centercenterHeather Tymms | BSc (Hons) Mathematics Student

centercenterHeather Tymms | BSc (Hons) Mathematics Student

Sainsburys Supermarkets Ltd | 33 Holborn, Ec1n 2htPlacement Year Report 2020/2021

analytics intern

8820090900Heather Tymms | BSc (Hons) Mathematics Student

Sainsburys Supermarkets Ltd | 33 Holborn, Ec1n 2htPlacement Year Report 2020/2021

analytics intern

OverviewFrom June 2020 to July 2021, I worked at Sainsburys Supermarkets Ltd as an Industrial Placement Employee (IP) for a 13-month placement year. The Office is located in Holborn, London but I was mostly working remotely due to the COVID pandemic. I worked in the Sainsburys Tech Department as firstly a Retail Experience Analyst and later as a Data Visualisations Analyst.

I worked on several high priority projects for the business. As an IP I viewed the restructure process during March and May. This gave me insight into the company structure. I got exposure to the Senior Leadership Team often as they enjoy feedback from IPs to improve the IP schemes in the future.

I developed my knowledge and skills in SQL, Python and Microsoft Applications (Excel/ PowerPoint/ Word) and I developed my softer skills too. My presenting knowledge and skills have increased a lot due to presenting my findings and insights every two weeks. I also offered to co-host a section of a Sainsburys Tech wide training day in the LOVE DATA conference.

My time with Sainsburys has been invaluable, not just because of the interesting workload to increase my skills from experience but the team I have been working with is so supportive and knowledgeable. Sainsburys is very colleague driven, it strives to be a Great place to work and shop and in my time here, they are living up to their values.

Table of Contents

TOC o "1-3" h z u Overview PAGEREF _Toc77670601 h 1Structure of the Business and Description of Work PAGEREF _Toc77670602 h 2Overview of Projects PAGEREF _Toc77670603 h 3Retail Experience Projects PAGEREF _Toc77670604 h 3Extra Activities PAGEREF _Toc77670605 h 4Data Visualisations Projects PAGEREF _Toc77670606 h 5In Depth Projects PAGEREF _Toc77670607 h 6Till Upgrade Should we upgrade our self-checkout tills? PAGEREF _Toc77670608 h 6SmartShop Connectivity Is Wi-Fi an issue when using the SmartShop Mobile App? PAGEREF _Toc77670609 h 8Simulation Trial Does this Simulation work for our larger stores? PAGEREF _Toc77670610 h 9Training in MSTR What does Analytics need to know to use MSTR competently? PAGEREF _Toc77670611 h 10Storytelling Game Plan How can we improve Storytelling in ADA? PAGEREF _Toc77670612 h 12Development Time PAGEREF _Toc77670613 h 12Aims and Objectives PAGEREF _Toc77670614 h 14Overall PAGEREF _Toc77670615 h 16Appendix PAGEREF _Toc77670616 h 18Appendix 1 Storytelling Game Plan PAGEREF _Toc77670617 h 18Appendix 2 Simulation Launch PDF PAGEREF _Toc77670618 h 19Appendix 3 September Link Conversation Feedback PAGEREF _Toc77670619 h 20Appendix 4 February Link Conversation Feedback PAGEREF _Toc77670620 h 21Bibliography PAGEREF _Toc77670621 h 24

Structure of the Business and Description of WorkSainsburys is a huge business according to the University of Hertfordshire Placement Pack. It is bigger than I perceived before I joined. I embarked on my 13-month placement at Sainsburys in June 2020 as an IP along with 12 other IPs. We all went through induction together and after 2 weeks, we got split up into different parts of the business. We had a 12-week program to train and prepare for working at Sainsburys alongside shadowing our team to understand how we work in our areas. I was one of a few who got placed into Applied Data and Analytics (ADA) team. Within ADA, there are 3 teams: The Analytics Team, The Data Visualisation and Automation Team and The Data Science Team. I was in a Squad in the ADA Analytics Team that dealt with Retail Experience Analytics. Figure 1 is a flow chart showing where the Retail Experience Team is placed in the business.

Figure 1: Flow Chart for Organisation Hierarchy

As a Retail Experience Analyst, I worked on how to improve customer experience and efficiency in stores around the Sainsburys estate. During my time there, we increased our workstream and became a multiband team where we provided insights to improve our operational performance in both Sainsburys and Argos stores.

Our team works with stakeholders from a variety of places around the business that look after processes like front-end (shop floor processes) and SmartShop logistics. They ask us questions like Should we upgrade our self-checkout tills? and we look at the data to find insight so we may answer these questions and give recommendations to show the best way we can improve the business.

We split out our project into 3 sections: Scoping, Analysis, and Documentation (Figure 2). During scoping, we find the data sources required (mainly using SQL and Sainsburys knowledge sharing area), understand the tables and the question given to us. We then move onto the Analysis part, which entails analysing the data, making graphs and visualisations, and sometimes building models to find insights. We then move ono the Documentation part where we give a presentation for our stakeholders to gain relevant information from the piece of work, collaboratively QA each others code/outputs and write up our piece of work onto the Sainsburys secure area. Lastly, we present our work back to stakeholders and colleagues then if necessary, schedule a meeting with the stakeholders to answer any remaining questions.

Figure 2: Flow Chart of project progression

I stayed in Retail Experience Analytics team for 11 months. After this, I was fortunate enough to gain experience in the Data Visualisations Team that took care of Sainsburys use of an analytical platform called MicroStrategy (MSTR). I was there for roughly a month and then gained some exposure of another Data Visualisation Team which took care of Digital Reporting on the MSTR platform. Data Visualisations had a longer time frame to complete projects so was slower paced than Analytics was.

Overview of ProjectsDuring my time in the Retail Experience Analytics Team, I was part of some interesting projects. Our task was to find insight from data to improve our customers instore experience, so we mainly focus on checkouts opening hours and facilities we provide in our Sainsburys and Argos stores.

As an analyst in ADA, we get a project brief from the Product Manager, we then speak with our stakeholders to make sure what they are asking for, we do the analysis, and we present our work in a Demo and we schedule another meeting with the stakeholders to ensure our insights help drive impact in the business.

