| Course | BU 443 Marketing Analytics |
|---|---|
| Unit | Unit 5 |
| Paper type | Campaign evaluation |
| Length | About 990 words, 6 pages |
| Format | APA 7 student paper |
| School | Herzing University |
| Program | BS in Marketing |
| Updated | September 2026 |
Free sample paper for BU 443 Unit 5
Probably Helped, Probably Did Not Pay: A Campaign Evaluation of a Composite Online Houseplant Retailer's Three-Week Pet-Safe Plants Push
Student Name
Business, Herzing University
BU 443: Marketing Analytics
Instructor Name
May 13, 2026
Probably Helped, Probably Did Not Pay: A Campaign Evaluation of a Composite Online Houseplant Retailer's Three-Week Pet-Safe Plants Push
The Campaign and the Question
From April 6 to April 26, 2026, Greenline Plant Co., the composite Portland houseplant seller followed in this course, ran a campaign promoting its pet-safe plants: species generally listed as non-toxic to cats and dogs. It combined two emails to the full list, paid social ads to pet owners in the continental United States and a new collection page. The campaign cost $4,500: $3,800 in ad spend and $700 for design and photography. Weekly sales of the pet-safe category rose from an average of $9,200 in the eight weeks before to $16,400 during the campaign.
The owners asked whether the campaign worked. This evaluation asks a narrower question: what is the rise in pet-safe sales evidence of? It establishes what sales were doing before the campaign, states the comparison used and why, lists everything else happening in the same window, and scales its verdict to what the evidence bears. All figures belong to the composite case.
What Sales Were Doing Before
A before-and-after comparison alone would credit the campaign with the whole rise, from $9,200 to $16,400 a week, a 78 percent increase. That comparison ignores the season. Houseplant sales climb every April as people spend more time on balconies and in newly bright rooms. In the same three weeks of 2025, with no campaign, pet-safe sales averaged $11,800 a week, already well above their late-winter level. Much of the rise in 2026 would have happened anyway.
The Comparison Used
The evaluation compares the change in pet-safe sales with the change in all other plant categories over the same weeks, year over year. Other categories received no campaign but faced the same season, the same site and the same shipping conditions, so their growth estimates what pet-safe sales might have done without the push. Shadish et al. (2002) describe this kind of design, comparing a treated group with an untreated comparison group before and after an intervention, as a way to rule out threats such as seasonal trends that affect both groups.
| Category | Weekly sales, April 6 to 26, 2025 | Weekly sales, April 6 to 26, 2026 | Change |
|---|---|---|---|
| Pet-safe plants | $11,800 | $16,400 | +39% |
| All other plants | $41,600 | $50,750 | +22% |
If pet-safe sales had grown at the 22 percent rate of other categories, they would have averaged about $14,400 a week. They averaged $16,400, so the estimated campaign effect is about $2,000 a week, or about $6,000 over three weeks. At Greenline's gross margin of about 55 percent, that is roughly $3,300 of gross profit, against a campaign cost of $4,500. On this estimate, the campaign added sales but did not recover its cost within the three weeks.
Everything Else in the Same Window
Four other events overlapped the campaign, and each could have moved pet-safe sales. Easter fell on April 5, the day before the campaign began, and holiday gifting may have lifted early April sales in all categories. On April 14 a widely shared video by a pet-care creator, unconnected to Greenline, warned cat owners about toxic houseplants; search traffic to the pet-safe page rose sharply that week, and some of that rise would have happened without Greenline's campaign. A competitor ran a 20 percent off sale in the week of April 20. And April 2026 was warmer than April 2025 in much of the country, which may have raised plant sales generally.
The comparison design handles the events that affect all categories alike, such as Easter and warm weather. It cannot handle the viral video, which affected pet-safe plants specifically, just as the campaign did. The estimated $2,000 a week is therefore an upper bound on the campaign's effect, since part of it may belong to the video.
Separating Email From Paid Social
The campaign used two channels, and the owners asked which did the work. The data allow only a rough split. The two emails were sent to existing customers and subscribers and cost almost nothing beyond design; tagged links from them led to 212 pet-safe orders during the three weeks. The paid social ads led to 164 tagged orders at a cost of $3,800, about $23 per order. But tagged orders are the attribution view the channel analysis warned against, and email recipients were people who already buy from Greenline, many of whom might have ordered in April anyway. What can be said is that the email portion cost little and plausibly added sales, while the paid portion carried almost all of the campaign's cost. If the campaign is repeated, running the emails to everyone but the ads only in some regions would isolate the part that costs money, which is the part whose value is in doubt.
The Verdict, Scaled to the Evidence
The evidence supports three statements, in decreasing order of confidence. First, pet-safe sales rose faster than other categories during the campaign, by about 17 percentage points of growth. Second, the campaign probably caused part of that difference, though some belongs to the viral video. Third, the campaign probably did not pay for itself within three weeks, since even the upper-bound estimate falls short of its cost by about $1,200 of gross profit.
The evidence does not support claiming the campaign caused the 78 percent rise, or that it failed. New customers acquired during the campaign may buy again later, which the cohort analysis will show by summer. Lewis and Rao (2015) show how hard it is to measure advertising returns precisely even with large experiments, and an observational comparison of three weeks cannot do better. A negative or modest result is a legitimate finding; the evaluation recommends that any repeat of the campaign be run as a test, with the ads shown in some regions and withheld in others, so that the next evaluation can measure the effect rather than estimate its upper bound. Farris et al. (2010) recommend judging campaigns on incremental profit rather than on sales during the campaign period, and this verdict follows that rule. Greenline and every figure are part of the invented case.
