A3. Advertising Content Audit – Individual Report [CLO1, CLO2, CLO3, CLO4]

Due: Sunday April 27^th, 11.55 PM / Weighting: 40% / Length: Max 2,000 words (+/- 10%)

Description:

This assessment provides you with the opportunity to undertake an advertising content audit to evaluate whether a company’s current advertising strategy is proper and to provide recommendations for further improvements.

You are asked to evaluate the consistency of beauty and cosmetics company Sephora’s Instagram (Australian) posts (https://www.instagram.com/sephoraaus) with its overall positioning objectives. In addition, by analyzing Instagram content and social media engagement data (e.g., the number of likes, comments), you can evaluate whether any sub-theme related to the company's positioning is more successful than other positioning sub-themes in generating engagement.

Details:

1) The focal company is Sephora which is a global beauty, cosmetics and personal care brand (we will focus on Sephora’s Australian operations - https://www.sephora.com.au). Identify the company’s positioning or value proposition statement (or a suitable summary) and provide it (with relevant citation) on the title page of your report.

Follow the steps below to complete your task:

2) Consistent Advertising with Positioning: We want to identify relevant themes across posted images. For this assignment, we will assume five relevant main themes which are similar (but not identical) with the main themes introduced in the Bodyshop case study (Lecture Week 8); Product, People, Landscapes, Other, About Positioning.

3) You have been provided with the “Sephora” instagram images, Apify metadata for the images and data extracted from Google Vision (label, multiple objects, ocr, logo).

Main Folder – Assignment 3_T1_25.doc  Link:

Images sub-folder – 3980 Sephora Instagram post images;

Extracted Vision – label_data.xlsx, multiple_objects_data.xlsx, ocr_data.xlsx, logo_data.xlsx

Apify metadata - Sephora_Apify_T1_25_ Ass3.csv

Initially, you will need to;

(i) Construct OpenCV data using the images (2.3 Image processing using OpenCV)

(ii) Process the captions in the Apify metadata file, to obtain Valence_overall and length_description and add them to Apify metadata file. (4.2 Text Analysis Variables – will help).

Since the relevant themes are somewhat consistent with the Bodyshop case, you have some base for constructing a dictionary for the main themes. You can add to this base or modify, by sorting detected features from Google Vision according to their usage frequency and adding any items which are specific to Sephora (possibly deleting items which are specific to the Bodyshop), if required. (This may be more relevant for the “About Positioning” theme since the positioning of the Bodyshop may vary from the positioning of Sephora) Using the constructed dictionary, determine which themes are present in the posted Sephora images.

(Do some preliminary googlevision processing with the multiple_objects, label, logo and ocr googlevision data – 5.2 Construct Variables from Googlevision will help)

From the preliminary googlevision processing, construct your data dictionary for the 5 key themes (use the bodyshop dictionary as a rough guide) which you will then use in the 8.1 Main Category_Variables_googlevision notebook.

4) Report how many posts (frequency and percentage) are related to each theme. Based on usage frequency, discuss the company’s advertising content strategy. In addition, evaluate whether you feel the frequency of the positioning-related theme is proper. Why? Or why not?

5) Advertising Effectiveness: (Evaluate whether the positioning-related theme is more successful compared to other popular themes in enhancing social media engagement by doing the following tasks):

Construct the regression data set which contains OpenCV, Apify metadata and theme categories as dummy variables (also include relevant control variables: (1) text length in the caption, (2) text sentiment in the caption, (3) OCR text length within an image, (4) OCR text sentiment within an image, and (5) Posting time dummies: Year, Month-of-Year (January, …, December), Day-of-Week (Monday, ..., Sunday), Time-of-day (Morning, Afternoon, Evening, Night).

Report summary statistics (count/frequency, mean, median, minimum, maximum) in a Table for each of the X and Y (like count, comment count) variables.

For Y equal to “like count” use normal OLS regression (considering outliers and multicollinearity) on either like count or Ln (like count+ 1) or both. Report the result of your model (s) in a Table (columns: X variables, coefficients, p-value, and VIF- Highlight the significant variables). Interpret the results focusing on the significance and relative importance of the separate themes in impacting viewer engagement.