In my role, I was helping with some project at the start and as time went on, I did some independent projects for the business too. This consisted of tasks such as: collecting data from a wide dataset using SQL, understanding project briefs, and using analytical techniques including correlation analysis, pre-post analysis and trial-control analysis.

In the following pages I describe the projects I have worked on in my time at Sainsburys Tech. There are three main area: Retail Experience, Data Visualisations and Extra-curricular activities.

Retail Experience ProjectsReplenishment How can we optimise the replenishment (shelving) process?

While working collaboratively with several team members and other squads, our aim was to track a trial to see if a new Data Science built tool was effective. I was tasked to write several sections of code to pull data that would track the performance of stores and to check the compliance of stores (did stores stick to the schedule?). I created and updated an excel dashboard to track the performance of stores and distributed it around to the relevant people, stakeholders, Data Scientists, and fellow analysts every week for over two months.

Till Upgrades Should we upgrade our self-checkouts?

While working with my line manager, Rachel McNally, we answered a question should we upgrade our self-checkout tills? We brainstormed our ideas and made plans for what each of us were doing. I investigated if the new models were faster than the older models and then checked if the older ones had more interventions from engineers than the newer models. We then presented this back to the stakeholders and colleagues. This is described in depth in the next section.

Till Mix Does till participation have a relationship with store attributes?

I was a part of creating a model to find out how many tills should go into Argos stores and what type of tills. Before a colleague could make a clustering model, I was tasked to find out if there were any relationships between till participation and other store attributes. I was using correlation analysis in Python and using the Pearsons Correlation Coefficient method to find any strong relationships between till participation and various store attributes. With the information I got, we were able to make an informed decision on what the cluster on and how. This piece of work increased my correlation analysis skills and visualisations skills.

ARI Debugging What is making this alert fire unnecessarily?

I was tasked to find and resolve an issue with an automated alert. This code was written in both SQL and Python. I had to understand the code and find the issue within a timeframe. I found he issue and made changes to the master code that was saved on GitHub. This task improved my knowledge of automated Python scripts and increased my understanding of GitHub.

SmartShop Connectivity Is there an issue using the Wi-Fi instore on SmartShop Mobile?

My colleague and I were tasked to use a raw data table to find whether Wi-Fi connectivity was a problem with the uptake of the SmartShop Mobile App. After my colleague cleaned the table, I reviewed his SQL code using the GitHub processes then I analysed the data provided and made visualisations in Excel to show the insights to stakeholders. This piece of work is described in more detail below.

Simulation Trial How can we optimise the manned tills effectively?

An algorithm was constructed by my colleagues. While they were finalising the algorithm, I helped list the key performance indicators (KPIs) we should track to define a successful trial. I then made a MicroStrategy (MSTR) dashboard to visualise the performance metrics to distribute the outputs to stakeholders. I wrote a SQL code that would update every day and MSTR dashboard would be automatically updated from this piece of SQL code. This piece of work is described more in the below section.

CSAT Review What is the customers perception of Checkout Ease and Checkout Speed?

I was tasked to find out how customer satisfaction (CSAT) had changed in the past two years and to see if the two survey questions related to Checkouts were perceived different or the same by the customers. I wrote a SQL code to collect the data, transformed it and made visualisations in Excel. I presented it back in a Demo. This increased my curiosity for data and looking for insights.

Closing times Is there a benefit to close some Convenience stores an hour early?

I was tasked to make a tool to easily show if Convenience stores were profitable enough in their last hour of trading or if it was more beneficial to close them an hour earlier than usual. For this piece of work, I found the necessary information about profitability in the Sainsburys business and found the data set required to analyse. I then made a responsive Excel Worksheet that flagged certain stores (one store per row) when it was under the inputted threshold in three different time periods. This piece of work was taken to the Retail Leadership team the week after the project was finished.

SmartShop Lapsing Customers What do lapsed customers do after using SmartShop?

I was tasked to create a Key Performance Indicator (KPI) Tree showing the participation of SmartShop transactions, customer types and where the customers go after they stop using SmartShop for over a month. I collected the data, checked other sources to make sure my output was reliable. I then made visualisations and presented this back to the stakeholders in a demo with a corresponding email after.

Extra ActivitiesIn Sainsburys, we are very wellbeing aware, this was evident in the number of activities around wellness and wellbeing there were during my time at Sainsburys. My line manager was a Wellbeing Champion in Sainsburys Tech and I asked if I could get involved with some activities she was running.

I was also a part of an initiative in ADA, called the Game Plan, which aims to improve aspects of work. There were 6 areas of the Game Plan, I was a part of the Storytelling game plan group where our aim was to increase our knowledge and awareness of Storytelling through our work in the ADA Analytics Team. From this, I helped in making a document consisting of ways to improve Storytelling. I was then part of upskilling the rest of the Analytics team. I made the contents for three sessions within ADA to improve storytelling.

These are the talks I have given that were extracurricular:

Burn-Out December

I lead a 45-minute discussion on the definition, warning signs and possible initiatives to avoid burning out. This was received well and had positive feedback. This talk was presented in December 2020 because it is a high burn out time for many people. I increased my research skills and presenting skills.

Accessibility April

I co-hosted an interactive discussion about Accessibility while working from home with aspects of how it can improve Storytelling too. We talked through the importance of the topic and had an activity in which the audience in groups of 3 should change a badly formatted PowerPoint slide with a graph and make the information more accessible. This was well received, and people have started to use the advice in their work.

Importance of Sleep June

I co-hosted a talk on the Importance of Sleep. This was with my line manager, a Wellbeing Champion. I volunteered to co-host this talk because I am fascinated and passionate about the subject and wanted to let others know how important sleep genuinely is. This conversation was about the science of sleep and what sleep can affect in our bodies. It also had tips to improve your sleeping pattern. There was a lot of engagement during this meeting, and it had positive feedback too.

Data Visualisations ProjectsAfter 11 months of being in the Retail Experience Analytics Team, I gained some experience in Data Visualisations Analytics team in my last months on my placement. In this time, I helped with finalising the Training Syllabus of MicroStrategy for the other ADA teams. After moving to a different Data Visualisations team, I investigated the migration from one data source to another for our reporting in Sainsburys. This is where I was trying to improve the process of a report.

