META MARKETING ANALYTICS PROFESSIONAL CERTIFICATE

Course 5: Marketing Analytics with Meta

Week 5: Marketing Analytics in Action

Coursera Study Guide

Click to Enroll in Coursera Meta Marketing Analytics Professional Certificate

CONTENT

In the final week, you will dig deeper into how to best assess your goals and KPIs through formulating and testing hypotheses. You’ll learn how to use test results to formulate next step recommendations for your marketing campaigns. Later in the week you’ll be given a template to create presentations for communicating your results effectively to stakeholders as well as best practice tips for conducting the presentation itself.

Learning Objectives

  • Present results of your marketing analysis
  • Make data-driven recommendations
  • Recommend a test and analysis method
  • Choose a measurement method to evaluate your hypothesis
  • Obtain and evaluate data for your analysis
  • Formulate a hypothesis based on your business goals and KPIs

PRACTICE QUIZ: ASSESS YOUR DATA AND HYPOTHESIZE

1. When setting a goal, it’s best to keep it SMART or specific, measurable, achievable, relevant and…

  • Time bound (CORRECT)
  • Trackable
  • Topical
  • Timeless

Correct: That’s right! Having a start and end time for a goal is important; it helps to keep the goal constrained and manageable.

2. Jamie has a goal to get more website visits in the month of May. Is this goal SMART?

  • No (CORRECT)
  • Yes

Correct: That’s right! It’s a good start to a goal, but it’s not very specific. Adding a specific number of website visits like 500 or 5000 would make this goal more useful.

3. True or false: Third party data is more valuable than first party data.

  • False (CORRECT)
  • True

Correct: That’s right! First party data is preferred. It’s collected under a privacy agreement that you made with your customers and third party data collection is increasingly blocked.

4. What is a test hypothesis?

  • A hypothesis that has already been experimentally tested.
  • A hypothesis that includes the way you intend to run your experiment.
  • A hypothesis that includes an anticipated relationship between an independent and a dependent variable. (CORRECT)

Correct: That’s right! These very specific hypotheses are generally structured as an if/then statement.

5. What are some of the limitations of Marketing Mix Modeling? (Select all that apply)

  • It isn’t very complex
  • It’s hard to show incremental effect if sales increases are small (CORRECT)
  • It requires collaboration between modelers and an econometric model (CORRECT)
  • It doesn’t infer causality, only correlation (CORRECT)

Correct: That’s right! This is a limitation of marketing mix modeling.

6. A test where you show two different versions of an ad and see which performs better is called a what?

  • Randomized Control Test
  • Independent Dependent Test
  • Two Type Test
  • A/B Test (CORRECT)

Correct: That’s right! This type of test can be helpful when you’re trying to optimize an ad.

PRACTICE QUIZ: RECOMMEND MEASUREMENT SOLUTIONS AND PERFORM AN ANALYSIS

1. What are some strengths of A/B testing in Marketing Analytics? (Select all that apply)

  • Allows for in-depth interviews
  • Finding incremental effects
  • Fast to conduct (CORRECT)
  • Good for testing advertising strategy (CORRECT)

Correct: That’s right! A/B testing is generally good for making quick, strategic decisions about an ad or campaign.

2. True or false: A brand lift test is a good way to evaluate changes in brand awareness.

  • True (CORRECT)
  • False

Correct: That’s right! Lift tests are an excellent way to find incremental changes and a brand lift test in particular is good for brand equity metrics like brand awareness.

3. If you need to combine data from multiple sources, you can use a…

  • Join clause in Java
  • Join clause in SQL (CORRECT)
  • Combination clause in Python

Correct: That’s right! A join clause in SQL can combine rows from two or more tables, based on a related column between them.

4. Jessie is using a join clause to see all the conversions they made in a one month period. To see all of the conversions from two tables for that one month period, what type of join should they use?

  • Right join
  • Full outer join (CORRECT)
  • Inner join

Correct: That’s right! This type of join gathers all the applicable data from two tables. As long as there was a purchase, Jessie would see the data in a new joint table.

