Almost every market, from food to telecom, bundles products into package deals. But while intuitively sensible, does bundling add real value in practice?
Do you want fries with that? Do you want leather seats, seat warmers, a camera for your side-mirror? How about a wireless mouse with your laptop?
Companies bundle: restaurants offer value meals, airlines market getaways, and telecom providers pair cable with internet service. The practice has become ubiquitous. It is “arguably the most flexible element of product strategy, since the component products are already available,” write Vineet Kumar of Yale SOM and Timothy Derdenger of Carnegie Mellon in Marketing Science. But the effectiveness of bundling as a sales strategy is weakly understood.
In light of this, Kumar and Derdenger partnered—an academic bundle?—to undertake one of the first empirical analyses of the practice. They hoped to clarify three basic strategic questions. First, does bundling result in the cannibalization of component product sales? And, either way, does it increase or decrease overall revenue? Second, are so-called mixed bundles, in which products can also be purchased individually, more effective than pure bundles, in which products can only be purchased together? And, third, does product complementarity affect whether or not to bundle?
To get at these questions, Kumar and Derdenger examined the portable video game market between 2001 and 2005, during which Nintendo produced two very similar hand-held gaming consoles and a shared library of games. They consolidated data on the quantity of consoles sold and total revenue for each; they also looked at three bundles released during that period—packages that included Mario Kart, Mario Advance 2, or Mario Advance 4 with the purchase of a console. From this information, Kumar and Derdenger then built a consumer behavior model in which consumers, over time, chose whether or not to buy a console or a bundle, and whether or not to buy video games if they already owned a console.
Mixed bundling offers a unique and previously unexplored method for extracting more value from consumers over the product life cycle.
In short, they found that bundling works as a sales strategy, but the results are nuanced.
For instance, bundling did lead to the cannibalization of over half a million consoles—that is, the introduction of bundles led to a loss in sales of about 500,000 plain consoles. But the overall effect was an increase in hardware sales and revenue.
Bundling also had the unanticipated effect of goading some consumers into making a purchase earlier than they otherwise would have. In particular, consumers who don’t place a high value on the console or specific games, who might otherwise wait for the hardware to drop in price, jump at the opportunity for a bundled discount that provides both a cheaper console and earlier access to the software. This shift translates into a longer potential timeline for game purchases, and therefore more total game purchases.
“By pulling hardware sales forward, the firm is able to get these consumers to the software market earlier, leading them to purchase more software over time,” write Kumar and Derdenger. In this way, mixed bundling offers a unique and previously unexplored method for extracting more value from consumers over the product life cycle.
Investigation into their other two questions provided the authors with equally useful insights. In the case of the video game market, they found that mixed bundling “dominates” pure bundling as a mechanism for revenue; revenue losses under the pure bundling scenario were at least 35 percent compared to the mixed bundling scenario.
More generally, Kumar and Derdenger found that bundling for most companies—those in which products are not integral to each other—operates as a kind of costless R&D. “Bundles serve a role similar to an additional product in the firm’s product offering,” they note, “since consumers do not value the bundle identical to the sum of valuations of the component products.”
The takeaway? Given the right market, bundling “allows firms to easily create entire product lines where only one product existed.” For a vast spread of industries, say Kumar and Derdenger, the institutional settings support the intuition.