Data drives the modern world and data science is increasingly prominent in our everyday lives — even if we don’t stop to think about it. This post looks at 5 examples of everyday applications of data science and what is being done to make the customer facing analytics both useful and engaging. After all, isn’t this what we all want from our data presentations?

1: Banking Visualisations

Speaking personally, I have two bank accounts. One is with one of the most traditional and established UK banks and the other is with a new company who do not have a single brick on the Highstreet, let alone a building. While both bank’s apps display my balance and allow me to do the day-to-day banking needs of managing direct debits and transferring money, the newer bank has a far greater array of services. It is because of these and the increased insights in my money they provide that I have swapped which of the pair is my main account.

Different bank apps provide different data visualisations for your account, but common examples include monthly spending comparisons, a breakdown of spending based on category, and a goals tracker visualising your progress towards a savings target with the possibility to see where you fell short and why.

These data visualisations are making people more aware of their spending habits and thus able to make better decisions thanks to the insights. I am still in my twenties, and yet I can remember having to go to ATMs on days and nights out to keep checking my balance to see how much was left so I could get home. Now we have the data not just at our fingertips, but visualised for us in real-time.

What I think is so impactful about this use of data is that the apps give us real-time answers to the questions we want answering, without us having to go to any effort to actually ask those questions. Instead, we are given the insights at a click of a button whenever we need them.

2: Health/Fitness Tracking

While we once needed an additional purchase and that deliberate decision to track our health and physical activities during the day, now many of us have phones that do it automatically. Furthermore, the inbuilt apps can then take that data and turn it into actionable insights for us.

While I eat far too many packs of mince pies a week to be considered a fitness buff, I still regularly look at this data through intrigue. For example, without me actively having to tell my phone to record or analyse anything, I know from the simple graph visualisation that I have been walking considerably less this month than last month, but that I’m still averaging the same overall for the year as I did in 2020.

Delving a little deeper into my steps, there are more analytics available too. While the step count is interesting and something I tend to look at after every outing and at the end of each week, the analysis of those steps is arguably more interesting. This is because the step data is broken down into the steps themselves to reveal my ‘Double Support Time’, ‘Step Length’ and ‘Walking Asymmetry’. These all show different things, but together they can work to suggest the ‘health’ of my walking.

For example, at the start of October I sprained my ankle for around the thousandth time. Based on the data breakdown, I can see that compared to the months before this clumsy accident, I have had a smaller stride, spent longer with both feet on the ground, and have had a far higher walking asymmetry meaning that the timing of my steps has been incredibly uneven and laboured. In fact, my walking asymmetry is so bad it has shot up from my usual average of 1.2%, up to a painful 50% average for my 4,944 steps yesterday. (And yes — if I had planned this post yesterday I would have definitely walked 56 steps around my room before bed.)

While I am fully aware that my walking is more difficult and painful currently and I know the reason for this, it is still interesting to see the data of my walking to analyse and confirm it. Furthermore, while I know the cause this time, it is something others have used as a reason to seek advice on what is causing the change in their walking.

As with the banking visualisations, this data offers easily accessed insights without us having to ask any of the questions. Indeed, when it comes to the deeper insights around my walking, I wouldn’t know which questions to ask. What is additionally useful with this data presentation, however, is that you are supplied with the context and needed information along with the data so that you can understand it. This means that instead of just being presented with a graph plotting my daily average walking asymmetry, I also have an explanation as to what it means and how to interpret it. If it did not have this information, I probably simply would not have bothered to look at this data!

Talking to others, I know this is a common response to inaccessible data presentations with people today wanting to be able to quickly understand the information presented.

3: Energy Usage & Carbon Footprint Calculated

As we all increasingly grow our awareness of our carbon footprints, either for social or financial reasons, there are several ways that data science is working for us. Perhaps the most everyday of these has been with smart metres. These devices record our energy usage for us with this data then analysed to grant insights. While different providers offer different information, these commonly include comparisons of your energy use compared to similar homes/households and ways you can reduce your usage. Indeed, providers routinely state that users of their apps cut their usage and thus their bills directly as a result.