For Y equal to “comment count” use normal OLS, Negative Binomial and Poisson regression. (Don’t worry if the NB and/or Poisson do not converge –some additional code may be needed)

Explain your choice of regression models. For “comment count” report the regression results of the best model (out of the OLS, Poisson and Negative Binomial if results are available) in a Table (columns: X variables, coefficients, p-value, and VIF).  Interpret the results focusing on the significance and relative importance of the separate themes in impacting viewer engagement.

Effective Ad Content Exploration: Now, you are required to break down the positioning-related theme into several (between 3- 5) sub-categories or sub-themes (See the examples in the class materials). For example, there may be sub-themes based on “Organic” or “Inclusion” which represent how the organization wants consumers to view the brand. You can refer to the mission statement and your overview of the organization’s positioning to provide potential sub-categories (sub-themes). Provide your rationale for this sub-categorization (ensure it is specific to Sephora and its positioning).

Report on the number of posts, and summary statistics for each sub-category.

For Y equal to “like count” use normal OLS regression (considering outliers and multicollinearity) on either like count or Ln (like count+ 1) or both.

Report the result of your model (s) in a Table (columns: X variables, coefficients, p-value, and VIF). Interpret the results focusing on the significance and relative importance of the separate positioning sub-themes in impacting viewer engagement.

For Y equal to “comment count” use normal OLS, Poisson and Negative Binomial regression (considering outliers and multicollinearity). Explain your choice of candidate regression models. For comment count report the regression result of the best model in a Table (columns: X variables, coefficients, p-value, and VIF).  Interpret the results focusing on the significance and relative importance of the separate positioning sub-themes in impacting viewer engagement.

Construct your data dictionary for the 3 to 5 sub-positioning themes (use the Positioning dictionary as a base and sub-divide into 3 to 5 sub-categories) which you will then use in 8.2 Sub Category_Variables_googlevision notebook.

Merge all the data (OpenCV, Apify metadata, Main category data (from 8.1), Sub-category data (from 8.2) into one regression ready file (8.3 Merge_All_Variables_Regression_data)

Use the regression file which will come from 8.3 to do the various regressions.  Use 8.4 Regression_models with main & sub-category variables (T1_25) to conduct the regressions. (You may need to run this notebook several times (modifying the code) to obtain the required regressions for each part)

6) Conclusion: Provide conclusions on which main themes and sub-themes the company seems to be focused on in their social media posts. Is their advertising content in their social media posts consistent with the company’s positioning? Which positioning (sub) themes are successful? Considering the above, recommend what the company needs to keep doing or improvements it needs to make to enhance engagement and consistency with positioning.

In completing this task, apply appropriate data analytics and consider the concepts introduced in class. Your report should not exceed the word limit, excluding the title page, relevant images, tables or charts.

Title page (1 page) includes (1) Company & Positioning statement, (2) Word count, (3) An executive summary (One paragraph) of your report, (4) Course name, tutorial session and tutor’s name, (5) Your first and last name & zID

Reference: (If any) Cite academic papers, newspaper articles, blogs, or industry reports using Endnote. Use APA (American Psychological Association) style. in-text citations and a reference list at the end. https://student.unsw.edu.au/apa

Format: Use word file (.doc), 12pt, 1.5 lines spacing, at least 2.5cm margins on all sides.

Submission instructions

A. Submit your report to the Moodle Turnitin Report submission box

B. Use the File name format: Tutorialsessioncode_your first and last name_ zID_A1.doc”

(e.g., M15A_Your_Name_zXXXXX_A1.doc).

C. Submit other supporting files (data, image, paper and code) to the Moodle Supporting Files - submission box. (1 mark deducted for each of 1) and 2) missing)

1) .ipynb contains all the relevant code to get the results in your report. Make a zip file by combining all colab/python files.

2) .xlsx file contains the datasets on which you run the regression.

3) You can also provide a document with a link to a shared G-drive folder containing all your relevant files (ensure the link works)

Marking Criteria

Your assignment will be marked based on the following marking criteria:

1. Analysis: Quality of advertising image data analytics

2. Evaluation and Recommendations: Quality of evaluation and recommendations

3. Written Presentation: Quality of written report

For further information, see the below marking rubric.

Marking Rubric for Assessment 3: Advertising Content Audit – Individual Report