5. What are some questions you should ask when comparing data from different sources? (Choose all that apply)

  • Does this data confirm my hypothesis? (CORRECT)
  • What time frame is used? (CORRECT)
  • What attribution setting is used? (CORRECT)
  • Which events are included and are they accurately measured and comparable?

Correct: That’s right! This is an important question to ask when comparing data from different sources.

6. True or false: The final step in the OSEMN process is interpretation.

  • False
  • True (CORRECT)

Correct: That’s right! The “N” in the OSEMN process stands for iNterpret and it’s the last part of the cycle.

GRADED QUIZ: MARKETING ANALYTICS IN ACTION

1. True or false: to formulate a business or marketing goal, keep the goal broad to capture more possibilities.

  • True
  • False (CORRECT)

Correct: That’s right! The first letter of the SMART acronym is specific. It’s important to clearly state what you’re hoping to achieve.

2. Blake plans to use likes as a KPI. Is this a good choice?

  • Yes
  • No. (CORRECT)

Correct: That’s right! Likes are a proxy metric. They do not have a direct effect on a business outcome.

3. What are some limitations of an A/B test? (Choose all that apply)

  • Very complex and time consuming
  • Might deliver biased outcomes
  • Reliable only if the confidence level is at least 75% (CORRECT)
  • Does not assess incremental impact (CORRECT)

Correct: That’s right! This is a limitation of an A/B test, but it can still be very useful for gathering data.

4. Blake is planning an experiment. Their hypothesis is that:

Ads targeted to women 30-55, interested in architecture, living in Cities in France will increase website visits.

What could help to make this hypothesis stronger?

  • A goal for this test
  • A time frame they are looking at the behavior (CORRECT)
  • A null hypothesis
  • A budget for the test

Correct: That’s right! Your goal should be specific, measurable, achievable, relevant and timebound.

5. What is the preferred method of multi touch attribution to get the most valid results?

  • Model-based attribution
  • Direct attribution
  • Data-driven attribution (CORRECT)
  • Last touch attribution

Correct: That’s right! This method is more expensive and can be challenging, but it’s likely to produce the best results.

6. Blake wants to start exploring their data. Which of the following is a good place to start?

  • Descriptive Statistics (CORRECT)
  • Visualizing The Data
  • Creating a Model

Correct: That’s right! Finding the data’s mean, median, mode etc. is a very good way to begin summarizing data.

7. When generating insights after an analysis, what are some questions you can ask yourself? (Choose all that apply)

  • What will this cost to incorporate into my analysis?
  • Any differences among audiences? (CORRECT)
  • How did I reach my goal? (CORRECT)
  • Did I reach my goal? (CORRECT)

Correct: That’s right! This is an important question to ask when determining insights from your research and analysis. This is a key insight to share with stakeholders!

8. What are some types of recommendations you can make after you have generated insights from your findings? (Choose all that apply)

  • Product Design Optimization
  • Planning and Forecasting (CORRECT)
  • Marketing Optimization (CORRECT)
  • Purchase Funnel Optimization (CORRECT)

Correct: That’s right! This is likely to be an area where your colleagues and stakeholders will seek your expertise.

9. When creating a presentation, what’s the main thing to keep in mind as you create it?

  • Make sure your presentation supports your narrative (CORRECT)
  • Crafting the best slide deck ever
  • How much time you have to present
  • How much money your campaign made

Correct: Correct! You always want to ensure that your presentation is helping you to tell your story.

10. How could you close your presentation?

  • Finish a story you started at the beginning
  • Answer a question you asked at the beginning
  • Add an interesting fact
  • All of the above (CORRECT)

Correct: Correct! You could end your presentation with all of the above.

11. Blake has a goal to get 1000 more conversions. Is this a SMART goal?

  • No.  (CORRECT)
  • Yes

Correct: That’s right! This goal is a good start, it’s specific, measurable, achievable, and relevant. However it’s not time bound which is an important part of setting a goal.