For those who want further insights, you can also use simple phone apps that can visualise your energy usage to compare it not just to the average use, but to show you the most cost-effective times for your different uses. For example when it is the cheapest time to put on a clothes wash or use the dishwasher. Furthermore, people with smart home appliances are able to gain greater insight from data science regarding cutting costs and carbon footprint. This is because based on usage trends and desired outcomes, data analytics can predict your needs and set the appliances to work accordingly.

Speaking personally, the data storytelling I enjoy most from my energy provider is the one focussed on my emissions. It states how much less CO2 emissions I have generated because of my energy choices, and how many trees it would on average have taken in the Kielder Forest to have absorbed that amount. The benefit of this is that the visualisation takes complicated data that is difficult to understand unless you are an expert, and turns it into something engaging and accessible.

In a field like energy suppliers, this accessibility is of great value to customers because historically there has been a sense of confusion over price tarrifs and costs. While some of this may be down to the complexity of the industry, there has also historically been a belief that it is deliberate by suppliers in order to overcharge customers. By allowing customers to view their energy data and have it compared and analysed, customers can feel a greater sense of security in their provider with trust rebuilt.

4: Our Media Habits Visualised

This category is perhaps the one that has generated the most social media traction over the last few years thanks to Spotify Wrapped. Indeed, just this week there were a few full articles on sites including the Metro asking ‘When does Spotify Wrapped 2021 come out?’ Considering that Spotify wrapped is a data visualisation of a customer’s usage over 12 months, it is incredibly impressive that it has managed to capture the imaginations quite as much as it has.

The reason it captures the imagination and intrigues of people is that the data is given to us as a complete data story — from the big picture, to small specific facts and snippets that we will be specifically interested in. The visualisations support these data stories and so it works as a complete package for us. Without us asking any questions or having to do any personal investigations, we are supplied with insights packaged as a narrative focussed on us and our own likes and interests.

While Spotify tells us a story with our data in this annual event, all the entertainment providers use it continually to optimise our user experiences. This includes using data analytics to offer us personalised recommendations as we watch, and targeting marketing at what the analytics say we will want to see — from the product through to the specific art chosen.

While some find this invasive, others find it a vastly improved customer experience to be shown only items that we are interested in as opposed to having to scroll through videos that we have no interest in. For example, I have two friends who are avid TikTok users. Both scroll through the videos throughout the day on their breaks, and yet it is incredibly rare that they see the same videos. Instead, what they see is tailored to their interests and what they have previously enjoyed watching — or at least, most engaged with watching.

5: More Immersive Gaming

The gaming industry has also increasingly utilised data in their marketing — including targeted marketing and using AI to generate recommendations for future purchases. However, the industry is also increasingly using data storytelling to keep people playing the games too. On one level this happens within the games themselves with the players being given live updates on their gaming with progress data and summaries visualised. However, companies are now also taking this data out of the game and sending it to gamers as part of their marketing.

For example, I recently had a few weeks off from playing a game because of a combination of being bored with it and life getting in the way. Then, one Friday afternoon, I received an email with the subject ‘Your journey is not over yet’. The email contained a little data story of my game experience and progress so far, including things like the percentage of the map I had discovered. While I knew this was a targeted advert trying to get me to play the game again, I was enticed by the fact that I had so much more to complete and so that evening I played for a couple of hours when I otherwise wouldn’t have. The game had been put into my mind with me triggered to think about my personal game story by being supplied it in visual data.

Meanwhile, data visualisations are becoming a more detailed and important part of the gaming experience itself too. This is most often the case currently with competitive online games where stats are vital for both in-game levelling up/victories, but also in personal player development. While this exists within the game, there is also a whole industry surrounding third-party gaming statistics platforms giving you additional data visualised to offer the insights gamers want: from their biggest weaknesses needing improving, to how they compare globally for each stat from kills and weapon accuracy, to individual map performances.

While these stats serve a purpose, they are also becoming a part of the game experience itself with people spending considerable time reviewing and enjoying the stats for the sake of doing so. This has been achieved by presenting the data in a way that gives insights that people want and offering it within the context of comparisons and progress.

