Difference between revisions of "ANLY482 AY2017-18 T2 Group15 Reflections"
Jump to navigation
Jump to search
Analytics Practicum was a course which allowed us to experience working with real data, with a real business, which has real business problems. It was an invaluable experience which brought together all we have learnt in the course of our study in the School of Information Systems.
(5 intermediate revisions by the same user not shown) | |||
Line 16: | Line 16: | ||
| style="padding:0.3em; font-family:Arimo; font-size:110%; border-bottom:4px solid #aaaaaa; border-top:4px solid #aaaaaa; background:#d8d8d8; text-align:center;" width="10%" | | | style="padding:0.3em; font-family:Arimo; font-size:110%; border-bottom:4px solid #aaaaaa; border-top:4px solid #aaaaaa; background:#d8d8d8; text-align:center;" width="10%" | | ||
− | [[ANLY482_AY2017- | + | [[ANLY482_AY2017-18_T2_Group15_Recommendations|<font face ="Helvetica" color="#3f3d3d"><strong> RECOMMENDATIONS</strong></font>]] |
| style="padding:0.3em; font-family:Helvetica; font-size:110%; border-bottom:4px solid #aaaaaa; border-top:4px solid #aaaaaa; background:#ffcc16; text-align:center;" width="10%" | | | style="padding:0.3em; font-family:Helvetica; font-size:110%; border-bottom:4px solid #aaaaaa; border-top:4px solid #aaaaaa; background:#ffcc16; text-align:center;" width="10%" | | ||
Line 25: | Line 25: | ||
|} | |} | ||
+ | |||
+ | <br/> | ||
+ | |||
+ | <div align="Left"> | ||
+ | <div style="background: #ffebba; padding: 12px;font-family:Helvetica; font-size: 15px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #ffcc16 solid 10px; border-right: #ffcc16 solid 10px; text-align:center "><font color="#3f3d3d">OUR RECOMMENDATIONS</font></div> | ||
+ | <br/> | ||
+ | ::This section will detail our recommendations for our client, as well as our plans for the following stages of this project and beyond. | ||
+ | |||
+ | ::'''a. Data Extraction and Preparation''' | ||
+ | ::::For future data processes, we strongly recommend our client to export and process their data in Relational form as mentioned, which will improve efficiency in data visualization and analytics on all platforms. | ||
+ | ::::In addition, we will provide our client with templates outlining the correct data format(s), as well as data process documentation for them to easily replicate our work. | ||
+ | ::::We would like to make data transformation as automated and seamless for our client as possible, but we are limited by the duration and scope of the project. Moving forward, our client may want to consider using faster methods of multiple transposition of data (such as OpenXlsx with R, “Stacks” with JMP) to speed up data cleaning and transformation. The automation of data cleaning may be commissioned to the next batch of ANLY482 students, possibly to a team adept in R, which may have data cleaning capabilities our client will find useful. | ||
+ | ::::We hope our client uses the results of this study as a SOP and create a data connection directly from the data warehouse to the visualization software for greater efficiency in updating their visualizations. | ||
+ | |||
+ | ::'''b. Cost Concerns''' | ||
+ | ::::Tableau is a costlier option to our client; license purchase is needed, unlike Qlik which is already purchased and implemented by our client’s overseas counterparts. If Tableau proves to be an unfeasible option, we recommend our client to use our Excel Pivot solution temporarily while the staffs familiarize themselves with Qlik. | ||
+ | |||
+ | ::'''c. Challenges Encountered''' | ||
+ | ::::This subsection will detail the challenges encountered in this study. | ||
+ | |||
+ | :::::1. Data Collection | ||
+ | ::::::This is a challenge faced by the client. For certain components of data, members of the sales department must manually input sales data into iPads. The client has mentioned that these iPads often crash, causing data to be lost. In addition, this manual input method is error-prone. | ||
+ | :::::2. Data Cleaning | ||
+ | ::::::For our client, the current method of data cleaning involves manually retrieving data by copy-pasting, which is both time-consuming and error-prone. | ||
+ | :::::3. Data Visualization | ||
+ | ::::::As we have limited experience with visualization software, more time and effort, as well as research, was needed to create the visualizations we wanted, and to optimize how they were displayed. This challenge will also apply to our client who are first-time users. However, guidance was readily available online and the time taken to create new visualizations decreased drastically with each one created. If this becomes an issue, we would recommend our client to hire new staff specializing in analytics | ||
+ | |||
+ | <br/> | ||
+ | |||
+ | <div align="Left"> | ||
+ | <div style="background: #ffebba; padding: 12px;font-family:Helvetica; font-size: 15px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #ffcc16 solid 10px; border-right: #ffcc16 solid 10px; text-align:center "><font color="#3f3d3d">OUR REFLECTIONS</font></div> | ||
+ | <br/> | ||
+ | <center>Analytics Practicum was a course which allowed us to experience working with real data, with a real business, which has real business problems. It was an invaluable experience which brought together all we have learnt in the course of our study in the School of Information Systems.</center> |
Latest revision as of 16:03, 15 April 2018
OUR RECOMMENDATIONS
- This section will detail our recommendations for our client, as well as our plans for the following stages of this project and beyond.
- a. Data Extraction and Preparation
- For future data processes, we strongly recommend our client to export and process their data in Relational form as mentioned, which will improve efficiency in data visualization and analytics on all platforms.
- In addition, we will provide our client with templates outlining the correct data format(s), as well as data process documentation for them to easily replicate our work.
- We would like to make data transformation as automated and seamless for our client as possible, but we are limited by the duration and scope of the project. Moving forward, our client may want to consider using faster methods of multiple transposition of data (such as OpenXlsx with R, “Stacks” with JMP) to speed up data cleaning and transformation. The automation of data cleaning may be commissioned to the next batch of ANLY482 students, possibly to a team adept in R, which may have data cleaning capabilities our client will find useful.
- We hope our client uses the results of this study as a SOP and create a data connection directly from the data warehouse to the visualization software for greater efficiency in updating their visualizations.
- a. Data Extraction and Preparation
- b. Cost Concerns
- Tableau is a costlier option to our client; license purchase is needed, unlike Qlik which is already purchased and implemented by our client’s overseas counterparts. If Tableau proves to be an unfeasible option, we recommend our client to use our Excel Pivot solution temporarily while the staffs familiarize themselves with Qlik.
- b. Cost Concerns
- c. Challenges Encountered
- This subsection will detail the challenges encountered in this study.
- c. Challenges Encountered
- 1. Data Collection
- This is a challenge faced by the client. For certain components of data, members of the sales department must manually input sales data into iPads. The client has mentioned that these iPads often crash, causing data to be lost. In addition, this manual input method is error-prone.
- 2. Data Cleaning
- For our client, the current method of data cleaning involves manually retrieving data by copy-pasting, which is both time-consuming and error-prone.
- 3. Data Visualization
- As we have limited experience with visualization software, more time and effort, as well as research, was needed to create the visualizations we wanted, and to optimize how they were displayed. This challenge will also apply to our client who are first-time users. However, guidance was readily available online and the time taken to create new visualizations decreased drastically with each one created. If this becomes an issue, we would recommend our client to hire new staff specializing in analytics
- 1. Data Collection
OUR REFLECTIONS