Marc Andreessen defined platform as “a system that can be programmed and therefore customized by outside developers — users — and in that way, adapted to countless needs and niches that the platform’s original developers could not have possibly contemplated, much less had time to accommodate.”
Clearly platforms as described above are intermediaries that connect two or more distinct groups of users and enable their interaction and generate multiple revenue streams by charging both sides of the market.
Reading about Uber bundling its Uber Eats offering in its main app and increasing its Total Addressable Market (TAM). Or even Ola bundling Foodpanda made me learn more about bundling. Even, Microsoft has long followed this strategy first bundling explorer with windows and then bundling all its productivity software in its Office365 offering, though some of its businesses are not platforms.
This practice of bundling products is called a Platform Envelopment Strategy.
The strategy proponents: When network effects are strong and positive, users will converge on fewer platforms; a new platform will have little appeal. The number of players serving a market will be small and users will face high costs when multi-homing. With few firms, platform providers enjoy market power.
Though high margins would attract other firms, incumbent platform providers are often well protected. Factors that restrict the number of platforms in the first place makes it difficult and expensive for other firms to develop a new platform.
However, bundling can offer an entry path to new firms.
Platform Envelopment involves bundling the functionality of the incumbent’s platform with the functionality of the target’s platform to take advantage of the overlapping user base. For example, Microsoft Windows and Internet Explorer bundle.
The strategy is effective because it forecloses the target platform’s access to the shared user base. Since platforms engender Economies of Scale, foreclosing the access increases the profit and the cost of the incumbent and target firm respectively. In the above example it foreclosed Netscape’s access to user base.
The Leverage in an envelopment strategy can come from Two sides Revenue and Cost. From Revenue the entrant firm can have pricing power (when there is high overlap) and/or can have share gains by discounting (when there is low overlap, but the correlation of potential customer valuations for the two platforms is weak). From the cost side the entrant firm enjoys economies of scope because of the shared user base and/or components.
We can see that the Envelopment strategy depends on variables like User Overlap, Economies of Scope, Correlation in the Potential consumer valuations for the two platforms, and the relationship between the two platforms.
UserBase Overlap
There will be a High user base overlap when both the platforms are mature and marketed to mass-market consumers. For example Direct2Home TV and Telecom Companies have a high user base overlap; MS Word and MS Excel among knowledge workers and/or sometimes usage of the target platform requires the use of attacking platforms like in the case of windows and explorer bundle.
Economies of Scope
Bundling can harness economies of scope only when cost-sharing opportunities are available. These opportunities are mainly available in production and marketing or maybe in distribution costs.
Correlation and Heterogeneity
Heterogeneity
Single Platform Case
Let’s start with a single platform case. The market is populated by two consumers one with a high valuation of $100 and the other with a low valuation of $50.
Now the platform provider has two options either it can price the product at $50 or at $100 (assuming the provider cannot distinguish the consumers). In both cases the provider makes $100 and leaves the consumer surplus of $50. To extract consumer surplus, the firm would want two consumers with a valuation of $75.
Multi-Platform Case
Through bundling, platform providers can extract the consumer surplus. To see that let’s assume a firm wants to bundle two of its platforms A and B. The market is populated by two consumers. With valuations, the table would look like this
Bundling Platform A and B into the platform (AB’). The firm can price AB’ at $150 which is the sum of the optimal prices of the platforms when sold separately. At $150 price level the firm will earn $300 in revenue. Hence bundling can reduce customer heterogeneity.
Mathematically bundling reduces heterogeneity because the variance in the sum of product valuations is lower than the sum of variances in product valuations (Adams and Yellen [1976], and modeled in further detail [Schmalensee, 1984, McAfee et al., 1989).
Correlation
Correlation of valuations for the two platforms means the relationship between them. From the above example, it’s clear a user might attach one value to Platform A and a different value or a similar value to Platform B. Various other combinations that follows are:
Value of Platform A = V(A)
Value of Platform B= V(B)
Users in the upper right and lower left quadrant have a high correlation of valuations for both the platforms. And vice versa for users in the other quadrants.
Perfectly negative correlation or low positive correlation is better when bundling platforms. We can understand this from our previous example by changing some values. Suppose C1 values both the platforms at $100 and C2 values both the platforms at $50.
In this example users’ value for the two platforms is perfectly positively correlated. By charging $150 only consumer C1 will buy and the revenue would be only $150. Consumer C2 won’t buy as the price of the bundle AB’ exceeds the value for consumer C2.
Relationship
The relationship between two platforms can be of three types
Complement to each other
Weak Substitutes
Un-Related
An entrant that bundles complementary platforms is most likely to succeed when the user base overlaps significantly and there is a weak correlation of potential customer valuations for the two platforms.
An entrant that bundles weak substitutes is most likely to succeed when bundling offers strong Economies of Scope.
An entrant that bundles un-related platforms is most likely to succeed when there are significant economies of scope and platforms should not have a strong positive correlation.
Hence,in general, the Platform Envelopment strategy is most effective when following features are present
a) Platforms are complement to each other.
b) There is high userbase overlap.
c) Correlation in Potential Valuation of users for both the platforms is negative.
References
Platform Envelopment by Thomas R. Eisenmann, Geoffrey Parker, and Marshall W. Van Alstyne