Summing Up

A common thread throughout all these examples is that the user is supplied insights they want and need, without them having to go out of their way to ask the questions. In the case of specialised platforms like banking and fitness, this is done by the provider being experts in the field and knowing what the most useful insights will be. Elsewhere data science is used so the presented information can be tailored to the viewer through customer profiling. Either way, these insights are then presented by communicating them as a story which is both specific to the user, and easy and quick to understand.

In short, the data has a great impact on the customers because it is accessible and focussed on them. It is not complicated and it is not generic. Moving forward, this is definitely something I will keep in mind with my own data storytelling and uses of data in marketing.

What We Can Learn From 5 Everyday Uses of Data Science was originally published in Big Data Today on Medium, where people are continuing the conversation by highlighting and responding to this story.

For the last couple of years, I have predominantly worked as a freelancer. While a lot of this work has been done online, I have reached out to find clients in my local area too. Partially due to the covid situation and partially just down to preference, I have predominantly had these meetings at coffee shops as opposed to private homes. I have also then done quite a lot of the work at coffee shops instead of from home too as sometimes I quite honestly just need to escape my four walls.

This post explores the post-pandemic coffee shop office as remote workers flock back to their lattes with the mantra ‘anywhere but home’.

As a side note, I would like to point out that this looks at those who are able to work from home and perhaps even encouraged to do so by their work. There are of course many industries where working from home is not an option or where employers are determined to get people back to their cubicles.

An Introduction to Coffee Shop Hot-Desking

Firstly, this is absolutely nothing new. Way back in 2012 when I was still doing my Undergrad I remember reading articles on this topic and most of my friends would go to a nearby café to study.

However, as with many working habits and expectations, recent times have thrown up questions over how things will be after the pandemic. In many industries, there was an almost hard reset with new habits and expectations forming. A huge part of this is the shift in the ratio of those working at the office, compared to those working from home — at least some of the week. For example, a 2021 McKinsey report analyzed the trends and concluded that:

Considering only remote work that can be done without a loss of productivity, we find that about 20 to 25 percent of the workforces in advanced economies could work from home between three and five days a week. This represents four to five times more remote work than before the pandemic and could prompt a large change in the geography of work
https://www.mckinsey.com/featured-insights/future-of-work/the-future-of-work-after-covid-19

While many have been able to negotiate and construct perfectly serviceable home offices, others have struggled hugely with the rigmarole of working from home. Common issues include distractions from people they live with, noisy neighbors and streets, a lack of productivity from being ‘at home’, and a sense of isolation and loneliness. Despite these obstacles, these people may still not be able to go to the office because of reasons including protocol and changed situations, and yet they would rather be anywhere but home.

This leaves some of us feeling stuck — we cannot go to the office, but we cannot stay at home.

The Problems of ‘The Nowhere Office’

Two of my favorite studies into post-covid working culture come from DEMOS: Post-Pandemic Places and The Nowhere Office. The latter report looks at the difficulties and problems that come from people having to do the same work, but without the established resources and infrastructure of the office. This quote puts it quite well — the idea of working from home meaning forever being stuck at the office:

One of the good things about going to work is just that: the beginning, middle and end of it. The Nowhere Office presents a future where these boundaries have dissolved and this needs careful management. Before Covid-19, our smartphones and laptops were already bringing our emails and notifications into our homes and even our bedrooms, but now the whole office is there.
https://demos.co.uk/wp-content/uploads/2021/03/The-Nowhere-Office.pdf

One solution for this is to use a coffee shop as a substitute office — allowing you to leave the home and stop associating your four walls of home and relaxation with productivity and the stresses of work.

However, while this is an option for this sense of going to work and for reducing the potential isolation from being at home by being among people with the buzz of public space, coffee shops have their issues for workers. For example, my last coffee shop meeting was with an academic who I was working with to help turn a number of their academic papers into more accessible and engaging public-facing pieces. We ended up needing to leave the coffee shop because the ‘buzz’ was too loud and we were unable to hear one another. Instead, we ended up on a bench along the river with our frozen breath mingling into the cold November sky before us as we shivered and turned blue. Other issues include privacy, with some work not being at all suitable for external eyes, and internet speeds.

