Multi-Device Consumption of Digital Goods: Optimal Product Line Design with Bundling

ABSTRACT

A contemporary business challenge for bundling theory is the distribution of digital content such as media, entertainment, software, and other information goods. Consumers can use a number of devices to interact with software, online services, music, video, news, and other forms of digital content. For instance, Netflix videos and Kindle ebooks, initially accessed only on television sets and computers, respectively, are now also consumed on smartphones and tablets. Print products are now distributed and consumed both in print and digitally. Firms offer product line designs that include prices for single device access as also bundle discounts for multi-device access, so that consumers can choose a device or even multiple devices. This paper provides guidelines regarding such multi-device product line design and pricing. A starting point is to note that there are many ways to practice bundling, from offering prices for single-device consumption as well as a bundle discount (mixed bundling), to forcing multi-device prices (pure bundling), and designs in between (partial bundling). Picking the best design, and associated prices, is a complex problem due to how single-device demand functions relate to multi-device demand. Recognizing that some dual-device purchases can occur even when there is no bundle discount, the paper develops intuition around the net gain or loss incurred from giving a bundle discount to entice single-device access buyers to dual sales. One surprising finding is that inducing dual sales through bundle discounts can be profitable even when intent to consume multiple times is quite weak, because in this case dual sales would not occur organically. Less surprisingly, such inducement is also profitable when multi-consumption intent is strong, and least attractive when the intent is moderate. When one of the devices is an emerging one or has weak own demand in the short term, then it can be useful to offer a partial bundle, tying sales for the stronger device into a bundle comprising access to both devices. When device valuations are such that consumers generally agree on the rank-ordering of the devices (e.g., if a location-based app offers greater value on a smartphone than on a tablet), then it is best to employ some type of bundling, unless the intent for multi-device consumption is proportionally lower among low-value consumers than for high-value consumers.

KEYWORDS:

• Digital bundling
• bundling strategy
• digital goods
• economics of IS
• management science
• multi-device consumption
• product line design
• subadditivity
• multi-device lines

Acknowledgments

I appreciate feedback from D.J. Wu and De Liu on earlier drafts. I am grateful to Rob Kauffman and Atanu Lahiri for reviewing input, Shyla Chopra for copy-editing and support in generating the final version of the paper, and Kayla Ho for assistance with the graphics.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

¹ I note that while the product being sold is access to content via a certain device, it is convenient to write as if it is the device that is being sold.

² The fundamental construct for modeling consumer demand in Bakos and Brynjolfsson [3] is that a consumer ω’s valuation for good i, if she buys n goods, is vn_i(ω). Thus the valuation is dependent only on the number—rather than the specific subset—of goods purchased. A complementarity or substitutability factor α is introduced (α is independent of specific goods) to allow for nonadditive valuations: vn_i(ω) = n^αv₁₁. As is evident from setting n to 1, this implies that a consumer has identical valuation v₁₁ for every good when that good is purchased alone (i.e., marginal demands are identical).

³ If the consumer does not already possess the device and must incur a fixed cost for acquiring it (e.g., buying a smartphone or a gaming console), then vi is considered to be net of the amortized or per-period fixed cost. See Li [25] for a good joint treatment of device and content purchase decisions.

⁴ Despite this ordering on the aggregate demand profile, a consumer may still have a higher value for access via device 1.

⁵ If the two devices have synergistic features, some consumers might view them as complements leading to super- additivity (v_B > v₁ + v₂). Analysis of super-additivity runs parallel to analysis of sub-additivity, the important consideration being to go beyond the “simple” case of additive valuations. Therefore, I restrict the discussion and analysis in this paper to the substitutes case in order to maximize clarity and focus and to avoid unnecessary tedium.

⁶ This explanation is based on how bundle affects demand (and consequent revenues) rather than due to economies in transaction or distribution costs, which are eliminated in the setup.

⁷ The exception here is when η(x) increases at a faster rate than v₂(x) for some customers near X_B: then, a discount could entice the cluster near x < X_B into dual purchase, without extending that same incentive (and hence and the revenue loss) to other customers.

Additional information

Notes on contributors

Hemant K. Bhargava

Heman Bhargava ([email protected]; corresponding author) is Jerome and Elsie Suran Chair in Technology Management and Director of Center for Analytics and Technology in Society at University of California, Davis. He earned his Ph.D. in Information Systems, Operations, and Economics from the Wharton School of the University of Pennsylvania. Dr. Bhargava is an academic leader in economic modeling and analysis of technology-based business and markets. His research focuses on decision analytics and how the distinctive characteristics of technology goods influences specific elements of operations, marketing, and competitive strategy, and the implications it holds for competitive markets and technology-related policy. He has published extensively in such journals as Management Science, Operations Research, Marketing Science, Journal of Marketing Research, Information Systems Research, and others. He is a Distinguished Fellow of the INFORMS Information Systems Society, and Department Editor (Information Systems) for INFORMS’ flagship journal Management Science. Dr. Bhargava co-founded the annual Theory in Economics of Information Systems Workshop and the UC Davis Master of Science in Business Analytics program. He was listed among the Global 100 Top Academic Data Leaders by Chief Data Officer magazine in 2020.