Direct Network Effects:

• Definition: Direct effects occur when the value

of a product or service increases as more people
use it.
• Example:
o Telephone: A telephone was almost useless
to the first person who owned one because
they had no one to call. However, as more
people got telephones, the value of owning
one increased significantly because each
additional user meant more people you
could communicate with.
Indirect Network Effects:
• Definition: The value of a product increases not
just because of the product itself, but due to the
availability of complementary goods or services.
• Examples:
o Computers and Software: A computer
without software is of limited use. The real
value of a computer comes from the
availability of various software that can run
on it. As more software becomes available,
the more valuable the computer becomes.
 o Smartphones and Tablets: Similar to
computers, the value of smartphones and
tablets increases with the availability of apps
and other services that enhance their
functionality.
WHERE DOES THE VALUE COME FROM?
1. Exchange:
o Explanation: Every product or service that
benefits from network effects facilitates
some kind of exchange, whether it’s
communication, transactions, or
information sharing.
2. Staying Power:
o Explanation: The more users a network has,
the more likely it is to stay relevant and
continue growing. This is tied to:
 Switching Costs: The difficulty or cost of
switching from one network to another
can lock users into a network, thereby
strengthening the network’s value as a
strategic asset.
3. Complementary Benefits:
o Explanation: These are additional products
or services that increase the value of the
 primary network. For instance, a gaming
console becomes more valuable as more
games (complementary goods) are made
available for it.
o Platform Products: Products or services
that encourage the creation of
complementary goods are often referred to
as platforms.
COMPLEMENTARITY AND COMPATIBILITY
• Combinable Network Goods:
o Some products naturally work well together
without much effort (e.g., a phone charger
with a phone).
• Complex Products:
o Complex products often require specific
technical standards to be compatible with
other components. For instance, computer
software must be compatible with the
hardware and operating system to function
correctly.
• Compatibility Options:
o Companies can choose to make their
products fully or partially compatible with
others, or they can opt for proprietary
 designs that refuse to interconnect with
other brands (e.g., Apple's ecosystem).
ONE-SIDED OR TWO-SIDED MARKETS
1. One-Sided Market:
o Definition: In these markets, most of the
value comes from a single class of users.
o Example: A social network where all users
are in a single category (e.g., individual users
on Facebook) benefits from same-side
exchange, where more users increase the
value for everyone.
2. Two-Sided Market:
o Definition: These markets involve two
distinct categories of participants.
o Cross-Side Exchange Benefits: When an
increase in users on one side of the market
boosts value for the other side, like in credit
card networks where more cardholders
attract more merchants, and vice versa.
MODELS FOR NETWORK EXTERNALITIES
1. Sarnoff’s Law (V ∝ N):
o Explanation: The value of a network grows
directly with the number of users.
 o Example: A TV network becomes more
valuable as more viewers tune in.
2. Metcalfe’s Law (V ∝ N²):
o Explanation: The value of a network
increases proportionally to the square of the
number of users.
o Example: Social media networks like
Facebook become exponentially more
valuable as more users join, because each
user can connect with many others.
3. Reed’s Law (V ∝ 2ᴺ):
o Explanation: The value of a network grows
exponentially as new subgroups form within
it.
o Example: WhatsApp groups where
subgroups can form, increasing the overall
network value significantly.
4. Widely Used Model (V ∝ N Log N):
o Explanation: This model suggests a more
moderate growth in value as a network
expands, compared to the sharp increases
suggested by Metcalfe’s or Reed’s laws.
 o Example: Online marketplaces like eBay,
where the value increases with more buyers
and sellers, but not exponentially.
VALUE OF A NETWORK: BROADCASTING
NETWORKS
• Sarnoff’s Law: Value ∝ N
o Explanation: The value of a broadcasting
network is directly proportional to the
number of viewers or listeners.
VALUE OF A NETWORK: CONNECTING PEERS
• Fully Connected Network:

• Metcalfe’s Law:
o Explanation: This law is applicable to
telecommunication networks, email, and
unicasting, where the value increases with
the square of the number of participants.
VALUE OF NETWORKS: GROUP FORMING
NETWORKS (GFN)
• Group Collaboration:
o Explanation: In these networks, the
collaboration value (interaction within
groups) is more critical than merely being
connected.
• Reed’s Law:
o Explanation: The value of a GFN increases
exponentially as the number of subgroups
that can be formed grows.
o Example: Peer-to-peer networks like
Facebook, group-based auction sites like
eBay, online chat groups, etc.
• Internet:
o Explanation: The Internet, as a network of
networks, has a value much higher than
individual broadcast or telephone networks
due to the vast number of possible
connections and interactions.
STRATEGIES FOR COMPETING IN MARKETS WITH
NETWORK EFFECTS
1. Move Early
2. Seed the Market
3. Subsidize Adoption
4. Expand by Refining the Market
5. Alliances and Partnerships
6. Encourage Development of Complementary
Goods
7. Rivals: Be compatible with the leading network.
8. Incumbents: Close off rival access and
continuously innovate.
TOO MUCH NETWORK EFFECT
• Congestion Effect:
o Explanation: When a network becomes too
large, it can lead to congestion, affecting the
quality of service (QoS). This can cause users
to drop out.
o Examples: Free internet services that
became too crowded or the initial public
offering (IPO) of Facebook when the
platform's growth led to performance issues.
NETWORK EFFECT PROMOTES STANDARDS
• Examples:
o Sony’s Betamax vs. JVC’s VHS: VHS won
despite being technologically inferior
because it was more widely adopted due to
less restrictive licensing.
o Sony’s Blu-ray vs. Toshiba’s HD-DVD: Blu-
ray won due to better adoption by
manufacturers.
o ITU’s H.323 vs. IETF’s SIP: SIP won due to its
ease of implementation and better
integration with various systems.
o WiMax vs. LTE: LTE won over WiMax despite
WiMax being technologically superior
because it was incompatible with other
networks.
--- Diffusion Theory: Bass Model ---
SEGMENTATION OF ADOPTERS
• Explanation: Different groups of adopters can be
segmented by factors such as social and
economic status, risk affinity, knowledge, and
interest in the product.
DIFFUSION EFFECTS
1. Intrinsic Factor (p):
o Explanation: This refers to something about
the product itself that attracts users,
independent of how many others have
adopted it.
2. Imitation Factor (q):
o Explanation: The network effect, where
more adopters increase the product’s value
or create an information-cascade effect,
encouraging others to adopt it.
3. Bass Model of Diffusion:
o Explanation: This model describes how the
number of adopters at any time ttt depends
on the intrinsic factor ppp, the imitation
factor qqq, the total potential market MMM,
and the cumulative number of adopters up to
time.
1. What is the Bass Diffusion Model?
The Bass Diffusion Model helps us
understand how a new product (like a new
phone or technology) spreads through a
population over time. It's used to predict how
 many people will adopt the product at
different stages.
2. Key Concepts:
• N(t): This represents the total number of people
who have adopted (bought or started using) the
product by a certain time.
• n(t): This is the number of people adopting the
product at a specific moment. It's like measuring
how many people bought the product today.
• M: This is the maximum number of people who
can adopt the product in the long run, or the total
market size.
• p: This is called the "innovation factor." It
represents the percentage of people who adopt
the product because they are naturally drawn to
new things or are early adopters.
• q: This is called the "imitation factor." It
represents how many people adopt the product
because they see others using it—like following a
trend.
3. How Products Spread:
• Non-Cumulative and Cumulative Graphs:
 o The graph on the left (non-cumulative)
shows how the number of new adopters
increases quickly, peaks, and then declines
over time. Early on, a few people adopt the
product, but as more people hear about it,
adoption speeds up. Eventually, almost
everyone who will adopt the product has
done so, and the number of new users starts
to drop.
o The graph on the right (cumulative) shows
the total number of people who have
adopted the product over time. It starts off
slow, grows quickly, and then flattens out
when nearly everyone who wants the
product has it.
4. The Equations:
• The Bass Model is described by two main
equations that help predict product adoption
over time:
o n(t) gives the rate at which people adopt the
product at time t.
o N(t) gives the cumulative number of people
who have adopted the product by time t.
• In simple terms, these equations are saying: “The
number of new adopters depends on the number
of people who haven’t adopted yet and how
strongly they are influenced by innovation (p) or
by others (q).”