However, for me, the biggest issue is the cost. While you can maybe nurse a coffee for an hour, you cannot do so for a full 9–6 workday. As such, you end up having to spend quite a bit for the privilege of working at the noisy sticky table…

My Personal Coffee Shop Office Costs

Over the last two years, I have kept fairly consistent records of how much I have spent during my freelance work trips to my local cafes. While I have not remembered to make a note every time, I have done so on most trips. And, quite honestly, the results are eye-watering.

2 Years. 281 coffee shop drinks. £843*

The cost is based on an average of 3 pounds a drink. This is based on a few searches online to find the average and the fact that my purchases are quite varied. For example, if I’m meeting with a client who is about to give me 30 quid for a first-edition 1922 modernist hardback, then I will splash out on my favorite dirty chai latte with a shot of pumpkin spice and squirty cream. However, if I’m meeting a potential new client for the first time with me a little unsure about the outcome, then I get a peppermint tea.

Regardless, that amount of money is staggering. And it does not include any food that I may or may not also buy during the outing…

For example, the other week I spent just under four hours with somebody at a coffee shop. I arrived before them and so bought a filter coffee as I got ready. When they arrived, they bought us both Americanos. As I now owed them a drink, I then bought their drink when I went up for food and we shared a wrap. Then because I had got the wrap as well as the coffee, they went and bought another round of drinks. Then, because we had empty plates and mugs for over an hour, I went and got a tea. They then got peckish and so they went and bought a cookie. Their cookie made me envious and so I went and bought one too. However, it was too dry for me on its own so I went and got a chai latte to go with it.

… This is a particularly extreme example of spending during a meeting, however, in truth, it is not a complete one-off. And, fortunately, I did make a profit from the time too with it initially just being for me to sell them a couple of books, but it then became me selling them several more and working with them as their new thesis editor for their Ph.D.

So… it worked out! But it probably does not always — especially when it is a pitch meeting with a new client or the upsells fail miserably as is more often the case.

Conclusion

From talking to others in a similar position to me, coffee shop working is incredibly common with many of us increasingly having ABH workdays. As far as I can make out and as far as a number of reports I’ve read seem to believe, the nowhere office culture is going to continue and so the cafe resurgence is here to stay.

As I now move from a freelancer to a contracted 9–6 man, however, I can no longer budget my cafe trips based on the income from that specific meeting. As such, I am going to need to start reconsidering my purchasing habits.

Nonetheless, the coffee shop office is here to stay. Thanks for reading.

ABH (Anywhere But Home) Workers and the Cost of the Coffee Shop Office was originally published in ILLUMINATION on Medium, where people are continuing the conversation by highlighting and responding to this story.

There is an accessibility revolution happening within Big Data where consumers are not just expecting a greater percentage of their data to be available to analytics, but for a greater number of their staff to be able to use the tools too. While there will always be groups who need dedicated data scientists conducting complex analytics from the ground up, many can now gain the insights with simple to use applications instead. However, while this is a great development in the field, it does mean there are team members who are now suddenly expected to work with data and its visualising who have had very little training to do so.

This post looks at how we can apply the same skills in design and marketing we use in other areas, to boost the impact of our data presentations.

1: Know Your Focus

This first point is about simply knowing what you are focussing on. In other words, what point you’re trying to get across. In most cases there will be a key takeaway and insight present in the data and this should be clear and centre stage. If this is not the case, then quite often further analysis is needed either because the answers have not been found, or the right question has not yet been asked.

It is usually desirable for a single data set to produce several different valuable insights and these will all need sharing. When this is the case, however, it is often more useful to present each insight or similar grouping of insights on their own — or at very least in their own dedicated sections. This means that the message is not lost with each having their own clear focus and moment of attention.

For example, if your study is into the utilisation of MRI machines at a healthcare centre, you will probably have a huge amount of data to explore and visualise. This could include every single machine, and all of its usage. Rather than including all of that on the dashboard used for your presentation, you need to see what the focus of that data is in relation to the key takeaways and insights you’re sharing.