Syllabus What should we include in our MSTR training programme?

I helped cut down a pre-planned syllabus and ensured only the necessary information that Analytics would need to know to understand and use MicroStrategy competently would remain. I then created a template of questions that would find out how the Analytics team work currently/what tools they used and why. I then set up meetings with over 10 people around the Analytics Team and spoke to them about the pre-determined questions. I finalised the modules and learning outcomes for the internal training course for Analytics to undertake. This piece of work helped hone my communication skills within the working environment. Also, it increased my knowledge of how the business works internally. I gained an insight into how to best set up a training program for a technical role.

MicroStrategy PowerPoint use case What will MSTR look like when making a PowerPoint?

I created a previous Analytics piece of work in MicroStrategy to showcase new features on MicroStrategy and a way to use the platform for analytics work. I made a dashboard and ensured the data was correct. I then formatted it into a way that would mimic a PowerPoint presentation so we could showcase the opportunities to use MicroStrategy. Finally, I made a similar piece of work but in one page that could scroll to showcase the scrollable feature and show another way we can share outputs with our stakeholders. I then presented this work to the ADA managers in July 2021 where I had lots of positive feedback and some interested Analytic Managers to start using MSTR within their work. This piece of work increased my knowledge on the platform MicroStrategy. It also gave me ideas into how best to visualise various types of data

Digital Data Visualisations How can we improve our processes?

During my time in the Digital Reporting team, I was tasked to improve a manual process in the SQL database to automate it. I used my previous SQL knowledge to amend and append some rows in a table to a new table. This process saved downloading and uploading a heavy CSV (comma separated variable document that could be opened in Excel). This task improved my knowledge on automation and problem-solving skills to improve processes in pieces of work.

They also were using a long process to upload CSVs from a machine to the SQL database with several steps in between. I created a new way to smooth this process. I created a Jupyter Notebook containing code that would upload a CSV from your machine to the SQL database in one document. I then created a SQL code to add the new information a productionised table. This task improved my descriptions of a technical process to others. It has also provided me with building a legacy document that can be used many times by different people for their own purpose.

In Depth ProjectsIn this section, I will go through 5 main projects I was assisting with:

Till Upgrade

SmartShop Connectivity

Simulation Work

Training on MicroStrategy

Game Plan Storytelling

Till Upgrade Should we upgrade our self-checkout tills?Within my time at Sainsburys, I worked on a project that had the potential to benefit the customer experience on the self-checkouts. Across the estate, we had over 3 different models of tills that were of different ages, between very new and over a decade old. A new model from a third party had been launched and Sainsburys had been rolling the upgrade through the stores but wanted to make sure it was worth rolling out to more stores and if so, they wanted to know which stores/tills were prioritised.

For this project, we had to take into consideration:

The amount of money we would spend to upgrade.

The amount of money we already spent so far.

The differences and benefits of upgrading to the new models.

From the question, we thought about what would affect the tills most. We then divided our workload accordingly. I spend 2 sprints (4 weeks) on this project, one sprint dedicating to investigate whether the new models were faster at transactions than the older versions and the other sprint on whether the older models required more engineering visits. We then thought about how long our period of analysis should be. We chose the 20 most recent weeks because these should not be highly impacted during COVID.

We first found the data sources, which werent all in an easy, accessible format. Due to this, we had to use several applications including Excel, Python and SQL to make it accessible to use in our work. This took a while I helped this process by using a previous documentation for uploading CSVs (a type of saved document) via Python to a SQL database. After this process, we were able to carry on with the work.

For the first sprint, the question was Are the new tills faster than the older tills?. For this, we found most of the data sources that were already accessible in a SQL database. My colleague uploaded the rest of it onto the database. I amended a previous analysts code to suit the new use for it. The outputted table consisted of rows of transactions, specific model information and how long each transaction took.

On the same SQL database, I then explored the table created to find top level statistical values, like mean time of transaction, standard deviation, and variance to familiarise myself with the table.

From the base table, I used Excel to group the like models together and saw if the mean time for a transaction was the same or different. I found the newer versions were faster. Figure 3 shows the graph I created in Excel to share with the stakeholders.

147510511553600

Figure 3: Bar graph showing new version usually are quicker per item

The second sprint was spent working on the question Do older versions require more engineer interventions?. This data was not readily available in the SQL database so in the first sprint, we spoke to a few people and found an Excel report which showed the number of engineer interventions were required around the estate. My colleague had a piece of Python code that would upload this information onto the SQL database, so I used this to include this information in the analysis. Now that the data was accessible, I made a SQL code to output the relevant data: a table that included model version and number of engineering interventions for that 20-week period.

Figure 4 is the Excel graph that we took to the stakeholders. We found that the older versions had more interventions with engineers. While I was doing this task, my colleague made a table of specific tills to upgrade first, defining down to model and store. We shared this with stakeholders, and they started rolling out the proposal soon after.

Figure 4: Bar Graph showing percentage of tills around the estate that require an intervention

I then documented this piece of work on GitHub using the command line to talk between the computer itself and GitHub. I wrote this project up on the Sainsburys web documentation site so that this project can be easily understood and replicated by other analysts in Sainsburys in the future. This is a necessary part of any project because we would like our work to be easily replicated in the future and we used other analysts work for this and it was invaluable to understand what she did previously with the business context.

From this piece of work, I increased my knowledge on constructing SQL code, reading in files into python and uploading them into the SQL platform. I also had more exposure of the business itself, I had to talk to several different people around the business to get the correct data before the analysis could happen. I also had several meetings with the stakeholders to keep them informed and to share our analysis.

I got more comfortable in the process that analytics use, I found it hard to adjust to not having instructions on an open question like this project, but I understood the process a lot better and found it easier from this point onwards. The process of thinking about what could influence the upgrade, what we want as a business and what type of data we would need to analyse and how. I gained the perspective of curiousness and thinking in a way that a stakeholder would.

