Team Y · Boycotts and store visits at Target, 2023 and 2025
Lena Park, Marco Silva, Nadia Haddad, Owen Brooks, Priya Natarajan
This is a hypothetical example. The team, its members and the dates of its work are invented. The boycotts, Target’s sales figures and store count, the data sources and the studies are real. It asks a question a company would pay to answer. The solar farm example asks one a town would act on. Both follow the instructions section by section. The PDF shows this one within the three-page limit, with the references and the AI disclosure on a page of their own.
1. Research question
Did the 2023 and 2025 boycotts of Target lower visits to its stores most in the counties whose politics matched each boycott? This is an evaluation. A retailer’s marketing team, deciding where to spend on promotions to win shoppers back after a boycott, would act on the answer. The 2023 boycott came from conservative shoppers angry about Pride merchandise, and the 2025 boycott from shoppers opposed to Target ending its diversity goals. We expect visits to fall a few percent more in Republican-leaning counties in 2023, and in Democratic-leaning counties in 2025. We are wrong if visits fall by the same amount everywhere, or if the same counties fall most in both years.
2. Why it matters
Both boycotts fell in quarters when Target’s sales declined. Its comparable store sales fell 4.3 percent in the second quarter of 2023 and 5.7 percent in the first quarter of 2025 (Target Corporation 2023, 2025). In 2023 the company said it could not separate the effect of the reaction from other causes (Associated Press 2023b). If the losses concentrate in counties whose politics match the boycott, a marketing team can aim its promotions at those stores. If they do not, it spends nationally. A retailer weighing a public stance could also estimate in advance which stores it puts at risk, and for how long.
Two studies come closest to ours. Hou and Poliquin (2023) used phone location data to count visits to the stores of several large retailers whose chief executives backed expanding background checks for gun purchases. Visits fell about 3 percent over four weeks, about 5 percent in the most conservative counties and not at all in the most liberal ones, and the drop faded within 10 weeks. Liaukonytė, Tuchman and Zhu (2023) studied the boycott of Goya after its chief executive praised President Trump. A counter-movement of supporters outweighed it. Sales rose about 22 percent, and the rise was gone within three weeks. The increase was largest in heavily Republican counties, with no strong offsetting fall in heavily Democratic ones. Each study follows stances on one side of politics. We study one retailer boycotted from opposite sides two years apart, which lets us test whether the same stores move in opposite directions.
3. Approach
Each observation is one Target store in one week. The outcome is the logarithm of the store’s weekly visits, from Advan’s Weekly Patterns Plus data in Dewey. The sample covers all of Target’s nearly 2,000 stores in the United States, from January 2022 to December 2025, which spans both boycotts. These are the stores a marketing team would choose among.
The treatment is the boycott. Every store was exposed to it, so we split stores by the politics of their county, measured by the Republican share of the 2020 presidential vote. We compare stores in the third of counties with the highest share against stores in the third with the lowest. We date the 2023 boycott from the week of May 22, when Target pulled some Pride merchandise after confrontations in its stores. The collection had drawn misleading videos for weeks before that (Associated Press 2023a), so we leave out the weeks from its launch in early May to May 22. We date the 2025 boycott from the week of January 20, when Target announced it would end its diversity goals, and we mark the 40-day boycott that began on March 5 (Hopkins and Gray News staff 2025).
The comparison group for stores in the Republican third is the same chain’s stores in the Democratic third, and the reverse. Their visits stand in for the counterfactual, what each group’s visits would have done without the boycott. Both groups share Target’s prices, products and national advertising, so anything the chain did to every store drops out. The counties differ in income, density and much else, so we compare each store’s change in visits over time rather than its level. The design is difference-in-differences, run as an event study for each boycott. It compares the change in visits at stores in the Republican third with the change at stores in the Democratic third, week by week, from 20 weeks before each boycott to 20 weeks after. We make the comparison among counties of similar income, so that a shock hitting poorer counties does not show up as politics. This design measures how much more visits fell in one group of counties than in the other. It cannot measure the boycott’s total effect on Target, because every store was exposed. The identifying assumption is that, within each income group and without the boycotts, visits in the two groups of counties would have followed parallel paths.
The two boycotts also give us a check. If the boycotts drive the response, the gap between the two groups should change sign between 2023 and 2025. A shock that hits the same counties in both years, such as prices rising faster in poorer counties, would push the gap the same way both times. A shift in spending that follows county politics could flip it too, and section 5 explains how we separate that.
4. Data and access
| Dataset | Holder | Unit and period | Variables we need | Access time | Status |
|---|---|---|---|---|---|
| Weekly Patterns Plus (Advan) | Dewey, through Cornell’s subscription | Store-week, 2022–2025 | Brand, weekly visits, location | Days | File opened Oct 15 |
| County Presidential Election Returns 2000–2024 | MIT Election Data and Science Lab, public | County, each election | Republican share of the 2020 vote | Now | File opened Oct 12 |
| American Community Survey, 5-year estimates | U.S. Census Bureau, public | County, 2020–2024 | Median household income | Now | File opened Oct 12 |
We already hold all three datasets. If the visit data cover too few stores in rural counties, the question narrows to stores in metropolitan counties.