2: Know Your Story

Once you know your focus, the next step is to know the story. Data storytelling is listed as one of the most important skills for data scientists in today’s market and this remains true even if you have found yourself as an ad-hoc data presenter.

While you may be confident and competent around data, not everyone seeing your presentation will be. Those people will be desperately looking at everything on the screen other than the statistics and graphs. Data storytelling is about targeting information at specific audiences through a compelling narrative — in other words, to get everyone to appreciate the insights, not just those who drool at the sight of a cumulative frequency table.

All too often data storytelling is surmised and defined as simply ‘communicating your insights effectively’. However, in my opinion this is not effective storytelling as you are not giving any of the needed information around that point surrounding the ‘why’. As said, the data needs to have a focus and that focus is what you are wanting to communicate with your storytelling.

For example, using the same MRI scan example, your focus and data story will be about what those analytics mean in terms of actionable outcomes. As such the visualisation should not simply be a graph of all the utilisation. Instead, it should be a graph highlighting the results that promote and reinforce your actionable outcome: perhaps that X is not cost-effective, or that Y is being overused in comparison to its effectiveness.

Once you have the focus and story, you can start to put together what information you need to include so that the insight is communicated most effectively. This means getting rid of anything that is not directly supporting that point, and making sure that you’re reinforcing and supporting the data you do present with any needed context or comparisons.

3: Match the Visualisation Design to the Focus and Story

In my experience, this third step is often people’s first or second step: They get excited and want to dive straight into the fun of the dashboard colours and graphs for the visualisation.

However, the design should match the first two elements. Put simply, this is because your visualisation is there to make your presentation of the insight and focus as strong as possible. As such, you should not produce a dashboard for your data presentation until you know the focus. You can of course play around with dashboards beforehand to help you process the data and discover your focus, but these are backend sources for you and maybe a few others — not the actual presentation.

Perhaps the most common mistake I have often seen with visualisations is that the presenter’s eyes have grown wider than the screen when they see all the options available and they try to fit them all into each presentation. Doing this often leads to an overwhelming experience where the main point that’s being made gets lost among the colourful statistical clutter.

Instead, you should incorporate the core design principles used elsewhere regarding white space, a reduction in visual clutter, and a clarity of vision.

It is also important to remember that while your data analytics tools probably have a lot of data visualisation capabilities, you do not have to use them exclusively for your dashboard. Instead, you can look at your focus, story and graphs and see what assets are needed to help sell that screen. You can then make them externally and bring them in. For example making a heading and a summary asset externally and then bringing them in as an image.

You may also then want a supporting image to help quickly and visually make the case you are making. Often these can be incredibly simple, but incredibly effective. For example, if your takeaway from the analytics and the data presentation is that staff should stop using the most popular machine and swap to another because it is better, then you could simply have the two machines dropped into the dashboard with one having a green tick, and the other a red cross. This is incredibly basic, yet it is a strong takeaway message presented visually in the design away from the potentially complicated data visualisations you are talking about.

Conclusion

As software and plug-ins for data analytics open up data science and its application to more and more people, more and more people will be expected to partake in data presentations. This could mean that people are expected to incorporate dashboards into their presentations, or it could mean they are expected to include customer sentiment into invoice sheets. Whatever the scenario, there will be people who are unfamiliar with or even scared of data who are suddenly facing it in their workplace.

Because of this, it is incredibly important to make our data presentations as accessible and engaging as possible. Rather than us having to develop a whole new skillset for this, we instead just need to apply the same simple design principles we use elsewhere to the data:

Focus on the key point we are making. Sell a story around that point so the message is communicated clearly. And design the presentation so it is clean, easy to understand, and engaging.

Thank you for reading.

3 Simple Tips for Boosting the Impact of Your Data Presentations was originally published in Big Data Today on Medium, where people are continuing the conversation by highlighting and responding to this story.

