| Course | DBA 839 Enterprise Data Complexity |
|---|---|
| Topic | Topic 2 |
| Paper type | Post-acquisition integration analysis |
| Length | About 970 words, 2 pages |
| Format | APA 7 student paper |
| School | Grand Canyon University (GCU) |
| Program | DBA |
| Updated | October 2026 |
Free sample paper for DBA 839 Topic 2
Three Systems, Three Ideas of a Customer: A Post-Acquisition Integration Analysis for a Regional Bank, Recommending a Cross-Reference Hub
[Author Name]
College of Doctoral Studies, Grand Canyon University
DBA-839: Enterprise Data Complexity
Topic 2 Post-Acquisition Integration Analysis
[Instructor Name]
[Date]
The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.
Three Systems, Three Ideas of a Customer: A Post-Acquisition Integration Analysis for a Regional Bank, Recommending a Cross-Reference Hub
The Request and Its Two Uses
A composite regional bank has grown through three acquisitions in eight years and now runs three core banking systems, holding about 1.9 million customer records between them. The retail and commercial divisions have asked for one view of each customer. They want it for two purposes. Retail wants to cross-sell: to know that a mortgage customer from one system also holds a deposit account in another, so that offers are relevant and nobody is solicited for a product they already have. Commercial credit wants exposure aggregation: to know every obligation a borrower and its related parties owe the bank, so that lending limits hold across all three systems. The two uses do not want the same answer, and the analysis begins by saying why.
Three Systems by Their Assumptions
The systems differ less in technology than in what each assumes a customer is. The oldest, the bank's original core system, keys everything to the account. A person with a checking account, a savings account, a car loan and a credit card appears four times, linked only by a shared tax identification number when one was captured. The second, inherited with a community bank, groups accounts into households, treating people who share an address and surname as one relationship; it was built for branch bankers who knew families. The newest, inherited from a digital lender, identifies people by login and knows nothing of households at all; two spouses with separate logins are strangers to it, while one login used by a small business owner for both personal and business accounts is one customer.
Five Customers Seen Three Ways
Working through sample customers shows what the systems disagree about. None of the three pictures is an error. Each was right for the business that built it. The problem is that they cannot all be the bank's picture.
Table 1
Five Sample Customers as Each System Sees Them, Illustrative
| Sample customer | Original core | Community bank system | Digital lender system |
|---|---|---|---|
| Married couple, joint mortgage, separate cards | Three records | One household | Two people, not linked |
| Widowed parent and adult child at same address | Two records | One household | One login, the child's |
| Owner of a small bakery and the bakery | Two records under two tax numbers | Two households | One login for both |
| Customer who moved and reopened an account | Two records, old and new address | Two households | One person |
| Two unrelated people sharing an apartment | Two records | One household | Two people |
Every Matching Rule Decides Something
Linking records across the systems requires matching rules, and the theory of record linkage makes clear that any rule trades false matches against missed ones (Fellegi & Sunter, 1969). Tightening the rule to avoid linking two different people will leave some records of the same person unlinked; loosening it does the reverse. That trade-off is technical, but the rules built on top of it are business decisions, and each belongs to the executive whose risk or revenue it changes.
Whether spouses with a joint mortgage share an exposure limit belongs to the chief credit officer. Whether a small business owner and the business are one relationship, which affects both lending limits and which banker serves them, belongs jointly to the heads of commercial and retail banking. When records disagree on an address or phone number, which source survives in the combined view belongs to the operations executive responsible for customer contact. And whether two people at one address are offered products as a household belongs to the head of retail marketing, within limits the compliance officer sets. Khatri and Brown (2010) argued that data governance should assign decision rights across domains such as quality, metadata and access; here the domain is identity, and the assignment must be made before any integration is built.
Three Integration Approaches
Three approaches are available. Converting everything onto one system and one customer model would give the cleanest result but would force a single answer to every question above at once, take an estimated four years and carry the risk of a large conversion. Keeping the three systems and linking them through a cross-reference hub, which stores matches and survivorship rules without changing the systems, would be faster and reversible, but would leave three customer models in place. Rebuilding around legal identity, verified individuals and registered businesses, with accounts and households as relationships between them, would be the most flexible model for both uses but would require identity verification the original core never performed for older customers.
Batini et al. (2009) surveyed data quality methodologies and noted that integration projects tend to fail when quality and identity questions are deferred to the technical build. Each approach should therefore be judged by how it handles the decisions above, not only by cost.
Recommendation
The bank should build the cross-reference hub now. It serves exposure aggregation, which is the more urgent need because lending limits are a safety and regulatory matter, by linking records under tight rules set by the chief credit officer. It serves cross-selling through looser household rules owned by retail, kept separate so that marketing convenience never loosens credit limits. A first version linking commercial borrowers could be running within a year, and every rule it applies would be recorded with its owner and the date it was approved. The hub defers the single customer model, the retirement of any system and a firm-wide answer to what a household is. Each deferral is deliberate, because each requires a business decision that the bank's leaders have not yet made and should not make under the pressure of a technical project. The choice should be reopened if the bank decides to retire the original core system, which would make conversion to a model built on legal identity the natural next step.
