, in a session with the group
responsible for its online services, there was a sense that the model replicated some of
their past experience very well. The Herald's online staff also felt that the model offered
deeper insight about why things happened as they had in the past and what strategies
might help them going forward.
In another workshop with the staff of the Belo newspaper chain, there was also the sense
expressed that the simulator could be a valuable tool for helping newspapers better
understand the online services business and the requirements for an effective transition to
it. There were additional presentations to the staffs of the Arizona Republic and Arizona
Central Group of newspapers, the Trinity Mirror Group, and potential industry partners
such as A tex Corp. and Northern Light. All of these meetings produced favorable
reactions that this was a new approach that could help newspapers deal with the many
strategic problems they were facing.. Presentations at the annual conference of Ifra, an
international association of newspapers, in Amsterdam in October of 2000, were also
enthusiastically received.
Despite these favorable reactions and the potential uses of the simulator, there was not
the widespread adoption that its developers had hoped for. There may have been several
reasons for this. One is that the anticipated shifts from print newspapers to online
delivery of the news materialized to only a very limited extent. Planning for this shift
had been an emphasis of the simulator and a hoped for selling point. Another reason is
that newspapers, by their very nature, are operationally oriented and focus on getting the
26
paper out every day. There is only limited appreciation for strategy and for long-term
thinking. It was therefore difficult to get them to spend money on a tool for developing
strategy.
Finally, though newspapers are basically the same kind of business throughout the world,
they do have some important variations in their products and how they are manufactured
and distributed. Reflecting the differences among newspapers was important for the
simulator to be useful as a planning tool, which would increase its value beyond simply
being a learning environment. Newspapers we spoke with wanted a certain amount of
customization, but were unwilling or unable to pay for custom model building.
Expanding the set of pre-programmed options offered in the simulator would have taken
more time and money. The latter was in short supply since the simulator was being
developed at a time when the venture capital bubble of the late 1990's had burst and
capital was hard to come by for Intemet-oriented ventures. The lack of capital also kept
us from engaging in the lengthy customer education and selling effort required for any
new technology that is unfamiliar to an industry.
Conclusions
This paper has described a simulator for evaluating newspaper strategies. It's principal
contribution is to allow newspaper companies to overcome the traditional view of
managing a set of functional silos and enable them to develop strategies for the entire
enterprise based on how those functional areas are likely to interact over time. They can
also test strategies under a variety of different circumstances to see which ones are most
resilient and don't depend on conditions being "just right" in order to succeed. One
purpose for the simulator's development, managing the transition of the print newspaper
to largely online services, has proven less relevant with the passage of time. The online
example has demonstrated that a simulator can nonetheless be a valuable tool for
newspapers that want to diversify their business activities and need to understand the
investments and policies required for the new venture to succeed and also how the
venture should interact with the existing newspaper business.
Acknowledgments
The authors would like to thank Mr. and Mrs. Andres Mata Osorio, publishers of El]
Universal in Caracas, for their support of this work and the staff of El Universal for their
help in formulating and testing the model. We would also like to thank J. T. "Tom"
Johnson, Professor of Journalism at San Francisco State University and co-director of the
Institute for Analytic Journalism for his contributions to our work.
References
The simulator described in this paper and results of a number of simulations can be
downloaded from the following web site: http://cronkite.pp.asu.edu/jagis2/simvenweb3/ The site
also includes other relevant information on the newspaper industry.
27
Grossler, A. Structural Transparency as an Element of Business Simulators. In 16th
Intemational Conference of the System Dynamics Society, Quebec '98, (pp. 39). Quebec
City, Canada: System Dynamics Society.
Hall, R. L., A System Simulation Model of a Magazine Publishing Company. In 1973
Summer Computer Simulation Conference. Montreal: Simulation Councils, Inc. (P.O.
Box 2228, La Jolla, CA 92037).
Hall, R. I. (1974). Managing a Magazine Publishing Company--A Decision Making
Game. In Constructing Instructional Simulation Games (pp. 22-29). U. of Manitoba,
Natural Resource Inst.
Hall, R. I, A System Pathology of an Organization: The Rise and Fall of the Old
Saturday Evening Post. Administrative Science Quarterly, 1976, 21(2), 185-211.
Hirsch, G. B., & Immediato, S. Design of Simulators to Enhance Learning: Examples
from a Health Care Microworld. In 16th International Conference of the System
Dynamics Society Quebec '98, (pp. 2). Quebec City, Canada: System Dynamics Society.
Hirsch, G. B., & Kemeny, J. M.. Mastering the Transition to Capitation. Health Care
Forum Journal, 37(3 May-June), 89.
Machuca, J. A. D., Castillo, J.C. R. d., Carrillo, M. A. D., & Zamora, M. d. M. G.. Our
Ten Y ears of Work on Transparent Box Business Simulation. In 16th International
Conference of the System Dynamics Society Quebec '98, 1 (pp. 5). Quebec City, Canada:
System Dynamics Society.
Senge, P. M., & Lannon, C. Managerial Microworlds. Technology Review (July, 1990).
Sterman, J. D.. Teaching Takes Off: Flight Simulators for Management Education.
OR/MS Today (October, 1992), 40-44.
Sterman, J. D. (1988). People Express Management Flight Simulator: Simulation Game,
Briefing Book, and Simulator Guide. Available from author, MIT Sloan School of
Management, Cambridge, MA 02139.
Sterman, J. D. (1988). Strategy Dynamics: the Rise and Fall of People Express (Memo
No. D-3959-1). MIT System Dynamics Group.
Back to the Top
CC BY-NC-SA 4.0