Turning the data of my longest Football Manager game into a memoir

When the pandemic was particularly all-encompassing, I found myself pouring more time than ever into gaming. At first this was as enjoyable as ever. However, after a while the dark blanket of social isolation and the painful realisation of my fruitless unproductivity bit hard. Rather than stopping gaming, however, I decided to turn it into a personal project. After all, it was not the gaming that was making me feel bad. It was just that it was all I was doing because of the stops in work and socialising. As such, I needed to find a way to bring a sense of productivity and focus to my hours spent playing.

This post looks at how I turned the facts, stats, and data of a football manager game into a memoir — arguably my biggest piece of data storytelling to date.

This post looks at the memoir as a piece of data storytelling and I am not trying to sell anything. The memoir was a personal project and it was bung on Wattpad for free. I also have no affiliation with the game other than as a player and fan.

A Brief Introduction to the Game

Football Manager is a football (soccer) simulation game where you can play as a manager of nearly any club or nation. This means being in charge of transfers, tactics and the running of the club in regards to training and press interviews. While you can watch the matches with the game increasingly boosting their graphics capabilities, I play it in a less visually stunning way of staring at the statistics of matches and the written commentaries as they flash across the screen, reading the staff reports on players with the progress graphs, and negotiating contracts down to the smallest detail.

In other words: It’s a game of data.

The game itself acknowledges this to a degree with the new version of the game coming out in November 2022 showcasing this:

The new Data Hub powers your squad analysis with the same metrics and reporting methods that clubs use in the professional world of football
https://www.footballmanager.com/games/football-manager-2022#desktop

Turning the Game Data into a Narrative

I had many hundreds of hours of gameplay to turn into a story and millions of facts and statistics to use: from 50 in-game years of tactical decisions and transfers, to the careers of all the players I’d worked with across the 30 odd clubs and nations during the save file. I needed to condense the information down and to find an overarching narrative. This is something crucial in any form of storytelling: Finding the driving thrust of the story.

A really effective way to do this when it comes to data storytelling in my opinion is to really focus on the personal stories and emotion of the data. This is because while a table of numbers and facts can be interesting to some, to others it is anything but. Some people need that sense of why the data is interesting, and one of the easiest ways to do this is to pull on the emotion of the reader by showcasing and illustrating the emotion in the data.

For this I kept it simple and focussed on ambition and the competitive drive to be the best. To showcase this, the narrative follows the chronological career of a manager who starts unemployed with no reputation, and works their way up to be the best of all time. (With various failings and challenges along the way otherwise it’s too easy and boring.)

There was still an awful lot of data exploring this aim however as pretty much every decision made in the game was to that aim and thus every piece of data was technically relevant. To condense this down more, therefore, I focussed on a few smaller stories within the bigger narrative which I found by looking at the common trends within the data as a whole. From this I found that a few figures were most recurring in the data and I added backstories to these of friendship and rivalry. For example, I turned the fact that Troy Deeney became a manager at a few clubs of a similar level to mine into a huge driving rivalry lasting decades with their failings and successes continually highlighted in opposition to my own. Similarly, I noticed there was a player I continually signed at a few different clubs and I turned this into a father and son dynamic of respect and love with their career highlighted beyond the 10,000+ others who could’ve also been noted.

To really cement the ‘personal story’ vibe as opposed to it being like a wikipedia page of results and facts, I decided to then write it as a first-person narrative. This meant I could add as much opinion and emotion around the data as I wanted with it becoming ‘the manager’s story’ as opposed to ‘the story of a manager’. The first-person voice is a good way to turn something data heavy into an engaging account through additional insights and comments from the narrator.

With this narrative voice and the emotional focus of the story chosen, I then just had to work through the data season by season to write what is now a top 5 football manager piece on Wattpad*.

*When you do a search on Wattpad for football manager stories there are only 6.

An Example from the Gaming Memoir

A Football Manager Career - Mj Manson-Hing - Wattpad

Here is an extract as an example of the storytelling. It’s the start of chapter 35 and looks at a period where I start unemployed having just been sacked back to back by Barcelona and then Beijing:

I returned from my holiday on the sofa in the summer to discover that my name was being linked with two pretty decent options. The first was offering me a return to Italy with Inter, whilst the second would see me heading to a new country with Borussia Dortmund. Both options definitely had potential and in many ways they were very similar as they were both sides who were challenging for the league in their respective countries without being the number one favourites. Due to their similarities, Inter were probably the most sensible option as I was already fluent in the language and had experience of the league, albeit over a decade before.