SmartShop Connectivity Is Wi-Fi an issue when using the SmartShop Mobile App?In Sainsburys, we have 3 types of checkouts: manned ills, self-checkouts and SmartShop. SmartShop is a form of self-checkout where you scan your items as you walk around the shop and then you can pay at a till (or on your mobile in some stores) and that is the transaction complete. There are two types of SmartShop currently, a Sainsburys SmartShop handset or a mobile device app. SmartShop has grown in the past few years, including during the COVID pandemic but the business was concerned about the use of Wi-Fi in stores. Previously they thought people using their mobile app required a Wi-Fi connection due to each superstore being large with solid construction. They also thought that their third-party Wi-Fi didnt have much traffic going through. So, they asked us to answer the question Is there an issue with connecting/reconnecting to Wi-Fi in our Sainsburys stores while using the SmartShop Mobile App?.

We were using a relatively new dataset for this piece of work. My colleague and I had a lot of data validation to do before we progressed onto our analysis stage. We made a few SmartShop mobile app transactions to validate it and there were lots of correspondence with the data engineering teams to clean and validate the data. We amended the issues after 3 weeks of reconciliation and started our analysis.

My colleague made the necessary base table from this new data set and I reviewed his work using the GitHub pull process and gave some feedback while QAing the piece of code. From the base table I made smaller datasets and analysed key areas that would define an issue with the SmartShop App Wi-Fi issues.

The areas I focussed on were: Proportion of people using handset to mobile app, proportion of completed shops on each type of Mobile Connection, proportion of rescans that a customer did on each type of connection and the differences in device type (Android/iOS). Figure 5 are two of the slides from the presentation we gave to the stakeholders. I made the slides using icons and shapes for a more engaging experience.

Figure 5: Sample slides in presentation for stakeholders

After the analysis, we shared our insights and recommendations to the stakeholders, and they used this information to progress talks with the store Wi-Fi provider. It was a new dataset, so the stakeholders were surprised about what we presented and recommended. We saw more people on Mobile Network without any connection failures and the same proportion of completion rate for both Wi-Fi connection and Mobile Network connection. This was significant because we found a different area to scrutinise instead of the bandwidth/strength of the connection it was more likely to be customers awareness of the free instore Wi-Fi or the log in process was too long/irritating to complete.

After we spoke to the stakeholders, I documented this work on GitHub and the Sainsburys knowledge dictionary so that if anyone else wanted to do a similar project in the future then it would be accessible and useful. A few months after the project, the new table was ready for everyone in Analytics to use so I was helping other colleagues understand the table and use it effectively.

From this project, I found a fascination for problem solving and curiosity with the data. This project was interesting because I was one of the first to use this table and it was eye opening to look at how data engineers understood and moved data and their processes. It was the first time I had used such granular data and it was a great experience.

Simulation Trial Does this Simulation work for our larger stores?In our team, we were working with several different teams around the business to see if a new and innovative way we can optimise our manned tills (mainbanks). This idea was proposed by another analyst in the team over 2 years ago. The initial simulation was being built within the team before it got automated with the engineers. Prior to productionising the tool, a trial was proposed to make sure this simulation would benefit the business. I came in to join the two team members that were making the simulation to help with the trial analysis.

I was a part of many meetings and correspondence with the stakeholders for this project. It was vital to work closely with the stakeholders because it was a proof-of-concept (PoC) piece of work that is new to the business. I emailed the lead stakeholder regularly and once the trial had begun, I sent over a launch guide in PDF form, which is provided in the Appendix.

I was tasked to identify the metrics we should track during the trial so we can gage the performance of the trial. The metrics for this trial that were identified are in the launch document in the Appendix. After identifying the metrics required for the trial tracker, I then wrote an SQL code to collate the key metrics with historic data as well as current data from stores in the trial. My colleague found the control stores using clustering analysis so I could add the control stores into the code.

I constructed a dashboard in MicroStrategy (MSTR) to track the trial automatically. This dashboard consisted of a chapter to look at all trial and control stores and another chapter consisting of interactive filters where you can choose a specific trial-control pair. This was the best way to show the performance of the trial because we can look at the whole trial and specific stores that would be compliant. Compliance in this context means the stores are opening the number of tills we advised for each hour of the day. While constructing this dashboard, I kept in mind how this tool was going to be used, bearing in mind the usability of the dashboard in a stakeholder perspective.

After this dashboard was constructed, we needed to distribute it to stakeholders and store managers. Using MicroStrategy in the team for the first time meant we had no prior process to distribute it because it was on a secure area. We corresponded with many teams, within both ADA-Data Visualisations and Automation team and Data Engineering teams. After a few weeks we finally got the ability to send this out. We sent the launch PDF (in the Appendix) to the stakeholders showing how to gain access to the dashboard.

The simulation was a piece of python code which needed to be run for all days of the week and for each store. Our trial had more than 5 stores so required everyone in the Retail Experience Analytics team to run some stores for the week. I participated in running these python codes with several inputs. This is where I increased my knowledge on how Python works and improved my use of the computers command line. The simulation code had a section to upload straight to the SQL database but sometimes this wouldnt work. I was required to upload the unsuccessful outputs to the SQL database using a previous python code I saved from the Till Upgrade work.

The trial went live on 7th March 2021 in some superstores around the estate. This went on for several weeks, running the python code every two weeks and updating the MSTR dashboard daily. We found low compliance for the majority of stores due to the COVID impact, so the trial got paused after 7 weeks.

From this piece of work, the business gained a resource and a working piece of code to restart this trial. This project has been moved to a Data Science Team to work through the issues that got raised during the trial and make it more efficient for next time. It also helped gage the pain points for using MicroStrategy in Analytics and we started talking about how best we can utilise MicroStrategy in the long term.

From this piece of work, I increased my knowledge of the business, having multiple conversations and correspondence with stakeholders. I also got experience on the analytical platform MicroStrategy and how to make automated dashboards. I gained insight into the processes of tracking a trial and how the business keeps their data reliable and secure. It was also the pre-requisite for the move to Data Visualisations, I worked with managers in DVA to get the dashboard fit for stakeholder use and ensured the distribution in a secure and swift manner.

Training in MSTR What does Analytics need to know to use MSTR competently?I moved over to a Data Visualisations team in mid-May to gain some experience in the area. I was working with two colleagues regarding the platform MicroStrategy. MSTR is a web browser application that holds data and is a quick and easy to make visualisations in either reports or dashboards.