12. When tracking a goal, a key performance indicator should be: (Select all that apply)

  • Simple
  • Directional (CORRECT)
  • Directly Related (CORRECT)
  • Measurable (CORRECT)

Correct: That’s right! This is one of the ways to make sure your KPI is tracking progress towards your goal.

13. What are some limitations of observational methods? (Choose all that apply)

  • Very complex and time consuming
  • tests may not have sufficient statistical power
  • Might deliver biased outcomes (CORRECT)
  • Doesn’t take contextual variables into account (CORRECT)

Correct: That’s right! This is a limitation of observational methods, but it can still be very useful for finding insights about your customers.

14. Blake is planning an experiment. Their hypothesis is that:

Ads targeted to Cities in the US will increase website visits in August.

What could help to make this hypothesis stronger?

  • Data sources for the test
  • A budget for the test
  • An audience to be studied (CORRECT)
  • A proxy hypothesis

Correct: That’s right! Your goal should be specific, measurable, achievable, relevant and timebound.

15. True or false: it’s not very useful to compare your results against competitors.

  • True
  • False (CORRECT)

Correct: That’s right! It can be extremely helpful to compare your results against the competition and see where you’re doing well and where you can improve.

16. When creating a presentation, what should you focus on first?

  • Future plans
  • Goals, Objectives, and KPIs (CORRECT)
  • Data results
  • How you set up your campaign

Correct: Correct! Open your presentation with these things.

17. What might be a good tactic when thinking about opening and closing your presentation?

  • Open and close by stating your results
  • Connect the opening and closing with a story, a question, or an interesting fact (CORRECT)
  • Look at them as separate, unrelated parts of your presentation
  • Skip the opening and closing — they’re not needed

Correct: Correct! Try to connect them in some way.

18. When setting a SMART goal, what do the S and M stand for?

  • Simple and Measurable
  • Specific and Measurable (CORRECT)
  • Specific and Manageable
  • Simple and Manageable

Correct: That’s right! Your goal should be specific, measurable, achievable, relevant and timebound.

19. What are some limitations of a randomized control test? (Choose all that apply)

  • Does not assess incremental impact
  • Requires input to assign participants to groups
  • Tests may not have sufficient statistical power (CORRECT)
  • They can’t account for the unknown (CORRECT)

Correct: That’s right! This is a limitation of a RCT, but it can still be very useful for gathering data in an unbiased way.

20. Blake is planning an experiment. Their hypothesis is that:

Ads targeted to men 25-40, interested in dogs, will increase website visits in Q1.

What could help to make this hypothesis stronger?

  • A goal for this test
  • A location they are evaluating this behavior in (CORRECT)
  • Data sources for the test
  • An alternative hypothesis

Correct: That’s right! Your goal should be specific, measurable, achievable, relevant and timebound.

21. True or false: randomized control trials are designed to measure causality.

  • False
  • True (CORRECT)

Correct: That’s right! RCTs are carefully controlled to keep all variables constant, except for the treatment variable. This careful control allows you to infer whether or not an ad increased conversions.

22. When comparing data gathered from multiple sources, what are some questions you should ask yourself? (Choose all that apply)

  • Does this data prove my earlier theories?
  • What will this cost to incorporate into my analysis?
  • What time frame is used? (CORRECT)
  • What measurement methodology is used? (CORRECT)

Correct: That’s right! This is an important question to ask (and answer) when combining data from multiple sets.

23. Blake has found that people abandon the purchase process and they believe this has to do with asking too much information from the customer before they complete their purchase. What might be a recommendation?

  • Update the checkout process and adjust the number of fields that need to be filled out before completing the purchase. (CORRECT)
  • Make no recommendation as this is not directly related to a KPI.
  • Change the target audience to one that’s more amenable to the current checkout process.

Correct: That’s right! This is a recommendation based on an insight and would be very helpful to share.

24. When creating a presentation, what should you focus on last?

  • How you set your campaign up
  • The data results of your campaign
  • Ways you’ll iterate future campaigns (CORRECT)
  • Demonstrating causal relationships

Correct: Correct! Also end by stating how what you learned will affect the future.

