Thumbtack Analysis Project

This sheet will help guide you through the steps that you need to perform in order to complete the analysis of visitor requests and quotes on the Thumbtack dataset. However, be sure to refer to the separate instructions document for additional information.

Cells in light blue indicate task information and other hints, while darker blue cells indicate questions and tasks to answer.

Note: Over the course of the project, there will be fewer direct instructions. In other words, you’ll need to make more decisions for yourself in later questions!

Amber cells indicate where answers should be posted. Pay careful attention to whether you need to type out your answers or create visualization. Creation of additional sheets to perform your analysis is encouraged, but will not be graded.

Data can be found on the Visitors_Extd tab as a table named VisitorsExtd.

Visitors and Conversion Rate

How does the number of visitors change over time?

1 (a)

To help organize things, create a new Sheet, and name it VisitorsByDay.

1 (b)

Create a PivotTable on the VisitorsByDay sheet. Set up the PivotTable to show the number (count) of visitors on each day of the dataset. CHECK: You should see 2718 visitors on August 1st.

1 (c)

Create a line chart of the visitors by day. Copy / paste that chart into the space below.

1 (d)

Create another PivotTable on the VisitorsByDay sheet that shows the number of visitors by the day of the week. Then, create a line chart of visitors by day of the week and copy / paste the chart into the space below. NOTE: This will require creation of an additional column with day of the week information first!

1 (e)

Answer the main question: How does the number of visitors change over time? Briefly describe any patterns in the visitors you observe from the charts you created.

What is the “conversion rate”, i.e. the proportion of visitors that make a request? How does the category and device type affect the conversion rate?

2 (a)

Prepare this question by creating a new Sheet called ConversionRates.

2 (b)

Create a PivotTable on the ConversionRates sheet that computes the overall conversion rate. Type that value in below. NOTE: There are no autograded cells in this project!

2 (c)

Update the PivotTable to show conversion rate across category and device type. HINT: Place one feature on the rows and the other on the columns!

2 (d)

Create a bar chart that demonstrates the conversion rates by both category and device type. Copy / paste the chart into the space below.

2 (e)

Answer the main questions. What is the overall conversion rate? How is this affected by the request category? How is this affected by device type? Be sure to note the size of each feature’s effect in your response.

Quotes and Jobs

What is the distribution of the number of quotes received​ ​per​ ​request?

3 (a)

Create a new sheet, QuoteDist. On that sheet, create a PivotTable that tabulates the frequency counts of the number of quotes received by each request. CHECK: You should see 1924 requests with five quotes, and 1 request with six quotes.

3 (b)

Use the PivotTable to compute the average number of quotes received per request. Type the answer in below.

3 (c)

Update the PivotTable so that we see the percentage of requests that get each number of quotes. What is the proportion of requests that gets…

    … zero quotes?

    … multiple (two or more) quotes?

3 (d)

Create a chart that shows the distribution of the number of quotes received per request. Copy / paste that chart into the space below.

What factors contribute to certain requests getting more quotes than others?

4 (a)

Compute the average number of requests received by the category of request. Which type of request gets more quotes on average: cleaning or moving? What is that difference?

4 (b)

For Local Moving requests, is there an effect of the move distance on the number of quotes generated? Describe any patterns that you observe.

4 (c)

For House Cleaning requests, is there an effect of the number of bedrooms or number of bathrooms on the number of quotes generated? Describe any patterns that you observe.

4 (d)

Provide a recommendation to Thumbtack for at least one way that they might increase the number of quotes sent by pros to requests, with a focus on those requests that received only zero or one quote. Be sure to justify your recommendation using the data and what you found in your analyses!

4 (e)

Finally, create or select at least one chart from your analyses to support your recommendation in the previous part and copy / paste it below. If you want to select two or more charts, place one in the box below, then put any additional supporting charts to the right.

20201015041737thumbtackdata