In this team, I was helping create the Syllabus for training that ADA-Analytics would undertake on this platform. I was tasked two pieces of work in my time there: help create the syllabus by getting information from Analytics about what they currently do for pieces of work and create a proof-of-concept piece of work, a historical piece of analysis that I used Excel/PowerPoint for and reconstruct it on MSTR to make an analytical use case for MSTR. My time in this team, I worked more independently than previously.

Syllabus

During my time working on the Syllabus, I was tasked to create a template of questions, talk to colleagues in ADA Analytics to describe some projects that theyve been doing and update the template Excel Spreadsheet. During a meeting with the DVA managers, they found some interesting finds and have amended their syllabus to reflect what types of functions and visualisations Analytics use.

I then finalised the syllabus for the analyst learning path and constructed brief module guides and learning outcomes to describe well enough for a MSTR professional can construct the module content using MSTR and PowerPoint.

From this piece of work, Sainsburys ginned a syllabus with an analyst input. This syllabus will be rolled out throughout Sainsburys later this year in a few teams and will be updated when necessary.

From this piece of work, I gained insight into how Data Visualisations work, talking with DVA colleagues regularly. Also, I improved my communication skills by talking to many people around the Analytics Team and the Data Visualisation and Automation Team.

Proof-of-concept analysis

I was tasked to reconstruct a previous piece of Analytics work in MSTR. I chose the CSAT review that I presented a few months prior. This piece of work was chosen because it could be improved with the use of MSTR. In the previous analysis, I had to use 4 different data sets because there was such a large amount of data that couldnt all be analysed in Excel at once(over 1 million rows). In MSTR, I was able to get all the 2-year information with just one table and analyse the whole dataset at once.

I imported the dataset into MSTR and made a dashboard looking output first to make sure the data was correct, and the visualisations were well made. I then was tasked to make a PowerPoint looking chapter which uses the new Free Form Features which means you can resize and move objects around the page without it affecting other objects on the page. I then used another new feature, where you can scroll down a page instead of clicking to a different page, this meant it looked like more of an infographic style of document.

I presented this work to the ADA managers on Thursday 8th July. I got a positive response from this and some managers requested I present this to certain teams around the business, like: ADA Analytics, Customer Insights stakeholders and Marketing Analytics. I then got asked to present this piece of work to representatives from MicroStrategy, the external company that make and sell the platform, in which they responded with positive feedback too. I presented this piece of work to 4 different audiences, tailoring the content to each one.

The business benefitted from my assistance in this project because I had more knowledge about Analytics which was beneficial as I would know how to make the analysis output I did previously. Also, I know how Analytics works so can use that to sell the MSTR to them in a targeted way.

I benefitted from this project because I was more aware of the differences and similarities with Analytics and Data Visualisations areas in Sainsburys. I also increased my MSTR knowledge and understood working Agile better too.

Storytelling Game Plan How can we improve Storytelling in ADA?Within ADA Analytics, we have several teams but from September 2020 we started a process called the Game Plan which ADA Analytics colleagues can choose to be a part of improving an area of the working process while still working in their teams. There were 6 workstreams to choose from including: Training, Ways of Working and Shout About Everything We Do. I chose the Storytelling workstream to improve the ADA Analytics Teams Storytelling skills with data to present to different audiences around the business. I wanted to increase my presenting skills and have a passion for making our presentations and other forms of storytelling outputs more engaging and accessible.

During the first part of the Storytelling Game Plan, we investigated what Storytelling is and how to improve our data storytelling for everyone in ADA Analytics. We then had an idea to output this in a way that would help the wider team with a playbook, a PowerPoint with concepts to think about when preparing and telling your story. I was responsible for two sections of the playbook, the Audience Tailoring section and the Accessible section. I have included the slides I made in the Appendix.

We presented the playbook and the other outputs to the management team and then the wider analytics team in March. There was a lot of good feedback from this, and they enjoyed learning more about storytelling. This playbook is now widely used to prepare presenting insights to stakeholders and others around the business.

After making and rolling out the playbook, we came up with 3 areas to focus on: Panel, Communication and Upskilling the rest of ADA. As we made a reference document when making a story, we now were thinking of ways we could upskill the rest of ADA in an interactive way. I previously co-hosted the Accessibility in work/Storytelling, so I chose to help more with the upskilling of ADA. I made the content for storytelling in the forms of 2 other interactive sessions: Who/Why and What/How.

Development TimeSainsburys Tech value our training and development time as much as we do ourselves. We are allocated a day of work time every two weeks to train and develop skills of our choice. There are a range of things that I did on my development days. There are 4 main areas of development I worked on: external courses, team days, technical skills, and soft skills. The following is sectioned out to reflect this.

External Courses

Throughout my time working at Sainsburys, I was lucky enough to get specific training from several companies in the retail and technology sectors. In October 2020, I was able to participate and gain a qualification with IGD, a retail commercial analysis company where I learned about how the retail sector works and what they look for when comparing companies within the Retail sector. This qualification gave me insight into not only the retail sector processes but also it had a specific module covering personal development. I learnt how to write achievable goals and I gained insight into how to write a comprehensive personal development plan (PDP) which I used for writing one at my time in Sainsburys.

I was able to attend a Snowflake Sainsburys custom training in the SQL database Snowflake. This has improved my knowledge of how the cloud database works and how to efficiently write SQL code to pull data from this database effectively. This was a day course via Microsoft Teams with a PowerPoint to explain processes and interactive exercises in Snowflake SQL database after every module. This course provided me knowledge into the platform Snowflake, and it increased my confidence in speaking in front of new people.

I was able to train from MicroStrategy, the Analyst Pass, which entailed some eLearning and virtual classes from the colleagues at MSTR itself. This course taught me how to use the analytical tool, MicroStrategy, effectively and how data is used analytically behind the interface. This was a certification that was paid for by Sainsburys. It is the course I went on to remake for Sainsburys internal training. I gained a lot of knowledge from this course which helps me understand the use of the analytical tool.

Team Days

With my time in the Retail Experience Analytics Tram, I participated in 2 innovation days. The first innovation day, I was grouped with a senior colleague where we investigated our customer demographic and saw if there was a difference in customers using various checkouts. The other innovation day, in March 2021, I was in a group of 3 where we made a simple simulation in Python to simulate the queue length for our self-checkout tills in one store for one hour using Sainsburys Christmas data. In these innovation days, I increased my knowledge of the Sainsburys data and I improved my teamwork skills. I was working closely on these days with my team even when we were working remotely.

