DOC 715 Week 7 Feasibility, Scope and Boundaries Example

Can It Be Done, and Where Does It Stop? Feasibility and Scope for an Onboarding Dissertation

[Student Name]

University of Phoenix

DOC/715: Doctoral Seminar I

Week 7 Assignment

[Instructor Name]

[Date]

The learner, the company, its customers and all figures are composites written for a model paper.

Weeks 1 to 6 built the study piece by piece: an evidenced problem, aligned questions, a defended methodology and design and a sample with secured access. This paper asks whether she can finish it and where its boundaries lie.

Timeline

Locke et al. (2014) caution that proposals routinely underestimate the time research takes and advise building schedules from specific tasks with allowances for delay. The learner's timeline, counted from proposal defense:

Months 1 to 2: ethics board review; final data use agreement.

Month 3: data extraction and de-identification by the data team.

Months 4 to 5: data cleaning, propensity model, matching and balance checks, reviewed by the committee before outcomes are released.

Months 6 to 8: outcome analyses for RQ1 to RQ3, sensitivity and robustness checks.

Months 9 to 12: writing results and discussion chapters.

Months 13 to 15: committee review and revisions.

Months 16 to 18: buffer and final defense.

Three months of buffer are built in, about one sixth of the total.

Locking the Design Before Outcomes

Rubin (2007) argued that observational studies for causal effects should be designed as randomized experiments are, with the comparison groups constructed and their balance assessed without access to outcome data, so that the design cannot be tuned to produce a desired answer. The timeline follows that advice: outcome fields will be withheld by the data team until the committee approves the matched sample in month 5. This adds a step but protects the study's credibility.

Skills Audit

The learner is strong in data management and regression, having built retention dashboards for years, and completed doctoral statistics courses. She has limited experience with propensity score matching software and sensitivity analysis. Plans: a six-week online course in causal inference methods before month 4, practice on a public data set and guidance from a committee member who uses these methods. Writing is a strength; time for writing is the constraint, addressed by scheduling two writing mornings a week.

Resources

Data team time: about 40 hours for extraction and de-identification, approved in writing by the head of data.

Software: R and its matching packages, free; the company's secure analysis environment.

Committee: chair with expertise in service management, methodologist with causal inference experience and a third member in marketing.

Employer support: four hours a week of protected time, approved by the chief operating officer.

Risks and Mitigation

Customer relationship system migration planned for next year could make historical fields harder to extract: likelihood medium, impact high. Mitigation: extraction moved to month 3, before the migration, and a snapshot archived.

Rumored acquisition of the company: likelihood low to medium, impact high. Mitigation: permission letter includes a clause continuing access through ownership changes; de-identified data are already extracted early.

Limited power for RQ3: likelihood high, impact medium. Mitigation: stated in advance, with results interpreted cautiously.

Learner's job change: likelihood low, impact medium. Mitigation: data extracted and stored under the agreement; analysis can continue if she leaves.

The greatest risk to the study was not statistical but a software migration scheduled by another department.

Why Eighteen Months Is Realistic

Because the data already exist, the study avoids the slowest part of many dissertations: recruiting participants and collecting data. The longest phases are analysis, where the matching checkpoint adds a deliberate pause, and writing, where committee review cycles often take longer than expected. The learner checked the timeline against the experience of two recent graduates of her program who used existing organizational data; both finished their final phases in fifteen to twenty months.

If the Results Are Null or Unwelcome

Feasibility includes being able to finish whatever the results show. If proactive onboarding shows no association with retention, the study still answers its questions and fills the gap identified in Week 1; a well-designed null result is useful to firms deciding where to spend. If the results are unwelcome to company leaders, the permission letter already guarantees that the company may review for confidential information but may not alter findings. The learner discussed this possibility with the chief operating officer before the letter was signed.

Sustaining the Learner

The learner works full time and has two children. The plan protects four work hours and two early mornings a week and avoids the company's busiest renewal months for writing deadlines. She has arranged monthly check-ins with her chair and a writing partner from her cohort, since isolation is a common reason doctoral students stall.

Delimitations

The learner chose these boundaries: one company; service firms under 20 staff; subscriptions starting between 2023 and mid-2025; onboarding counted only when a manager met the customer in month one; renewal at a year and feature use at three months as the outcomes; and existing records only. Each keeps the study focused on the problem and achievable within eighteen months.

Limitations

Constraints outside her control: assignment to onboarding was not random, so unmeasured differences may remain; findings come from one firm's product, pricing and customers; records cannot capture customers' reasons or experiences; feature logs measure use but not value; and the size subgroup limits power for RQ3.

Do the Boundaries Still Answer the Questions?

Narrowing can go too far, leaving a study that is feasible but trivial. The learner checked each delimitation against the research questions. Studying one company limits generalization but still answers whether onboarding is associated with retention for that company's small-business customers, a question the literature has not answered for any firm. Restricting to service businesses keeps the population consistent with the problem. Twelve months captures the period when most churn occurs. Using records only excludes reasons for leaving, which no research question asks about. The boundaries narrow the study without hollowing it out.

The Committee's Role

Feasibility also depends on the committee. The learner shared this assessment with her chair, who asked that the matching checkpoint in month 5 be a formal meeting, not an email review, and that the methodologist attend. That request is now in the timeline. Committee availability during the summer, when two members travel, is the one scheduling risk the learner cannot control; month 6 to 8 analyses were placed to avoid needing committee input in July.

Trimming Scope

Two additions were tempting. A committee member suggested adding a customer satisfaction survey, and the learner wanted to model revenue expansion. Each would add months of new data collection or analysis to an already full plan. Hochstein et al. (2020) describe expansion revenue as a key goal of customer success, which makes it a worthy topic, but not for this study. Both are moved to future research.

Conclusion

With an eighteen-month timeline including a design-before-outcomes checkpoint and three months of buffer, an honest skills plan, resources secured in writing and risks mitigated, the study is feasible. Six delimitations define its scope, five limitations are acknowledged and two additions are deferred. Week 8 assembles the Phase 2 foundation.

References

Hochstein, B., Rangarajan, D., Mehta, N., & Kocher, D. (2020). An industry/academic perspective on customer success management. Journal of Service Research, 23(1), 3-7. https://doi.org/10.1177/1094670519896422

Locke, L. F., Spirduso, W. W., & Silverman, S. J. (2014). Proposals that work: A guide for planning dissertations and grant proposals (6th ed.). Sage.

Rubin, D. B. (2007). The design versus the analysis of observational studies for causal effects: Parallels with the design of randomized trials. Statistics in Medicine, 26(1), 20-36. https://doi.org/10.1002/sim.2739