MGMT 8815M Week 11 Design Argument Example

Show the Applicant What the Underwriter Needs: A Design Argument for a Regional Bank's Small Business Lending, Defending a Visibility Frame Against a Speed Frame and an Underwriter-Kept Checklist Against an Approval Engine and a Loan Coordinator, With a Design Proposition and the Result That Would Count Against It

Student Name

Walden University

MGMT 8815M: Managing by Design Theory

Instructor Name

Month Day, Year

Show the Applicant What the Underwriter Needs: A Design Argument for a Regional Bank's Small Business Lending, Defending a Visibility Frame Against a Speed Frame and an Underwriter-Kept Checklist Against an Approval Engine and a Loan Coordinator, With a Design Proposition and the Result That Would Count Against It

Introduction

This paper makes two claims about small business lending at Meridell Bank, a composite regional bank with $3.8 billion in assets and 42 branches across western Pennsylvania. The first is that the problem with its small business lending is best framed as applicants being unable to see what underwriters need, not as a slow approval process. The second is that a shared document checklist, kept up to date by the underwriters themselves and visible to the applicant and the loan officer, answers that frame better than the faster automated approval engine the bank is now evaluating. The argument draws on the problem framing, field observation and organizational analysis carried out over this term, and it closes with a design proposition and a test that could show it wrong.

Meridell makes about 1,100 small business loans a year, most between $50,000 and $500,000. The median time from application to decision was 34 days in 2025, against 9 days at the two largest competitors in its market. The bank lost an estimated 140 approved or likely applicants last year to faster lenders, according to its own exit calls.

The Frame and Its Strongest Rival

The strongest rival frame is the one the bank's leadership holds: decisions are slow because underwriting is slow. On this view, each application passes through credit analysis that takes too long, and an approval engine that scores most applications automatically would cut the time. The frame has real support. Underwriters at Meridell carry about 45 open files each, and credit analysis on a full file takes two to three working days.

The visibility frame rests on what happened to the files before analysis began. A review of 120 files from the fourth quarter of 2025, carried out with the credit department during the term's field research, found that analysis took a median of 3 days of the 34. A median of 19 days elapsed while files waited for documents the underwriter requested after the application was submitted. The median applicant submitted documents in 2.6 rounds. The most common late requests were interim financial statements, schedules of business debt and tax returns for any owner with 20 percent or more of the business. Each of these appears on a list underwriters keep for themselves but that applicants never see; the application form asks for "financial statements" and "tax returns" without saying which years, which owners or in what form.

The visibility frame explains three things the speed frame cannot. It explains why analysis time is small compared with the total. It explains why repeat borrowers, who learned the list from their first loan, received decisions in a median of 12 days. And it explains why loan officers, when interviewed, described their main job as "chasing paper." An approval engine would cut the three days of analysis. It would leave the nineteen days of waiting untouched, because an engine cannot score a file that is incomplete.

Dorst (2011) located much of design's work in finding a frame from which a different kind of response can be built. The speed frame leads to a technology purchase; the visibility frame leads to a change in what the applicant is told and when.

The Design and Two Competing Designs

The proposed design is a shared checklist for each application, generated at the first meeting from the loan type, the business structure and the ownership, and maintained by underwriters. When an underwriter decides that a particular file needs an additional document, the underwriter adds it to that file's checklist, which the applicant and loan officer can see in the bank's online portal, with a short plain-language reason. Underwriters also review the standard lists each quarter, adding items they find themselves requesting late more than a set share of the time.

The first competing design is the approval engine. Its strength is speed for small, simple loans; vendors claim decisions in minutes for loans under $100,000 with clean credit. Meridell's own data suggest that about a quarter of its applications might qualify. For those applicants the engine would be faster than any checklist. For the other three quarters, the engine changes nothing until the file is complete.

The second competing design is a loan coordinator, a new role in each region who assembles files on the applicant's behalf, much as a mortgage processor does. Its strength is that it removes the burden from the applicant entirely. Its weakness is cost and dependence: four coordinators would cost about $340,000 a year, and the underwriters' knowledge of what they need would still sit in their heads, now relayed through a second person instead of a loan officer.

The checklist is chosen because it is the only design that moves the underwriters' knowledge to the applicant at the start. It costs less than the coordinators and does not prevent the bank from adding an engine later for its simplest loans. Romme (2003) argued that design as a mode of research differs from science and the humanities in aiming to produce systems that do not yet exist; the comparison here is offered in that spirit, as a judgment among workable systems rather than a proof.

Fit With the Organization

The term's organizational analysis found two mismatches that the design must overcome. Underwriters are measured on files decided per month, and adding items to checklists takes time that the measure does not count. And the online portal, built for consumer deposit accounts, can show documents uploaded but not a list of documents still required. The first can be addressed by adding a measure of rounds per file to underwriters' reviews; the second requires a change to the portal, which the bank's technology team has estimated at eight weeks of work. Berger and Udell (2002) argued that relationship lending depends on soft information gathered by loan officers that is hard to pass up through a bank's hierarchy. The checklist does not replace that information. It passes the underwriters' hard requirements down to the applicant, which frees the loan officer's time for the soft information that only a relationship can supply.

The Design Proposition

Denyer et al. (2008) proposed that design knowledge be stated as propositions with four parts: the context in which the design applies, the intervention, the mechanism through which it works and the outcome expected. Stated that way, the proposition is as follows. In a regional bank's small business lending, where underwriting requirements vary by loan and owner and are known mainly to underwriters (context), a document checklist generated at application and kept current by underwriters, visible to the applicant (intervention), lets applicants supply complete files in fewer rounds because they learn the requirements before they begin gathering documents (mechanism), so that the time from application to decision falls, mainly through less waiting rather than faster analysis (outcome).

Limitations

What the literature supports and what this paper proposes should be kept apart. The literature supports the general claim that problem framing shapes the solutions considered and the argument that design propositions can be stated and tested across settings (van Aken, 2004). It does not show that checklists shorten lending times; that claim is this paper's own, drawn from one bank's files. The 120-file review covers one quarter, and the fourth quarter may differ from others, since many applicants are waiting on year-end statements. The finding that repeat borrowers are faster may also reflect their stronger credit rather than their knowledge of the list.

Evaluation

The design should run for six months in two of the bank's six regions, with the other four as comparison. The measures are rounds of document requests per file, days waiting for documents and total days to decision. A result would count against the design if the pilot regions' rounds per file fell but their days to decision did not, since that would show the waiting has another cause. A second result would count against the mechanism specifically: if applicants in the pilot regions report not having looked at the checklist, any improvement must come from something else, such as underwriters reviewing their lists more carefully. Either result would send the inquiry back to the frame.

References

Berger, A. N., & Udell, G. F. (2002). Small business credit availability and relationship lending: The importance of bank organisational structure. The Economic Journal, 112(477), F32-F53. https://doi.org/10.1111/1468-0297.00682

Denyer, D., Tranfield, D., & van Aken, J. E. (2008). Developing design propositions through research synthesis. Organization Studies, 29(3), 393-413. https://doi.org/10.1177/0170840607088020

Dorst, K. (2011). The core of 'design thinking' and its application. Design Studies, 32(6), 521-532. https://doi.org/10.1016/j.destud.2011.07.006

Romme, A. G. L. (2003). Making a difference: Organization as design. Organization Science, 14(5), 558-573. https://doi.org/10.1287/orsc.14.5.558.16769

van Aken, J. E. (2004). Management research based on the paradigm of the design sciences: The quest for field-tested and grounded technological rules. Journal of Management Studies, 41(2), 219-246. https://doi.org/10.1111/j.1467-6486.2004.00430.x