I was also a part of the Retail Experience Team Day in May 2021. On this day, all the team ran a half-hour session on either a soft or technical skill and the rest of us participated. I learnt a lot that day, some sessions included: personal brand, privilege, the use of a python environment called poetry and the Sainsburys competency framework. I ran a session on how to effectively use Jupyter Notebooks by using hot keys and main text cells. This day improved my confidence in the use of Jupyter Notebooks and made me more aware of personal brand and privilege in society.

Within the Data Visualisations Team, I was a part of the Team strategy day. Throughout the day there were many scheduled meetings like a Q&A with Helen Hunter (head of CDAO) and the strategy meetings where each Squad in DVA talked through their past years accomplishments and their road map for the next year. This was insightful to see when moving into DVA full time. I understood the breakdown of teams in DVA from this. I also participated in the Q&A and asked a question to Helen Hunter.

Technical Skills

When I first started at Sainsburys, I was on a 12-week programme to increase my technical skills in SQL, Python, R, MicroStrategy and hypothesis testing. After this, I used my development days to improve my technical skills using W3Schools.com and web documentation. I also used some YouTube videos to increase my knowledge on Pivot Tables in Excel.

In Sainsburys, other colleagues sometimes investigate one analytical approach in Python and write a runbook on it, a Jupyter Notebook document explaining an analytical application and show in the same document how to use it effectively. I used several of these runbook documents to learn about analytical applications like Correlation Analysis and Simulations in Python. These really helped me understand the statistical concepts and uses of these analytical approaches.

Soft Skills

When I started at Sainsburys, I was aiming to improve my presenting and communication skills. I was able to spend some of my development time to plan, construct and practice the wellbeing talks I offered to host/co-host. I investigated topics like accessibility, colour blindness, burn outs and sleep. This really helped me practice leading a conversation and researching topics on the internet.

I also went to several conferences during my time at Sainsburys, including: Women in Technology and LOVE DATA. Both were internal and had a networking session where I met people from around the data sectors. There were topics discussed like Imposter Syndrome and Data Security. It was insightful to attend these conferences, I learnt a lot. I took the opportunity to participate in presenting at the LOVE DATA conference in October 2020. Usually, the interns would plan and host a section in this conference, but this was the first year it was fully virtual. We had the opportunity to plan and host an ice-breaker session where we chose to run a scavenger hunt, finding objects around the house that would spell out LOVE DATA. This was a great experience to present at such a big event.

Aims and ObjectivesIn this section I will be talking about the extent this placement has met the aims and objectives for the professional placement year in the University of Hertfordshire. From the Placement Pack provided by UH, each sub-section of this part I will describe how this placement has met the aims and objectives.

Undertaken problem solving activities in the workplace

As an Analytics Intern I undertook many projects that required problem solving skills. As an analyst, we try to answer broad questions from our stakeholders like Is there a benefit to upgrading the self-checkout tills?. For these projects there are no instructions to what you should analyse, each analyst will approach it in a different way and will need to have a brainstorm session to cut each project into manageable chunks. These brainstorms would outcome in a plan of analysis that would show which data to analyse and how.

Another way I undertook problem solving skills in the workplace was when I was learning to code in SQL and Python. In these coding languages, there can be many errors in your code that means it wont run. Sometimes the errors that occur may not be useful to find the issue in the code. This requires problem-solving skills to fix.

Developed communication skills

This placement increased my communications skills across the board. It developed written communication skills in emails to stakeholders and other colleagues around the business. I was able to send emails without checking with my line manager from March 2021. Also, it developed my verbal communication skills in small groups and large groups. I was tasked to present my insights and findings every time I finished a piece of work, mostly every two weeks. This type of presentation was attended by between 10 and 30 people. I also put my name forward to present some wellbeing talks around ADA Analytics. This type of presentation would attract an audience of between 20 and 30 people.

Throughout my time at Sainsburys, I had 2 link conversations with my line manager to get feedback from colleagues so I can better target areas to improve. My first link conversation was in September 2020, where I had lots of feedback from team members and stakeholders saying I should improve on my presenting skills and confidence. Then in my February 2021 link conversation feedback was saying how I have had a change in my presenting skills and have improved but there is still room for more improvement.

Shown the ability to accept responsibility, work independently, and as part of a team, manage your own time and schedule your work

This placement has honed many skills and requirements in a work environment. Within my first few months, on the Replenishment work, I had to hit many deadlines and standards within this piece of work. I was responsible for updating and distributing an Excel dashboard via email every week. This was highly important due to it tracking an ongoing trial. I was distributing to the Retail Leadership Team, stakeholders and Area Managers that were in talks with Store Managers in the trial. This dashboard would show key performance indicators and compliance. In this context, compliance means the deliveries would be unloaded and unpacked in a certain order.

Within Sainsburys Tech, we work in an Agile way of working. This means that we usually work in 2-week sprints and will have enough work for those two weeks. From this amount of work, as analysts we break down the tasks into reasonable chunks so that we can distribute our workload throughout the 2 weeks. I had a lot of support and guidance to chunk out by work from the start of the placement but as I got more familiar with the pieces of work and the expectations of an analyst, I started to chuck out my work independently. Using agile ways of working has improved my time management skills and has let me schedule my work independently.

During my time at Sainsburys, I had projects that were collaborative and independent. I worked on the Closing Times piece of work independently, ensuring the piece of work was done to a high standard and the stakeholders got the output that they wanted, I analysed the data and built the tool within the two-week sprint, only speaking to my colleagues regarding questions about the data. I also worked collaboratively on some projects, including the SmartShop connectivity work. I worked with my colleague and we analysed different questions so we could come together and pool our insights for the same open question.

Understood implications of a formal working environment

This placement taught me the implications of a formal working environment. I started the placement in June 2020, during the COVID pandemic, so I was working remotely most of the year. Even with this way of working, I still gained valuable insight into the formal working environment.

I needed to be punctual and proactive when it comes to work. I was not late for meetings. This was reflected in my link conversation, I had feedback from a colleague talking about my good punctuality and work ethic.

