MBA FPX 5008 Assessment 2 Costed Recommendation From Data example

Reviewed by Rupert Danvers, MBA Applied Business Analytics Capella University Free custom sample in 24 to 48h

This MBA FPX 5008 Assessment 2 sample paper builds a costed recommendation from data, moving from an analytic finding on late deliveries to a business action at an online furniture retailer. It belongs to Capella MBA FPX 5008, Applied Business Analytics, which Capella MBA students take as MBA-FPX5008. The paper gives the finding in brief and names the gap between finding and action. It proposes reallocating shipments among carriers, scales the action to the uncertainty in the data and computes the return from the findings. It explains what the return means for the decision, covers operational considerations and states what the data cannot tell.

CourseMBA FPX 5008 Applied Business Analytics
AssessmentAssessment 2
Paper typeCosted recommendation from data
LengthAbout 1,080 words, 5 pages
FormatAPA 7 student paper
SchoolCapella University
ProgramMBA
UpdatedSeptember 2026

Free sample paper for MBA FPX 5008 Assessment 2

1

From Analytic Finding to Business Action: Reallocating Carriers After a Late-Delivery Analysis at an Online Furniture Retailer

Student Name

Capella University

MBA FPX 5008: Applied Business Analytics

Instructor Name

Month Day, Year

What this part is doingThe title names the finding's destination, a change in carrier allocation, and the business. The assessment is judged on whether analysis becomes a decision, so the decision is named first.
2

Introduction

Analysis creates value only when it changes a decision. An online furniture retailer shipping about 96,000 large orders a year had noticed rising complaints about late deliveries. An analysis of last year's orders, by carrier and delivery zone, found a concentrated problem: one of the retailer's two freight carriers delivered 31% of its rural orders late, compared with 9% for the other carrier in the same zones, while the two performed similarly in urban and suburban zones. This paper turns that finding into a proposed action, scales the action to the uncertainty in the finding, computes the expected return from the analysis, and states what the data cannot show. The retailer is a composite used for illustration.

The Finding in Brief

The analysis covered 96,400 delivered orders. Late delivery was defined as arrival after the delivery window promised at checkout. Overall, 12.8% of orders were late. In rural zones, which accounted for 18% of orders, carrier A delivered 31% late and carrier B 9%. Late orders were three times as likely to generate a customer service contact and twice as likely to result in a partial refund or return, and customers whose first order arrived late were less likely to order again within a year.

The Gap From Finding to Action

A finding describes; an action changes something. The finding that carrier A is late in rural zones suggests several possible actions: move rural volume to carrier B, renegotiate carrier A's rural service terms, widen the promised delivery window for rural orders, or add a regional warehouse closer to rural customers. The gap between finding and action is where many analytics projects stall, because the analysis stops at description and leaves managers to guess what to do. Davenport (2006) described firms that compete on analytics as those that embed analysis in operational decisions rather than producing reports that sit unused.

The Proposed Action

The proposed action is to route rural orders to carrier B wherever carrier B serves the zone, which covers about 80% of rural orders, and to renegotiate carrier A's contract to include on-time performance standards for the remaining rural zones. Carrier B's rural rate is about 14 dollars higher per order than carrier A's.

Uncertainty Scaled Into the Action

The finding is strong but not certain. Carrier B's better rural performance might not hold if its volume rises sharply, since its rural network could become congested. Rather than moving all rural volume at once, the retailer will shift rural orders in two regions for three months while keeping carrier A in comparable regions, then compare late rates. Scaling the action to the uncertainty means spending enough to learn whether the change works under higher volume, without betting all rural deliveries on an assumption. Provost and Fawcett (2013) describe data-driven decision-making as grounding decisions in analysis of data, and a staged shift generates the data needed to confirm the decision before it is made permanent.

What this part is doingThe action is staged because the finding may not hold at higher volume. Scorers reward recommendations whose scale matches the strength of the evidence.
3

Return Computed From the Findings

The return is computed from the analysis. Rural orders number about 17,350 a year, of which about 13,900 could move to carrier B. At carrier A's 31% late rate, about 4,310 of these would be late; at carrier B's 9%, about 1,250, a reduction of about 3,060 late orders. The analysis found that each late order cost an average of 41 dollars in partial refunds, service contacts, and redelivery, so the direct savings would be about 125,000 dollars. The added freight cost would be about 195,000 dollars (13,900 orders at 14 dollars). On direct costs alone, the change loses about 70,000 dollars a year. The analysis also found that customers whose first order arrived late reordered at a rate 8 percentage points lower. About 40% of rural orders are first orders, so the change would prevent about 1,220 late first orders, and at an 8-point higher reorder rate, about 98 additional customers would reorder. With an average reorder margin of about 380 dollars, that adds about 37,000 dollars, bringing the net to about a 33,000 dollar loss in the first year, with further retention value in later years.

What the Return Means for the Decision

On the numbers, moving all eligible rural volume to carrier B is not justified by first-year economics. The better action is narrower: move first-time rural orders to carrier B, where the retention benefit is largest, and renegotiate carrier A for repeat customers. That narrower action costs about 78,000 dollars in added freight for about 5,560 first orders and saves about 50,000 in late-order costs plus about 37,000 in retention, a small net gain. Computing the return from the analysis itself changed the recommendation, which is the point of doing the arithmetic before committing to the action.