References
Batini, C., Cappiello, C., Francalanci, C., & Maurino, A. (2009). Methodologies for data quality assessment and improvement. ACM Computing Surveys, 41(3), Article 16. https://doi.org/10.1145/1541880.1541883
Fellegi, I. P., & Sunter, A. B. (1969). A theory for record linkage. Journal of the American Statistical Association, 64(328), 1183-1210. https://doi.org/10.1080/01621459.1969.10501049
Khatri, V., & Brown, C. V. (2010). Designing data governance. Communications of the ACM, 53(1), 148-152. https://doi.org/10.1145/1629175.1629210
The DBA-839 Topic 2 assignment instructions
The Topic 2 assignment in DBA-839 asks doctoral students to analyze data integration after acquisitions. Strong analyses identify the business uses behind the request and explain why they may want different answers. Students should profile each inherited system by the idea of the customer it was built around, show with sample records where the systems disagree and list the matching and survivorship rules any integration must adopt. Each rule should be assigned to the executive whose risk or revenue it changes before integration approaches are compared, and the recommendation should state what it defers and what would reopen the choice.
Inside the DBA 839 Topic 2 example
The request that prompted the finished analysis comes first: the retail and commercial divisions want one view of each customer for cross-selling and for exposure reporting. It then describes the three inherited systems by their assumptions rather than their technology. The oldest keys everything to the account, so a person with four accounts appears four times. The second, bought with a community bank, groups accounts into households. The newest, from a digital lender, identifies people by login and knows nothing of households at all. Matching these is possible, but every matching rule decides something: whether spouses share an exposure limit, whether a small business owner and the business are one relationship. The analysis assigns each such rule to the executive whose risk or revenue it changes, and only then compares integration approaches.
Where the marks sit in the DBA 839 Topic 2 rubric
Faculty reward integration treated as a set of business decisions. An Excellent analysis separates cross-selling from exposure aggregation. Credit goes to profiling three systems by account, household and login, with a table of five sample customers seen three ways. The rubric values Fellegi and Sunter on matching trade-offs and assigning each rule to an owner, following Khatri and Brown. Strong papers weigh conversion, a hub and legal identity, citing Batini and colleagues on deferred quality questions. Top marks go to tight credit rules kept apart from looser marketing rules.
DBA 839 Topic 2 help before you submit
Where this analysis most often falls short is in comparing integration tools before anyone has said what a customer is. A paper that maps fields between systems and declares the integration designed has solved the format problem and left the identity problem to whoever writes the matching code. Drafts also lose credit by treating the oldest system as simply obsolete, when its account-level logic may be exactly what exposure reporting needs. Matching and survivorship rules presented as technical settings hide business decisions about credit limits and customer contact inside configuration. Recommendations for a single customer view that serve cross-selling alone ignore the division that must defend its aggregated exposure to examiners. A hub proposed without a record of what it postpones quietly commits the bank to a permanent interim, and a careful grader asks when the interim ends.
Get a DBA-839 Topic 2 example written to your instructions
Send the DBA-839 Topic 2 instructions and the rubric posted in your classroom, with the integration case or company your section uses. We write a custom example to them, with each system profiled by its assumptions, matching rules assigned to business owners and the chosen approach shown with what it defers, in 24 to 48 hours. The first one is free.
More DBA 839 papers and DBA samples
- DBA 839 Topic 1: Platform Postmortem Essay
- DBA 839 Topic 3: Definition Arbitration Memo
- DBA 839 Topic 4: Data Decision Rights Matrix
- DBA 839 Topic 5: Dashboard Disagreement Dq Post
- DBA 839 Topic 6: Architecture Commitment Analysis
- DBA 839 Topic 7: Unmade Decision Trace
- DBA 839 Topic 8: Data Strategy Defense Brief
- DBA 835 Topic 2: Materiality Assessment
- DBA 833 Topic 2: Data Preparation Decision Log
- DBA 831 Topic 2: Claim and Design Alignment Paper
- MGT 805 Topic 2: Grouping Basis Comparison
DBA 839 Topic 2 questions, answered
Why is a single customer view so hard after acquisitions?
Because each acquired system was built around a different idea of the customer, and those ideas carry business consequences. Merging account records into people, or people into households, requires rules about who counts as one relationship, and those rules change credit exposure, marketing consent and service entitlements. The example treats each rule as a decision for a named executive rather than a setting for the integration team.
What does master data management add here?
It supplies the discipline for maintaining a shared reference record, including how records are matched, which source wins when values conflict and who approves changes. The example applies the idea at enterprise level: a hub can hold the cross-reference, but the match and survivorship rules inside it still need business owners with authority to change them and accountability for what follows.
Is the recommended hub the right choice for every bank?
No. The bank is a composite, and the recommendation follows from its particular mix of systems, its reporting obligations and how soon it can afford to rebuild. A different institution with one dominant platform might sensibly convert everything to that model. Its value lies in the weighing of integration options that DBA-839 expects, and it offers no technology or regulatory advice to any real organization.
Where can I find a free DBA 839 Topic 2 sample paper?
The integration analysis is printed here in full: the request and its two uses, three systems by their assumptions, five customers seen three ways, matching rules and owners, three approaches and the recommendation. Your own DBA 839 Topic 2 prompt gets a free analysis.
Why keep credit and marketing matching rules apart in DBA 839 Topic 2?
Exposure limits need tight rules that avoid linking different people, while cross-selling tolerates looser household links. One shared rule would either weaken credit control or starve marketing.