MKT 443 Topic 2 customer data quality audit example

Customer Relationship Management Grand Canyon University Free custom sample in 24 to 48h

This MKT 443 Topic 2 sample paper audits a composite musical instrument retailer's customer file before any campaign is planned. It belongs to GCU MKT 443, Customer Relationship Management, listed as MKT-443 in the Grand Canyon University BS in Marketing. A random sample of 2,000 from about 48,000 records finds 18 percent without email, 9 percent of addresses failing validation, about one in nine duplicated and 31 percent inactive for three years. Each defect is traced to where it enters, from an email prompt that appears after payment at the till to lesson bookings that never search for an existing customer.

CourseMKT 443 Customer Relationship Management
TopicTopic 2
Paper typeCustomer data quality audit
LengthAbout 960 words, 2 pages
FormatAPA 7 student paper
SchoolGrand Canyon University (GCU)
ProgramBS in Marketing
UpdatedOctober 2026

Free sample paper for MKT 443 Topic 2

1

Missing, Wrong, Repeated or Stale: A Customer Data Quality Audit That Traces Each Defect to Where It Enters

[Author Name]

Colangelo College of Business, Grand Canyon University

MKT-443: Customer Relationship Management

Topic 2 Data Quality Audit

[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.

What this page is doingThe title names the four defects the audit measures and its distinctive step. The retailer, its file and every figure are a composite; the sources are real.
2

Missing, Wrong, Repeated or Stale: A Customer Data Quality Audit That Traces Each Defect to Where It Enters

Why Audit First

A composite musical instrument retailer with three stores, a lessons program and an online shop wants to segment its customers and launch targeted campaigns. Before planning any of that, this audit measures the quality of the customer file the campaigns would draw from. Segmentation and dashboards inherit whatever defects the file carries: a segment built on missing emails cannot be emailed, and a count of customers inflated by duplicates overstates the market. Bad customer data quietly raises costs everywhere it travels (Redman, 1998), reaching from day-to-day operations to strategic decisions to the trust staff place in their own systems, and few firms ever measure those costs. Wang and Strong (1996) showed that data users judge quality on several dimensions, including completeness, accuracy, consistency and timeliness, not accuracy alone. This audit measures four defects that map to those dimensions. All figures are illustrative.

The Sample

The file holds about 48,000 customer records. The audit drew a random sample of 2,000 and examined each, checking fields automatically where possible and reviewing duplicates by hand. With a sample of this size, each reported percentage is precise to within about 1 to 2 percentage points.

What the Sample Shows

Eighteen percent of sampled records have no email address, which projects to about 8,600 customers who cannot be reached by the channel the retailer plans to use most. Nine percent of postal addresses fail validation against a standard national address file, so mail to about 4,300 customers would be returned or lost. Manual review found that about one record in nine, about 11 percent, duplicates a customer already in the file. And 31 percent of records show no purchase, lesson or contact in three years.

Table 1

Customer File Defects in a Sample of 2,000 Records, Illustrative

DefectQuality dimensionShare of sampleProjected records of 48,000
No email addressCompleteness18%About 8,600
Postal address fails validationAccuracy9%About 4,300
Duplicate of an existing customerConsistencyAbout 11%About 5,300
No activity in three yearsTimeliness31%About 14,900

Where Each Defect Enters

A cleansing report would stop at the counts and recommend fixing records. This audit goes further and traces each defect to where it enters the file, because cleaning records without changing how they are created means cleaning them again next year.

Missing emails enter at the store till. Observation at two stores showed that the point-of-sale screen asks for an email address only after the card payment is complete, when the customer is already putting away a wallet and leaving. Clerks skip the prompt rather than call the customer back. The field is not lost later; it is never captured.

Bad addresses enter through a free-text address field at the stores, with no lookup or validation, so typing errors, missing apartment numbers and abbreviations go straight into the file.

Duplicates enter through the lessons program. A parent who bought an instrument in a store is registered there, then registers again when booking a child's lessons online, because the booking system does not search for an existing customer. Most duplicates in the sample follow exactly this pattern, with the two records differing only in email or phone format.

Stale records accumulate because the file has no rule for marking customers inactive. Customers who moved away, stopped playing or bought once a decade ago remain in the file as if current.

How the Defects Interact

The four defects are not independent. A duplicate created through the lessons booking often carries the email address that the store record lacks, so merging duplicates will also reduce missing emails. Stale records include many with failed addresses, because people who moved away years ago are the ones whose addresses no longer validate. Fixing the defects in the right order therefore saves work: merge duplicates first, then mark inactive records, and only then validate the addresses that remain, since many failed addresses will already have been set aside. The counts in Table 1 overlap for the same reason and should not be added together.

What the Defects Would Do to Planned Campaigns

If the retailer segmented the file as it stands, about 31 percent of every segment would be people who have not bought in three years, and campaign response rates would look worse than they are. Duplicates would inflate customer counts and send families two copies of each offer. And the email campaigns would miss nearly a fifth of customers entirely, skewed toward store buyers rather than online buyers, which would distort any comparison between the two channels.

Recommendations at the Source

Each recommendation addresses where a defect enters. Move the email prompt to before payment at the till, and give clerks a reason to ask, such as offering the receipt by email. Replace the free-text address field with an address lookup. Make the online lesson booking search for an existing customer by phone or email before creating a new record, and merge the existing duplicates in a one-time review. Adopt a rule that marks customers inactive after three years without activity and excludes them from campaigns unless they respond to a reactivation offer. Payne and Frow (2005) placed information management at the center of CRM strategy, and these changes are a small, practical part of it.