Weekly Patterns Plus counts visits to places and identifies no person, which normally puts it outside IRB review. We do not get to make that call ourselves, so we asked the IRB office on September 18. It confirmed on October 9 that the project needs no application. We need no data-use agreement, because we have no sponsor and the other two sources are public. A retailer would pay for this answer, so if one offered to sponsor the work, we would first ask whether Cornell’s Dewey licence allows commercial use.
5. Anticipated challenges
The main threat is omitted variables: a shift in spending that follows county politics in the same weeks, with no boycott behind it. Our 2025 window opens in September 2024 and spans the presidential election and the inauguration. If shoppers in Democratic-leaning counties cut spending at every store after either event, Target’s gap would move the way the boycott predicts, and the sign check could not tell the two apart. That would make our 2025 estimate too large. Visits to other large chains in the same counties separate the two, because a boycott of Target lowers visits at Target and a general cut lowers them everywhere. We will use chains that sell different goods from Target and took no public stance on diversity goals. If their gap moves the same way as Target’s, we will report by February 15 that our 2025 estimate cannot separate the boycott from a partisan shift in spending.
In 2023, Target moved its Pride merchandise to the back of some Southern stores (Associated Press 2023a). If that softened the reaction where conservative shoppers are concentrated, it pushes our 2023 estimate toward zero. We cannot observe which stores made the change, so we will state it as a limitation in the outline by November 11.
6. Team plan
| Member | Role | Delivers | By | Backup |
|---|---|---|---|---|
| Lena Park | Coordinator | Runs the check-in. Uploads each deliverable | Every Wednesday | Owen |
| Marco Silva | Data | Store-weeks for Target and the comparison chains, 2022–2025, cleaned and described | Nov 6 | Priya |
| Nadia Haddad | Events | Dated timeline of both boycotts, with a source for each date | Nov 4 | Lena |
| Owen Brooks | Design | The design section of the outline, with the check that the two groups of counties moved together before each boycott | Nov 4 | Marco |
| Priya Natarajan | Counties | Vote share and income matched to every store | Nov 6 | Nadia |
Missed work. A miss is a task not done by its date without warning. A member who warns the team before the date has not missed it, and we agree on a new date at the next check-in.
- First miss. The coordinator notes it on the check-in, with a new date.
- Second miss. The coordinator and the backup meet the member that week. The backup takes over the task, and the check-in records it.
- No change, or a week without a reply. We email the instructor, with the check-ins attached.
References
Associated Press (2023a). Target pulls some LGBTQ+ merchandise from stores ahead of June Pride month after threats to workers. May 23. https://www.wlbt.com/2023/05/23/target-makes-changes-lgbtq-merchandise-june-pride-month-after-threats-workers/
Associated Press (2023b). Target sales ebb after Pride month shopper backlash; company cuts profit outlook for 2023. August 16. https://www.nbcnews.com/business/corporations/target-q2-2023-sales-inflation-pride-month-shopper-backlash-rcna100187
Hopkins, R., II and Gray News staff (2025). Pastor calls for Target boycott after company ends DEI policy. Gray News, March 6. https://www.kold.com/2025/03/06/pastor-calls-target-boycott-after-company-ends-dei-policy/
Hou, Y. and C. W. Poliquin (2023). The effects of CEO activism: Partisan consumer behavior and its duration. Strategic Management Journal 44(3): 672–703. https://doi.org/10.1002/smj.3451
Liaukonytė, J., A. Tuchman and X. Zhu (2023). Frontiers: Spilling the beans on political consumerism: Do social media boycotts and buycotts translate to real sales impact? Marketing Science 42(1): 11–25. https://doi.org/10.1287/mksc.2022.1386
Target Corporation (2023). Target Corporation reports second quarter earnings. Form 8-K, Exhibit 99, August 16. https://www.sec.gov/Archives/edgar/data/27419/000002741923000032/a2023q2ex-99.htm
Target Corporation (2025). Target Corporation reports first quarter earnings. Form 8-K, Exhibit 99, May 21. https://www.sec.gov/Archives/edgar/data/27419/000002741925000096/a2025q1ex-99.htm
AI disclosure
Tools. ChatGPT (chatgpt.com).
Used for. Suggesting search terms for studies of boycotts and store visits. Suggesting how to date the start of each boycott.
Checked. We opened both studies, Target’s two earnings releases and the three news reports, and confirmed every number and date we quote. One study the tool suggested does not exist, and we dropped it. We put no Dewey data into the tool.