Operational Considerations

Routing first-time rural orders to carrier B requires a change in the order management system, which currently assigns carriers by zone and weight alone. The change adds a rule that checks whether the customer has ordered before, which the system can do from the customer account. The warehouse must stage carrier B shipments separately, which the warehouse manager estimates adds a few minutes per pickup. Customer service will be told about the change so that agents can explain delivery timing accurately. Naming these operational steps shows that the action is executable, not only financially sensible on paper.

What this part is doingThe recommendation is checked against the systems and people who would carry it out. Scorers reward analytics recommendations that account for implementation.
4

What the Data Cannot Show

The data cannot show whether carrier A's rural performance would improve under a contract with penalties, how customers weigh speed against price, or whether late delivery affects reviews that influence other customers. They also cannot show whether carrier B's performance will hold under more volume, which the staged test addresses.

Measures

The test will track late rate by carrier and zone, cost per order, service contacts, refunds, and reorder rate at six and twelve months. If carrier B's rural late rate rises above 15% under the added volume, the shift will pause and the renegotiation with carrier A will take priority.

Conclusion

The late-delivery finding becomes useful when it is translated into a specific action, tested at a scale matched to its uncertainty, and evaluated with a return computed from the analysis. Here the arithmetic narrowed the action to first-time rural orders, where the benefit is large enough to justify the cost.

5

References

Davenport, T. H. (2006). Competing on analytics. Harvard Business Review, 84(1), 98-107.

Provost, F., & Fawcett, T. (2013). Data science and its relationship to big data and data-driven decision making. Big Data, 1(1), 51-59. https://doi.org/10.1089/big.2013.1508

What the MBA FPX 5008 Assessment 2 instructions ask for

Assessment 2 in MBA FPX 5008 has you turn a data finding into a recommendation with costs and returns. You summarize the finding, explain what action it suggests, design the action, adjust its size to how certain the finding is, estimate the costs and the financial return using figures from the analysis, interpret what the return means for the decision, consider how the action would work in practice and state the limits of the data. A pilot or phased approach is appropriate when the evidence is not strong enough to justify a full commitment. Business analytics and finance texts support the recommendation.

How this MBA FPX 5008 Assessment 2 example is built

Finding, action, cost, uncertainty. The opening states the finding compactly, leaning on the earlier analysis instead of restating it. An action block says what should be done, specifically enough to be started. A cost block prices it, including implementation effort rather than only spend. A return block estimates the benefit using the figures the analysis produced, with the arithmetic visible. An uncertainty block carries the analysis's limits into the recommendation, scaling the action to the strength of the evidence. A short block names what the analysis cannot support, in case the reader was hoping for it. The closing states what additional data would justify going further. Every figure carries its basis, and no benefit is claimed beyond what the finding supports.

MBA FPX 5008 Assessment 2 scoring guide: what Distinguished looks like

The scoring guide weighs the finding, the link to action, the design, the treatment of uncertainty, the return calculation and the limits. The finding reaches Distinguished when summarized briefly with its strength. The link to action scores higher when the gap between knowing and doing is addressed. The design earns credit when specific, such as moving a share of volume from one carrier to another in named regions. Uncertainty is judged on whether the size of the action matches the strength of the evidence. The return should be computed from the findings, not assumed. Operational issues, like carrier capacity, should be considered. Limits should be stated. Distinguished recommendations are sized to what the data can support.

MBA FPX 5008 Assessment 2 help: mistakes that cost marks

Recommendations from data often leap from a correlation to a full rollout without asking how certain the finding is. Another weakness is estimating a return from optimistic guesses rather than the figures in the analysis. The action is vague, like "improve delivery performance." Costs are incomplete, leaving out contract penalties or switching time. Operational constraints are ignored. Recommending more data collection instead of a decision is treated as a safe answer. Sizing the first step as a pilot in the worst region, and computing its return from the analysis, keeps the recommendation defensible and prepares the executive synthesis in Assessment 3.

Get a MBA FPX 5008 Assessment 2 example written to your instructions

Send the Assessment 2 instructions and your MBA FPX 5008 scoring guide, along with the analysis your first assessment produced. We write a custom example against those criteria and return it in 24 to 48 hours. The first custom sample is free, and scaling the action to the strength of the evidence is the move worth taking from it. The paper above is an original model document written by our desk, not a submitted student paper and not an official Capella University document.

More MBA FPX 5008 papers and MBA samples

MBA FPX 5008 Assessment 2 questions, answered

How do I cost a recommendation from a dataset?

Work from what the action requires rather than from the data. Hours, licenses, a change somebody has to implement. The data gives you the expected benefit; the cost comes from the operation. Stating both with their bases is what allows anybody to judge whether the recommendation is worth taking.

What if the finding is weak?

Then recommend something small and reversible, and say that is why. A pilot, a trial in one region, a change that can be undone. Matching the size of the action to the strength of the evidence demonstrates better judgment than either ignoring a weak result or acting on it as though it were strong.

Should I recommend collecting more data?

As part of the answer, not instead of one. Saying what further evidence would justify a bigger move is valuable; recommending more analysis in place of a decision is the version that scores badly, since the manager asked what to do with what is already known.

Where can I find a free MBA FPX 5008 Assessment 2 sample paper?

This page presents the full carrier reallocation recommendation, with the finding in brief, the gap to action named, uncertainty scaled into the action and the return computed from the findings.

What if the finding is weak in MBA FPX 5008 Assessment 2?

Scale the action to match, such as a small pilot with a clear measure, rather than a full rollout. The recommendation can still be firm about the next step.