How the Fixes Will Be Checked

The audit will be repeated on a fresh random sample of 2,000 records six months after the changes. Because the defects are now tied to their sources, the repeat audit can check each separately: the share of new store records with an email, the share of new addresses that validate, and the share of new lesson registrations that duplicate an existing customer. If a defect has not fallen, the audit will show which entry point is still producing it.

References

Payne, A., & Frow, P. (2005). A strategic framework for customer relationship management. Journal of Marketing, 69(4), 167-176. https://doi.org/10.1509/jmkg.2005.69.4.167

Redman, T. C. (1998). The impact of poor data quality on the typical enterprise. Communications of the ACM, 41(2), 79-82. https://doi.org/10.1145/269012.269025

Wang, R. Y., & Strong, D. M. (1996). Beyond accuracy: What data quality means to data consumers. Journal of Management Information Systems, 12(4), 5-33. https://doi.org/10.1080/07421222.1996.11518099

What the MKT 443 Topic 2 instructions ask for

The Topic 2 assignment in MKT-443 asks students to assess the quality of customer data. Strong audits measure defects on a random sample with stated precision, classify them by quality dimensions such as completeness, accuracy, consistency and timeliness, and then trace each defect to the point where it enters the file. Students should explain how the defects interact, show what they would do to planned segmentation and campaigns, recommend fixes at the source rather than one-time cleansing, and describe how a repeat audit will check each fix. The audit should also say which defects overlap, so fixes are made in a sensible order.

How the MKT-443 Topic 2 example is put together

The finished audit works from a random sample of 2,000 records drawn from roughly 48,000, and every figure it reports is labeled illustrative. Eighteen percent of sampled records carry no email address. Nine percent of postal addresses fail validation against a standard address file. Manual review finds that about one record in nine duplicates a customer already on file, usually a parent registered once at a store and again when booking a child's lessons online. Thirty-one percent show no purchase, lesson or contact in three years. The audit then does what a cleansing report would not: it traces each defect to where it enters. Store clerks reach the email prompt after the card has been taken and the customer is leaving, so the field is skipped at the till, not lost later.

Reading the MKT 443 Topic 2 rubric

Graders award an Excellent audit for tracing defects to their source. The rubric rewards a random sample of 2,000 with precision stated and a table of four defects mapped to Wang and Strong's dimensions. Credit goes to finding that emails are skipped because the till prompt comes after payment and that duplicates come from online lesson booking. Faculty look for Redman on the costs of poor data. Strong audits explain how defects overlap and the order of fixing them. Top marks go to source fixes checked by a repeat audit.

Common MKT 443 Topic 2 mistakes, and how to avoid them

Nothing costs more here than rates drawn from a handful of records someone opened by hand, since no figure from them can be applied to the file. Reporting that the data is poor without a single measured percentage leaves every later topic without a baseline. Many drafts list the quality dimensions from the reading and never score the sample on any of them. A cleanup recommendation with no account of where defects enter repeats the problem within months, because the clerks and web forms producing it are unchanged. Treating duplicates as a storage issue misses that each one splits a customer's history and sends two renewal letters to one household. The least expected deduction comes from blaming staff for skipped fields when the screen asked at the wrong moment.

Get an MKT-443 Topic 2 example written to your instructions

Send the MKT-443 Topic 2 instructions and the rubric posted in your classroom, with the customer data or case your section supplied. We write a custom example to those criteria, with a sampled file scored on each quality dimension, every defect traced to its point of entry and fixes ranked above cleanup, in 24 to 48 hours. The first one is free.

More MKT 443 papers and BS in Marketing samples

MKT 443 Topic 2 questions, answered

What are the dimensions of data quality?

The audit uses five that most data management texts recognize under similar names: completeness, whether a field is filled; validity, whether the value fits its format and rules; uniqueness, whether a customer appears once; timeliness, whether the record is current enough to act on; and consistency, whether the same fact agrees across systems. Other lists add accuracy against the real world, which usually needs a costlier check.

Why sample the file instead of checking every record?

Because a well-drawn random sample gives a reliable estimate of each defect rate at a fraction of the effort, and some checks, such as judging whether two records are the same person, need a human eye. Two thousand records reviewed carefully tell the retailer more than forty-eight thousand scanned by a rule that misses the hard cases. The example states its sampling method so the rates can be trusted.

Why fix data at the point of entry rather than clean it later?

Because cleaning treats the result while the cause keeps running. If the till still asks for an email after payment, next year's file will be missing addresses at the same rate. A cleanup is sometimes worth doing once, and the example budgets for one, but it ranks changes to the screens and routines that create each defect ahead of it.

Where can I find a free MKT 443 Topic 2 sample paper?

The data quality audit appears here complete: why audit first, the sample, what it shows, where each defect enters, how the defects interact, what they would do to campaigns, recommendations at the source and how the fixes will be checked. Share your file's sample and a free audit can be built.

Where do the missing emails come from in MKT 443 Topic 2?

From the store till. The point-of-sale screen asks for email only after card payment, when the customer is leaving, so clerks skip it. The address is never captured, not lost later.