Skip to content

Group Presentations

Format

From Week 2 onwards, there will be two presentation slots in each Lab Section, each lasting 30 minutes.

Format:

  • Presentation: 15 mins, answering prompts allocated below.
  • Panel Discussion: approx. 10 mins of Q & A from another group.
  • Open Discussion: remaining time.

Panel Discussion

Who will be the panel?

  • For each week, the two teams take turns as panel leading the Q & A.
  • For example, when Group A presents, Group B takes the role of the Q & A panel and vice versa.

What should the panel ask?

Potential questions could be based around:

  • Developing a deeper understanding of the managerial / marketing implications
  • Details on empirical strategy
  • Interpretation of results

Groups

Students will give group presentations twice in their Lab Section, once between Weeks 2 and 4 and again between Weeks 5 and 7. Students are randomly assigned to two groups, one per presentation.

Group Allocations: See Canvas.

Topics

Use the knowledge developed in the course from readings and lectures as a starting point to construct an answer to the following prompts. You should use your research skills to go looking for additional material as necessary.

Scenarios are Hypothetical!

All scenarios are entirely hypothetical, and do not reflect actual scenarios or current areas on research that any named company is facing or working on to the best of my knowledge.

Topics will be posted 2 weeks before a scheduled presentation so that all teams have an equal amount of preparation time.

Week 7

Topic A:

After the Cambridge Analytica scandal Facebook (now Meta) wants to refocus the topics of conversation consumers are discussing online about the company. They propose airing a new TV ad that centres on their focus on their new rules around data privacy. Your team at Facebook Ads has been asked to test whether this ad is effective. Propose an analysis that would test the ad’s effectiveness. In your analysis carefully think through how the marketing strategy might backfire and how you could monitor whether this is the case.

Topic B:

In response to the onset of the COVID-19 pandemic in early 2020 the US Center for Disease control is encouraging the general population to stay at home. They have reached out to your Marketing Analytics team at a large consulting firm to assess the effectiveness of asking celebrities to tweet encouraging people to stay at home and use the hashtag #stayhomesavelives as opposed to similar messages being sent from the official CDC account. Develop an empirical strategy to test the relative effectiveness, and consider whether using celebrities might cause different shifts in the way the public discuss the pandemic.

Week 6

Topic A:

Peter Panda (@peter_pannnda) is about to launch a new startup called Plushfluence that helps members of the Instagram plushie community increase the likeability / virality of their posts. Propose an analytics framework to analyse what makes plushie posts more “likeable”, and a framework to test whether your findings can be supported.

Topic B:

Building on existing knowledge on improvised marketing interventions (IMI), the company ViralBrand has hired your team to develop an understanding about how and why IMI increases consumer brand engagement and perceptions. Develop a plan that allows you to investigate how IMI influences these key metrics and isolate what components of an IMI strategy are most effective.

Week 5

Topic A:

Your team has been hired by Influentials (a European Influencer agency) to assess the performance of various influencer segments as defined by their reach in generating customer engagement. The firm is open in letting you analyse any of their data, and is willing to listen to possible ideas for an experiment to be part of the analysis strategy. Propose an analysis strategy that will allow you to measure the performance of different influencer segments and explain the implications your findings could lead to.

Topic B:

Suppose Instagram (owned by Meta) has launched a project to understand how they can earn additional revenue from influencer posts. Assume that at present Instagram does not charge an influencer or the company the influencer is hired by for a sponsored post. Propose an analysis strategy that could deliver a causal estimate of the effect of monetizing influencer posts for Instagram. In developing your presentation be sure to provide some strategic guidance on how they could scale this up and implement your solution across the whole platform. Are there any guardrail metrics they should track?

Week 4

Topic 1

While working in Activision Blizzard’s (a large video game publisher) Marketing Analytics team you’ve been assigned the task to quantify the effect of Word of Mouth (Volume and Sentiment) on the demand for its video games. Propose an empirical strategy that would allow them to estimate these effects credibly without the need for experimentation. When pitching your strategy, discuss why an experiment is most likely not feasible.

Topic 2

Yelp.com’s Customer Experience team has discovered a stark difference in the ratings of national chain restaurants vs independent restaurants: The modal review for a Chain is 1 star out of 5 whereas for independent restaurants it is 5 stars of 5. After presenting this stylized fact to the larger data science group, they’ve identified that the next phase of analysis should aim to discover why this might be the case and have assigned your team to lead the analytics. Develop an empirical strategy to answer this question, and relate the (expected) findings back to consumer behaviour.

Week 3

Topic 1:

Bol.com’s Seller Experience team is looking to expand the number of third party vendors who sell on the site. One problem they are facing is that new vendors have less, and noisier reviews by consumers about their reputation. Propose a new strategy that they could adopt to help new seller’s reputations on Bol.com and an analytical framework that you could use to test whether this strategy is successful.

Topic 2:

The Dutch Authority for Consumers and Markets (ACM) has recently located and successfully shut down a market for fake reviews within the Netherlands that specialized in fake reviews for Washing Machines and Dryers. Cool Blue’s Head of Data & Analytics team has reached out to your group to understand the effect this might have on the company. Why would fake reviews potentially matter for CoolBlue and how might your team go about evaluating what the effects of the fake review shutdown are?

Week 2

Topic 1:

TikTok’s User Experience team is worried about the amount of lurkers on their platform and wants to design a new strategy to get them to engage with or post content. Propose a potential strategy that could lead to increases in either form of engagement by lurkers along with an empirical strategy to test whether the strategy is effective.

Topic 2:

Reddit’s Customer Analytics team wants to understand potential reasons why users contribute to their site, either by posting original content or commenting on the posts of others. Design a project that can potentially disentangle user motivations along with an empirical strategy to test for the prevalence of the each, and potentially classify a user by their motivations.

Week 1

No presentations.