How to include "theory" in contemporary design science research?

Top IS journals emphasize “theory.” But there is an inherent mismatch between the design science research method and canonical approaches to “theory” in management research (Walls et al., 1992).

Standard approaches to theory building focus on describing empirical relationships between constructs (Whetten, 1989). But in DSR, researchers focus on helping achieve goals rather than describing the world (Hevner et al., 2004). Given this core discontinuity, how should DSR authors fulfill the strong preference for “theory” at MISQ and ISR?

To investigate, I reviewed all self-described design science papers published in the two journals since 2023. This revealed that DSR authors adopt five broad strategies for incorporating theory in their work.¹

Strategies

1. Just don’t have theory

A surprising number of DSR authors did not include “theory” in their papers, or at least did not describe their contributions using this terminology.² This strategy was common at ISR, where I counted 8 recent DSR papers that did not mention the word “theory” at all. Examples include Zhang et al. (2024), Guo et al. (2024), and Xie et al. (2025). That said, I did not find any MISQ papers that did not at least mention some connection with theory.

2. Briefly mention theory

A handful of papers briefly mentioned the importance of theory in design science research, without explicitly laying out a formal theoretical perspective. Examples include Lin et al. (2024) at MISQ and Bossler et al. (2024) at ISR. For example, Ampel et al. (2024) briefly mention that a novel artifact can contribute to knowledge by informing a nascent design theory.

3. Offer “theory” as background

A few papers used the word “theory” to describe something that is almost like a literature review. For example, Kim et al. (2023) use the heading “Theory and Motivation” in a section that enumerates desirable traits for explainable AI, provides a 2x2 taxonomy of explainable AI systems, and then digs into the limitations of existing approaches within a specific quadrant of the matrix.³ Faik et al. (2024) also follow this broad pattern.

This use of theory as background is a little different from canonical definitions of “theory” like Whetten (1989) which make much sharper claims about the relationships between constructs (e.g., CEO personality → sales). In a traditional descriptive paper (rather than DSR paper) this kind of background is sometimes used to reason deductively about hypothesized relationships between constructs (see Wilson et al., 2024 in MISQ).

4. Use a kernel theory

Another common strategy was to use a “kernel theory” to guide the development of an artifact. A kernel theory is a description of reality which informs prescriptive principles about how to build an artifact.

This strategy is probably the most well developed approach in the design science literature. It seems to come from Walls et al. (1992), who formally articulated some of differences between classical business “theory” (Dubin, 1978) and “theory” in a goal-oriented DSR context.

Li et al. (2024), Wambsganss et al.(2024) and Liu et al. (2025) each use a kernel theory. For example, Li et al. (2024) use a theory of how people learn (“cognitive load theory”) to inform the development of educational software.

Note: There is a 2022 ICIS paper from Möller et al. on the application of kernel theories in design science research. Check there for more.

5. Develop a full design theory

Danatzis et al. (2024) propose a full “design theory” by applying a taxonomy from Gregor & Jones (2007). From this perspective, “theory” is a formally-defined, complex structure with eight different sub-components such as constructs and testable propositions.

Note that in Gregor & Jones the “kernel theories” from Walls et al. (1992) are called “justificatory knowledge.” So in a way strategy #5 contains strategy #4.

Phew. OK so that lays out the five strategies that DSR researchers use to apply “theory” in their work. I think there are two key takeaways.

Takeaways

Phew. OK so that lays out the five strategies that DSR researchers use to apply “theory” in their work. I think there are two key takeaways.

• First, based on this analysis, the word “theory” is a little bit overloaded in contemporary DSR. Theory can mean background knowledge, use of a kernel theory (Walls et al., 1992), or a full-fledged design theory Gregor & Jones, 2007). Because the meaning of “theory” is a little ambiguous within DSR, I think it makes sense for researchers to clearly communicate how they are applying “theory” in their own papers, and how this choice aligns with prior work.

• Second, based on this analysis, “theory” does seem to be optional in design science research, especially at ISR. In the past, I have abandoned projects because I could not see a path towards a theoretical contribution. But in some cases, it may be OK to submit a DSR paper without theory.

Footnotes