Working in Agile at Sainsburys has shown the importance of time management, deadlines and working transparently. I participated in a meeting every morning called a Stand Up where we talk through what we did the previous day and what we will be doing the current day. This meeting is to flag up any blockers and share our progress within our team. This was a good way of being transparent in our work and to keep to our deadlines too. If deadlines were coming up and we may be struggling, then someone could help with the output as there was that transparency and support within the team.

Appreciated the relevance of your academic work and applied this to new situations

In this placement, I appreciated the relevance of my academic work. I have used a lot of the knowledge I have gained in the university in the placement year. One was the use of Python. A level 5 module, Programming, taught the coding language Python with packages that would help mathematics and statistics. I was working with Python and Jupyter Notebooks (a Python interface) within my time in the Retail Experience Analytics Team. This really helped me pick up more uses of Python quicker than I would have. We got taught how to use a Python package NumPy in this module and at Sainsburys I learned many use cases for the NumPy package.

I also used many statistics concepts that were taught in Basic Statistics at level 4. I used line graphs, bar charts and scatter plots in my time in the Analytics team. Also, in Applications of Mathematics, another level 4 course, I presented my work and tailor content to the audience which helped me understand what was expected from presenting within a company.

Used new tools and techniques to supplement those covered on the degree

In this placement, I used new tools and techniques to supplement those covered on the degree. Within the Level 5 Programming module, I was learning about a Python and some of its packages like: NumPy, a mathematical package, and mat plot lib, a visualisation making package. I used both packages during my time at Sainsburys and I learnt a more commands in these packages. This will support my third-year work too.

I learnt the concept of hypothesis testing in level 4 Basic Statistics and level 5 Statistical Modelling which supported me on hypothesis testing with real data. I understood the concept and applied it in the coding language R at university and I applied it within Sainsburys using Python instead.

Within the level 4 module Applications of Mathematics we gained insight into tailoring content to different audiences. I presented to an audience that knew about mathematics well then wrote a report that was for an audience of low mathematical exposure. I used new tools and techniques in my placement to present at different levels of mathematical understanding which built upon my learning in the university.

Acquired experience in the use of computer systems in business and industry

In this placement I acquired experience in the use of computer systems in business and industry. Due to starting this placement during the COVID pandemic, I was required to competently download and access many computer applications setting up my work laptop. I used the computers command line to download and install applications like Git and Python. This gave me greater understanding of how the computer systems work and why it is set up this way.

We worked collaboratively within Sainsburys even with the remote working by using Shared folders and the use of Git Hub, a platform to upload and share pieces of code. I was able to learn how to use branches and review others work on the platform.

Gained a greater understanding of the career you eventually intend to peruse

I gained a greater understanding of the career I eventually intend to peruse during this placement. It has given me the insight of knowing the day-to-day tasks that an Analyst has during their work. I also was exposed to both Analytics and Data Visualisations and Automation areas within Sainsburys.

I was able to speak with several people around the business including: Data Science, Cyber Security and Business Transformation. This consolidated that data was the place to be and there are many flavours of data roles, especially in such a huge company. I was offered a role as a Data Visualisation and Automation Analytst when I have finished my studies so I have accepted this offer knowing that I will gain experience in this area and possibly move to different data roles when I gain experience in the critical thinking required for Data Analytics.

Overall

Overall, I greatly enjoyed the placement and have seen real progress in both my technical skills and my soft skills. I had a lot of opportunities to present my work and lead conversations on topics close to my heart and have enjoyed the challenge of pushing myself to do this. I have enjoyed working in a team throughout the year.

This placement has been relevant to many of the modules I have taken during my time at university so far and has equipped me with many new skills to help me with the rest of my time at university also.

Level 4

Basic Statistics It has given me exposure to an analytical tool, MiniTab, so that I could transfer my analytical tool knowledge to MicroStrategy while at Sainsburys. It also gave me some understanding of using visualisations for statistics with graphs, charts, and infographics.

Applications of Mathematics This gave me insight into how we should tailor to different audiences which I built upon during my time in the Sainsburys placement scheme.

Application of Computing I was taught the use of MATLAB in a mathematical way during this module. Even though MATLAB was not used, in Sainsburys, I gained transferable skills during this module like touch typing and the processes of computers.

Level 5

Programming It has shown me what Python is and how to use Python effectively. I was taught how to use loops (if and for loops), mathematical functions and visualisation packages in Python, all of which I used multiple times and built upon over my year in industry.

Statical Modelling This module showed me how to effectively use hypothesis testing and understand the mathematics behind it which I built upon at Sainsburys. I used R to model statistics at university whereas I applied this understanding and used Python to model statistics at Sainsburys.

Careers and Development I gained many skills and insights from this module at the university. I had a workshop on how to effectively network before the careers fair which gave me knowledge on how to network in all the various events hosted by Sainsburys. Also, I had a workshop on how to write CVs and cover letters which helped my application process to get this placement scheme. The mock interview day was very helpful to understand what is required and the process of a graduate/placement application process.

Overall, I had a great time at Sainsburys Tech. It developed skills I learnt at University and it gave me new tools and skills from the placement scheme too. I have felt very supported within Sainsburys and from the university. This years placement process was rather different from previous years due to the COVID pandemic. Even with such a short timeframe to keep everything going remotely, both Sainsburys and the University of Hertfordshire have kept the process smooth for placement students.

It was a great placement with a great opportunity to understand what Analytics jobs entail. I feel I had good exposure to huge business and a supportive team that helped me flourish within a working environment. This year has focussed my sights on future jobs in the Data sector and has given me many skills and a good networking community within the data area.

AppendixAppendix 1 Storytelling Game PlanFigure A1.1 Tips on Audience tailoring

Figure A1.2 Audience tailoring Example (Dummy Data)

Figure A1.3 Tips on Accessible

Figure A1.4 Examples on Accessible Slides (Dummy Data)

Appendix 2 Simulation Launch PDF

Appendix 3 September Link Conversation FeedbackAreas of strength Summary: Heather has shown lots of enthusiasm, willingness to learn and improve her skills. Its great that she is using her development days to improve her Excel knowledge. She is passionate about coding and already has a good understanding of analytics.