25. True or false: a KPI should be directional to help track progress towards a specific goal.

  • True (CORRECT)
  • False

Correct: That’s right! A KPI should move up or down (think a daily view count for instance) to keep track of progress towards a goal.

26. True or false: when bringing data together from multiple sources, the first step is to make sure the data is comparable.

  • True (CORRECT)
  • False

Correct: That’s right! Having data from multiple sources isn’t very helpful if you can’t compare them to each other.

27. True or false: even though you’ve completed your analysis, it’s not a good idea to make recommendations related to the effectiveness of your marketing.

  • True
  • False (CORRECT)

Correct: That’s right! This is definitely something you can make a recommendation on after you’ve completed a thorough analysis.

28. How could you open your presentation?

  • With a story
  • With an interesting fact
  • With a question
  • All of the above (CORRECT)

Correct: Correct! You could open your presentation with all of the above.

29. Blake is attempting to measure the effect of an ad across many channels, online and offline. What is the most appropriate type of analysis?

  • A/B Testing
  • Marketing Mix Modeling (CORRECT)
  • Multi Touch Attribution

Correct: That’s right! This type of modeling is best suited for data coming in from all sorts of different channels even though it can be complex and time consuming.

30. Blake is generating insights and feels they have already done a good job evaluating the results against goals and KPIs. What can they attempt to do next?

  • Propose alternative hypotheses for later tests
  • Add perspective and context to the results (CORRECT)
  • Run a different analysis

Correct: That’s right! These insights can be helpful to compare results against the past and make projections for the future.

31. Which test should James use to prove to his boss that the Instagram campaign had an effect on the subscriptions?

  • An A/B test
  • A conversion lift test (CORRECT)
  • A brand lift test

Correct: Correct! A conversion lift test lets James prove the incremental effect of the campaign on subscriptions.

32. Why would you look at competitors and their performance to generate insights?

  • Competitive and industry information can add relevant context and put your results in perspective (CORRECT)
  • It is better not to compare yourself to the competition and just make sure you reach your goal.
  • It can help you fill in the blanks in your first-party data.

Correct: Correct! It helps to evaluate your results against the performance of your competitors and the industry as a whole so you can understand how you compare.

33. What kind of template should you use for your presentation?

  • One with my own favorite colors
  • A blank template
  • A template with branded colors and fonts (CORRECT)
  • Something cool from the templates list

Correct: Correct! A branded template shows professionalism, and alignment with your business’s branding.

34. What kind of data should you show?

  • All the data from our dashboards
  • Only data that shows positive returns
  • Data that supports the story you’re communicating (CORRECT)
  • No data — it’ll be distracting

Correct: Correct! Only show data that will be useful to your presentation.

35. What should you include in your presentation?

  • Goals, objectives, and KPIs
  • How you set up your campaign
  • What results you found, and plans for the future
  • All of the above (CORRECT)

Correct: Correct! You would include all of these in your presentation.

36. What are the advantages of practicing your presentation?

  • Get comfortable presenting
  • Determine your run time
  • Determine your movements
  • All of the above (CORRECT)

Correct: Not quite. This is only one advantage.

37. What is a simple story framework to guide you in your presentation?

  • The actions you took and the data you discovered
  • List all the data, explain the problem, show the actions
  • The problem, the actions, plans for the future
  • The problem, the actions, what you discovered, plans for the future (CORRECT)

Correct: Correct! This is a simple structure you can follow for your presentation.

38. What are some things you need to know about the space before you present?

  • Room set-up
  • How to load the presentation
  • How to advance slides
  • All of the above (CORRECT)

Correct: Correct! Make sure that you know these things — and more — before the day of your presentation.

Previous – Week 4
Home

Subscribe to our site

Get new content delivered directly to your inbox.

Quiztudy Top Courses

Popular in Coursera

Liking our content? Then, don’t forget to ad us to your BOOKMARKS so you can find us easily!

Subscribe