Formula 1: The Rate of Adoption (n(t))

This formula tells us the number of new adopters

at any given time t. Let’s explain each part:
• n(t): The number of people adopting the product
at time t. It's measured as a rate, typically the
number of adopters per time period (e.g., per
day, per year).
• M: This is the market potential, or the maximum
number of people who will eventually adopt the
product. It’s a fixed number representing the
total size of the market in the long run.
• p: The innovation factor. This represents the
percentage of people who adopt the product
because they are early adopters or attracted to
the novelty. It’s measured as a proportion or
 probability (e.g., 0.01 means 1% of the
population adopts due to innovation).
• q: The imitation factor. This reflects how many
people adopt the product because others around
them have adopted it (i.e., peer pressure or
trend-following). It’s also measured as a
proportion (e.g., 0.3 means 30% of the people
adopt by imitating others).
• t: Time, which is usually measured in months or
years. t is the point in time at which we are
measuring the adoption rate.
• e^(−(p+q))^t: This is an exponential decay
function. It adjusts the formula based on how
much time has passed. Early on (when t is small),
this value is larger, and as time passes, it gets
smaller. It reflects how adoption slows down as
time goes on.
How This Works:
The formula shows that the number of adopters at
any time t depends on three key factors:
1. Market size (M): The larger the potential market,
the more people can adopt.
2. Innovation (p): A higher p means more people
adopt early on without waiting for others to do so.
3. Imitation (q): A higher q accelerates adoption as
more people are influenced by others' behavior.

Formula 2: The Cumulative Number of Adopters

(N(t))

This formula gives the total number of adopters by

time t, showing how the product spreads across
the population over time.
• N(t): The cumulative number of people who have
adopted the product by time t. It’s measured as a
count (e.g., total number of adopters).
• M: Same as before, the maximum market size or
total number of people who will ever adopt the
product.
• p: The innovation factor, which drives initial
adoption, measured as a probability.
• q: The imitation factor, which becomes more
important over time, also measured as a
probability.
• t: Time, measured in a similar way (months or
years).
• e^(−(p+q))^t: Again, this exponential decay
function adjusts for how much time has passed,
influencing how fast the number of adopters
grows.
How This Works:
• At the beginning (when t is small), the formula
predicts slow growth, driven mostly by p
(innovation).
• As time progresses, q (imitation) becomes more
influential, and the total number of adopters
increases rapidly.
• Eventually, as the market becomes saturated
and everyone who could adopt has done so, N(t)
approaches M (the market size), and the number
of new adopters decreases (as seen in the non-
cumulative graph).
Summary of Variables and Their Measurements:
Variable Meaning How It's Measured
N(t) Total Count of people
adopters by (cumulative number
time t of adopters)
n(t) New Rate (number of
adopters at new adopters per
time t unit of time)
M Maximum Total potential
market size number of adopters
p Innovation Proportion of
factor adopters driven by
innovation
q Imitation Proportion of
factor adopters driven by
peer imitation
t Time Measured in time
units (e.g., months,
years)

These formulas allow companies to predict how

many people will adopt a product at any given time
and how many people in total will have adopted the
product by a certain point.

5. Graphs with Different Values of q:

• The bottom two graphs show how the model
behaves with different values of q (the imitation
factor):
o When q is larger, the adoption rate grows
faster, meaning more people are adopting
the product because they’re imitating
others.
o The graph on the left shows the number of
new adopters at each point in time. You can
see that a higher q makes the adoption peak
earlier.
o The graph on the right shows the cumulative
number of adopters. With a higher q, the
product spreads faster, and more people
adopt it in the beginning, but both curves
eventually flatten when the market is
saturated.
6. In Summary:
• The Bass Diffusion Model is a way to describe
how new products spread through a population.
• People adopt the product either because they are
innovative (factor p) or because they see others
using it (factor q).
• The model shows us that adoption is slow at first,
speeds up as more people join in, and eventually
slows down again when most of the potential
market has adopted the product.