Areas to develop Summary: Key areas of focus are storytelling/presenting, approach to an analytics task and linking the analysis back to the business context this will also help the story telling in sprint demos. Lastly, leaning on more experienced team members for development, I find that this can be the best way to learn. It is very easy to try and tackle a problem by yourself when you are working from home, and particularly as we are the type of people that enjoy problem solving, we can sometimes get stuck going down a rabbit hole, so being able to recognise this and reach out to other team members for advice and to bounce ideas off of will be really valuable for your development.

Any other comments

Heather has shown great enthusiasm and fitted well into the team, she got stuck in straight away and I feel we got to know her quite quickly which is great

I think Heather has started really well and has a great understanding of analytical techniques and an interest in developing her skillset. Her presenting has improved a lot in a short space of time, and she is knowledgeable when answering questions. I think she can lack confidence when presenting but am sure this will come with time. She has already improved so much in the 2 months Ive been here and Im enjoying working with her.

Dont be afraid to ask stakeholders questions, they have a lot more business knowledge than we do so definitely go to them to understand the business problem more.

Heather has lots of questions which is great, and I think the more she does of something (e.g., presenting, documenting), the more confident she will become.

Appendix 4 February Link Conversation FeedbackAreas of Strength Summary: Heather asks lots of good questions when were working together and comes up with lots of ideas for what we could look at in the work She is curious and likes to explore and try new things. She knows lots of different techniques for coding in SQL and starting to pick up Python which is going well so far. Heathers code is well commented and structured. Heather is determined and will keep trying when given a task Heather takes on feedback and considers how to progress herself Real interest in self-development and improving her technical skills. Ive noticed improvements with her public speaking, the burnout session she ran was excellent and really memorable. Well done! Heather was brave to push herself outside her comfort zone

Heather has introduced some really useful ideas in the Story Telling work stream. She can see the value in storytelling and that comes through in the effort and time shes dedicated for the material shes produced for this. She has a good presence in these sessions, and certainly doesnt shy away from contributing. Overall, Id say shes contributing and performing well in this part of her role. Heather has been a solid contributor to the Story Telling work stream.

Other

Great enthusiasm and has really grown in confidence, for example putting herself forward for the burnout talk and being increasingly proactive with the storytelling work.

Shes positive and has a good outlook.

In the last quarter Ive been impressed with the work shes done on Wi-Fi and keeping more specific and focussed to the task.

Areas to Development

Critical thinking

Being careful to not go down rabbit holes with work try to keep focus on the question / objective of the work

Heather is able to write code well and calculate metrics from the data, but I think she finds it difficult to translate the analysis into insight for the business. When analysing data, she could try asking herself why does this matter to the stakeholder, what will they do with this information? to help focus on what's the value of the analysis.

Can tend to overcomplicate and get a bit lost in the code, sometimes losing sight of the overall business problem.

Time Management

I noticed that on occasions Heather has seemed to be limited on time at the end of sprints to prepare her demo and it may have been a bit rushed. Heather could try leaving one day to prepare the demo and even if she hasnt done all of the analysis, she wanted she can focus on what has been achieved to present in the demo.

At times it seems Heather still can get too much in the detail. Remember to keep tying back to the bigger picture of what youre trying to achieve. If youre spending too long on a task, put it in the group.

Presenting

Continue working on presenting as I have definitely noticed a big improvement and Heather coming across much more confident in recent demos and team meetings.

Presenting focusing on the key points when presenting what is the main takeaway from the slide / work

Technical

Keep using python and trying new methodologies and packages and really push hard on developing this over the next few months and I think youll be in a good place.

Confidence

I think at times Heather can worry about being judged or getting it wrong. For example, writing emails takes her a while, she usually wants to check things with the team. I recommend giving things a go and learning as you go along.

Mainly I think continue as you are for the most part. Keep putting yourself in uncomfortable situations because this has helped your development to no end.

Other Comments

Overall a pleasure to work with you and have you in the team

Ive really noticed how shes challenged herself and taken on lots of different responsibilities lately from speaking about Burnout, to organising the meeting we had on data requests and her work on Game Plan.

Really impressed with how youve taken on board comments and tried to use these to help you.

Very hardworking and always polite, helping her to build good relationships with stakeholders.

Heather has been doing some great work in Python and Snowflake for the SmartShop work and has come up with good ideas for what we can look at for the work

More willing to ask for help rather than struggle through a problem for too long

Appendix 5 Analytics Use Case in MicroStrategy

This Piece of work was made from a prior Analytics piece of work in Data Visualisation and Automation to showcase the new features that MSTR has. This piece of work was presented to many people around the business and beyond. People like: Head of Marketing, ADA Analytics and MicroStrategy representatives all saw and commented on this piece of work to show the potential the new features have when working in MSTR performing analysis and creating an output in the same tool.

BibliographyOver the past year, I have had exposure and experience to a lot of tools and training. As Sainsburys is a secure company, I am not able to reference the documentation I have written for the projects I have participated on. Here is a list of public webpages for tools and training I have participated and used this year. More information about training is on my Linked In page.

Company:

Sainsburys https://about.sainsburys.co.uk/

Tools used:

Snowflake SQL Database - https://www.snowflake.com/company/

PyCharm - https://www.jetbrains.com/pycharm/

Jupyter Notebook https://jupyter.org/about

Git - https://git-scm.com/about

Git Hub - https://github.com/about

MicroStrategy - https://www.microstrategy.com/en/company

Confluence - https://www.atlassian.com/software/confluence/guides/get-started/confluence-overview#about-confluence

Microsoft OneDrive/Share Point https://www.microsoft.com/en-ww/microsoft-365/onedrive/onedrive-for-business

Microsoft 365 - https://www.microsoft.com/en-gb/microsoft-365/business/compare-all-microsoft-365-business-products?&activetab=tab:primaryr2

Training I participated in:

IGD course https://www.igd.com/Portals/0/Downloads/Learning/Fast-Track-2020-guide.pdf

MicroStrategy Analyst Pass https://www.microstrategy.com/en/education

Snowflake training Custom to Sainsburys so unfortunately no web exposure to it. The syllabus was cloud databases, warehouses, writing SQL code, efficiency of SQL code, capabilities of SQL databases.