| Course | MT362 Artificial Intelligence Applications for the Marketing Professional |
|---|---|
| Unit | Unit 5 |
| Paper type | Personalization strategy memo |
| Length | About 830 words, 6 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | BS in Business Administration |
| Updated | October 2026 |
Free sample paper for MT362 Unit 5
What the Customer Told Us, What We Noticed, What We Guessed: An MT362 Unit 5 Personalization Strategy Memo
[Student Name]
Purdue University Global
MT362: Artificial Intelligence Applications for the Marketing Professional
Unit 5 Personalization Strategy Memo
[Instructor Name]
[Date]
The chain, director and rules are composite; the research cited is real.
Memo
To: Renee Castillo, Director of Lifecycle Marketing, Fernbrook Pet From: Marketing analyst Re: Personalization rules for the retention pilot
Decision Requested
Approve the eight rules below before the pilot begins in February. The rules table is attached.
The Recognition Test
One test organizes every rule: would the customer, reading the message, understand how the chain came to know what it says? Messages that pass use data the customer gave or would expect a retailer to hold. Messages that fail reveal an inference, and that revelation is the moment tailoring turns unsettling. White et al. (2008) found that highly personalized emails can provoke reactance when the personalization is not justified to the recipient, and Aguirre et al. (2015) found that covert collection of the data behind an ad reduced its effectiveness while overt collection did not.
Three Tiers of Customer Data
Told data is what customers entered themselves: pet names, species and weights in the app, and answers to surveys. Observed data is what the chain recorded from transactions: bag sizes, reorder gaps and schedule edits. Inferred data is what the model produces: churn risk, likely next purchase and predicted pet age. Each tier gets a different permission. Told data may appear in a message's wording. Observed data may set a message's timing. Inferred data may choose which offer is sent but may never surface in the copy.
Inferred data may decide what we offer, but it must never show in what we say.
Timing That Feels Like Service
The strongest use of observed data is a reminder sent when a 30-pound bag should be running low, calculated from the pet's weight and the bag size. Customers tend to read accurate timing as attention rather than surveillance, because the reason for the timing is obvious: they bought the bag, and the chain knows how much a dog of that size eats.
Where Tailoring Unsettles
Three message types fail the recognition test and are banned by name. A subject line mentioning a prescription kidney diet reveals a pet's illness that the customer never mentioned in any message, even though the chain recorded the purchase; the Unit 4 review already dropped prescription purchases from the model, and this rule keeps them out of copy too. A greeting that mentions a pet's predicted age reveals an inference. And any line hinting that the chain noticed a slowdown in orders, such as we have missed you, reveals that the customer is being watched for signs of leaving. Kim et al. (2019) found that ad transparency helps when the reason for targeting is acceptable to the consumer and hurts when it is not, and each banned message carries a reason the customer would not accept.
The Eight Rules
Rule one: told data may appear in wording. Rule two: observed data may set timing but not wording that describes the customer's behavior. Rule three: inferred data may select an offer but never appear in copy. Rule four: no message may mention a pet's health condition unless the customer raised it in that conversation. Rule five: cancellations citing a pet's death stop every product message for twelve months. Rule six: offers climb a fixed ladder before any discount. Rule seven: discounts are capped at 15 percent, no more than twice a year per subscriber. Rule eight: every rule is checked against the Unit 4 rulings, and nothing approved there is contradicted here.
After a Pet Dies
Rule five deserves its own argument. A subscriber who cancels because a pet died receives no product email, text or offer for twelve months. The nearest store sends one handwritten card. The memo accepts that some of these customers will later adopt another animal and return without any prompt, and that a few might have returned sooner with one. That cost is worth paying, because a discount arriving after news of a pet's death would read as the chain not listening.
A Ladder Before a Discount
Offers climb in a fixed order: first a delivery delay, then a change to a smaller bag, then free delivery on the next order, and only at the last step a discount, capped at 15 percent and limited to twice a year. The ladder answers the worry raised in Unit 1 that risk-based offers teach customers to hesitate. A customer who pauses learns that the chain will help adjust the schedule, not that pausing earns money, and loyal customers are not paying for others' hesitation.
Measures
Three measures will show whether the rules work, each read from the pilot rather than assumed: the complaint rate among message recipients, the unsubscribe rate and cancellations among recipients compared with the control arm.