References
Farris, P. W., Bendle, N. T., Pfeifer, P. E., & Reibstein, D. J. (2010). Marketing metrics: The definitive guide to measuring marketing performance (2nd ed.). Pearson Education.
Lewis, R. A., & Rao, J. M. (2015). The unfavorable economics of measuring the returns to advertising. The Quarterly Journal of Economics, 130(4), 1941-1973. https://doi.org/10.1093/qje/qjv023
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.
BU 443 Unit 5 instructions, in plain terms
BU 443 Unit 5 asks students to evaluate a marketing campaign. Prompts generally require establishing, with data, what the measure was doing before the campaign, choosing and justifying a comparison, identifying other events in the same period, estimating the campaign's effect and stating a verdict that matches the strength of the evidence. Many instructors ask whether students can ever claim causation and how to handle seasonality. A negative or modest result is usually acceptable if well argued. Composite data may be used. The evaluation typically pairs a comparison table with APA 7 reasoning, and strong submissions translate sales effects into profit against the campaign's cost.
How this BU 443 Unit 5 example is built
The evaluation describes the campaign, its cost and the headline rise, then narrows the question. A before section shows how last year's same weeks reveal the season's effect. A comparison section introduces the category-based design with a citation and a table of year-over-year changes, then converts the estimated effect into gross profit against cost. A section lists four overlapping events and explains which the design handles and which it cannot, making the estimate an upper bound. A section separates the email and paid parts of the campaign as far as the data allow. The verdict section grades three statements by confidence and recommends running any repeat as a regional test.
Where the points sit in the BU 443 Unit 5 rubric
Campaign evaluation rubrics at Herzing typically reward a clear baseline, a justified comparison, attention to confounding events, a quantified effect and a verdict proportionate to the evidence, with writing and sources. The baseline criterion earns full credit when seasonality is addressed. The comparison criterion rewards explaining why the comparison group resembles what would have happened without the campaign. Confounders score highest when each overlapping event is named and the student says which the design controls. The effect criterion pays for converting sales into profit and setting it against cost. The verdict criterion, often decisive, rewards graded language that claims no more than the evidence supports.
BU 443 Unit 5 help before you submit
Campaign evaluations commonly lose points by comparing sales during the campaign with sales just before, crediting the campaign with seasonal growth. Another frequent problem is ignoring events in the same period, such as holidays or unrelated media attention. Students often report revenue effects without subtracting costs or applying a margin. Some claim the campaign caused the result because sales rose. Others declare failure because profit fell short, without noting that new customers may buy again later. Many evaluations blur the contributions of different channels in one campaign. Finally, an evaluation that ends without suggesting how to measure better next time misses the most useful conclusion available. Writing the verdict in graded words, such as probably or unlikely, keeps the conclusion honest when the design cannot rule out every rival explanation.
Get a BU 443 Unit 5 example written to your instructions
Send the Unit 5 instructions and your BU 443 rubric, along with the campaign or dataset the case supplies. We write a custom example that establishes prior behavior first, justifies its comparison, reports the change before interpreting it, lists what else ran and sizes the verdict to the design. First custom sample free, back in 24 to 48 hours.
More BU 443 papers and marketing samples
- BU 443 Unit 1: Metrics definition exercise
- BU 443 Unit 2: Data collection and tracking audit
- BU 443 Unit 3: Channel performance analysis
- BU 443 Unit 4: Customer and cohort analysis
- BU 443 Unit 6: Test design and reading
- BU 443 Unit 7: Dashboard or visualization
- BU 443 Unit 8: Analytics recommendation
- BU 323 Unit 5: Brand narrative
- BU 439 Unit 5: Paid campaign plan
- BU 106 Unit 5: Operations or supply paper
- BU 217 Unit 5: Stakeholder analysis
BU 443 Unit 5 questions, answered
Can I ever say a campaign caused the result?
Only where the design supports it, which normally means something was randomized or a comparable group went untreated. Most course datasets are observational, and the honest verdict is that the movement is consistent with the campaign having worked while other explanations stay open. Markers reward the writer who names that limit, because it shows the design was understood rather than ignored.
What should I do about seasonality?
Name it, show it where the data reaches back far enough, and pick a comparison accounting for it rather than one ignoring it. Comparing the same span against the previous year handles seasonality and imports a year of other changes, which is a trade you state rather than hide. Where the history is too short, say the seasonal effect cannot be separated.
Is a negative result acceptable to submit?
Yes, and it often produces the better paper because there is no temptation to overclaim. Report what the measure did, apply the same scrutiny you would give a rise, and consider whether the design could have detected an effect at all at the volumes involved. A campaign that moved nothing, evaluated carefully, meets the criteria as fully as one that did.
Where can I find a free BU 443 Unit 5 sample paper?
The BU 443 Unit 5 campaign evaluation is available free on this page, with the comparison table, the confounding events and margin notes. If your campaign is different, send the details and a custom evaluation follows at no charge, usually within two days.
Can BU 443 Unit 5 say a campaign caused the result?
Only as strongly as the design allows. The sample uses a comparison with other categories to estimate an effect, calls it an upper bound because of an overlapping event, and grades its verdict as probable rather than proven.