References
Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K., & Wetzels, M. (2015). Unraveling the personalization paradox: The effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing, 91(1), 34-49. https://doi.org/10.1016/j.jretai.2014.09.005
Kim, T., Barasz, K., & John, L. K. (2019). Why am I seeing this ad? The effect of ad transparency on ad effectiveness. Journal of Consumer Research, 45(5), 906-932. https://doi.org/10.1093/jcr/ucy039
White, T. B., Zahay, D. L., Thorbjornsen, H., & Shavitt, S. (2008). Getting too personal: Reactance to highly personalized email solicitations. Marketing Letters, 19(1), 39-50. https://doi.org/10.1007/s11002-007-9027-9
The Purdue Global MT362 Unit 5 instructions
In Unit 5 students write a memo setting rules for how an AI-supported marketing program may personalize messages. The memo should state the decision requested, the principle that separates helpful tailoring from intrusive tailoring and the specific rules that follow from it. Instructors expect examples in pairs, one acceptable and one not, so the principle can be seen working. Rules should be consistent with the earlier privacy review. Sensitive situations, such as a customer's loss, often need their own rule. Offers and discounts deserve limits that address the incentives the course raised earlier. Measures belong at the end: the figures that will show, once the pilot runs, whether each rule helped or hurt.
How this MT362 Unit 5 example is built
Its first page asks for approval of eight rules and sets out the recognition test, citing White and colleagues and Aguirre and colleagues. Three data tiers follow, each with a permission, under a highlighted line that inferred data may decide the offer but never show in the wording. A section on timing explains why an accurate reminder reads as service. Three banned message types are named, with Kim and colleagues on when transparency helps. The eight rules are listed in order. The pet-loss rule is argued separately, accepting its cost. An offer ladder places a capped discount last, answering the Unit 1 worry. Measures close the memo: complaints, unsubscribes and cancellations among recipients, compared with the control arm.
MT362 Unit 5 rubric: what earns full points
Personalization memos are generally assessed on the clarity of the governing principle, the specificity of the rules, consistency with earlier privacy work and the quality of the measures. One principle applied to every message type earns more than a list of separate judgments. Paired examples show the principle in action. Rubrics reward rules specific enough that a copywriter could follow them unaided. Handling sensitive situations with care shows judgment about customers. Limits on discounts that address the incentive problem show strategic thinking. Measures tied to complaints, unsubscribes and retention show the rules will be tested. Memo format and research cited in APA 7 support the recommendations.
MT362 Unit 5 help from the desk
Personalization memos often stop at a general commitment to respect privacy, with no test a writer could apply to a draft message. Others ban personalization broadly, losing the useful timing that customers welcome. A frequent gap is allowing model predictions to appear in copy, such as references to a customer's likely age or slowing orders, which reveals the watching. Sensitive cases are often handled by the same rules as everything else. Discount rules are frequently missing, which leaves the incentive problem from Unit 1 untouched. Before you send the memo, draft three real subject lines and run each through your principle. If any one of them is hard to classify, the principle needs sharpening before the rules can rest on it.
Get an MT362 Unit 5 example written to your instructions
What customer data does the Unit 5 business hold, and which messages or offers would it tailor? Bring both, plus the memo instructions. Within 24-48h a first memo, free, draws its personalization line by stated principle, pairing a helpful message with an intrusive one for each rule. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More MT362 papers and BS in Business Administration samples
- MT362 Unit 1: AI Retention Offers Post
- MT362 Unit 2: AI Use Case Assessment
- MT362 Unit 3: Customer Data Inventory
- MT362 Unit 4: Privacy and Consent Review
- MT362 Unit 6: Seminar Reflection
- MT362 Unit 7: Bias Review
- MT362 Unit 8: Pilot Test Design
- MT362 Unit 9: Vendor Comparison Matrix
- MT362 Unit 10: AI Adoption Proposal
- MT361 Unit 5: Listing Presentation Slides
- MT359 Unit 5: Publicity Plan With PESO Model
- MT358 Unit 5: AI Ad Optimization Exercise
- MT357 Unit 5: Digital Content Calendar Plan
MT362 Unit 5 questions, answered
How do I define where personalization becomes intrusive?
Pick a principle you can apply repeatedly, such as whether a customer would recognize how the business learned what the message reveals, and test every message type against it. Graders tend to reward consistency more than any particular principle. Paired examples help: one message that passes and one that fails, drawn from the same data.
Should the memo cite research on customer reactions to personalization?
Yes, where it supports a rule, but keep claims proportionate. Studies on perceived intrusiveness and privacy concern are commonly cited, and some sections expect two or three. Summarize what a study measured and in which setting, and avoid stretching a finding from one product category into a universal rule about all customers.
Does a personalization memo need numbers?
It needs measures more than projections. State what the business will track to see whether tailoring helps or harms, such as complaints, unsubscribes and purchase rates among recipients, and where those figures come from. Predicted lifts without a testing plan tend to draw comment, since the course treats every tailoring claim as something to verify.
Where can I find a free MT362 Unit 5 sample paper?
The complete MT362 Unit 5 personalization strategy memo is printed here, with eight rules and a recognition test for a pet chain's retention messages.
Why can a churn prediction choose the offer but not appear in the copy in the MT362 Unit 5 sample?
Mentioning an inference, such as a slowdown the chain noticed, reveals that the customer is being watched, which turns tailoring from helpful to unsettling.