As has often been the case in the past, however, I had an emotional pull towards the less rational option. Troy Deeney had spent eight years there and was listed as a club icon and I quite liked the opportunity to go there and do better than him — just like I had done at Barcelona when I had outdone Pep (prior to being sacked anyway.) On the 20th August, 2048, I signed the deal and took over at the Signal Iduna Park.
https://www.wattpad.com/1086096431-a-football-manager-career-35-chasing-troy

The first-person voice offers an engaging way to frame this section which is ultimately: They became manager of Borrusia Dortmund. To boost the drama of the piece, the decision is framed around the influence of Troy Deeney and positioning him as the antagonist with the narrator focussed on beating them.

Data from the game used in this extract alone includes: email history from 2048 (10 years before I retired), club histories, manager histories and the manager profile regarding things like language proficiency.

Summing Up

In many ways this was nothing more than a frustratedly bored and sad guy turning their endless gaming during the pandemic into something even more pointless and geeky. However, I really enjoyed the challenge of looking at the stats, facts and data from these hundreds of hours of play and more than 50 years of in-game time to write a narrative. If nothing else, it was a good way to practice my storytelling in a longer form than articles with there a lot of data to be looked at and turned into something that was hopefully engaging.

While it is certainly not my best piece of writing with me not going through and giving it the extensive edit that all work of this length no doubt needs, I am pleased to say it has been read more than five hundred times on Wattpad and it does have a fan!

Indeed, Lisa_Ken left the comment:

How could nobody read this

Well, even if no others do, I’m thankfully that they did and I enjoyed writing it even if just for myself.

Thanks for reading.

A Gaming Data Story — My FM Memoir was originally published in Data Narrator on Medium, where people are continuing the conversation by highlighting and responding to this story.

I recently realised that spread across a few different phone chats I had a decade of data at my disposal. This post explains how I turned the information into usable data and gained insights from it.

The data in question is the stats from a game that’s been going on for ten years with a group of my friends. The rules are very simple: For every stranger that returns your smile, you get a point.

The post looks at the process of turning this data into something engaging and public facing. This is a case study for accessible and engaging data storytelling.

Gathering the Data

I started by going through the chats and seeking out the needed data. The way we have done it over the years is that we will occasionally show off a particularly successful day or month as a tease and brag, and then at the end of the year send each other our overall stats to see who the winner is. As such, I needed to go through and find these moments when we shared our results and put them all in one place.

In a couple of cases the results were no longer in the chat, but fortunately I’d archived some older ones where it was. In one case, I had to message the person to ask for theirs again.

However, it all existed. I just had to get my hands on it! As such, this wasn’t too hard.

Inputting the Data

As the numbers are pretty straightforward, I decided to simply put them into a spreadsheet. Along the top I put the years, and then beneath I listed the totals each ‘smiler’ got — identifying them by a letter as opposed to their names.

Once done, I totalled things up and added each person’s annual average too.

I now had a simple table showing everybody’s individual results and also their totals from the full decade giving us the overall winner. (Sadly, not me.)

It is at this stage that I realised I had the data of over 12,000 recieved smiles. This is a lot! And it seemed a waste not to do anything more with them. I thought about what might be interesting, and then decided that what could work nicely was to enable the date to be filtered through social categories. As the group are my friends, this was easy as I already knew the information of their gender, race, and sexuality.

For fun, I then decided to add a filter by attractiveness as I thought this could be interesting too. For this, I used an ‘objective’ and ‘scientific’ app that rates people’s faces… While this is not at all scientific, it is fun and it’s far better than me deciding which of my friends are attractive and which are not!

Interpreting the Data

Now I had the data all together, it was time for the analysis. For this, I put it into Tableau Public. After having a little play to see what was interesting, I focussed on two things:

1: The information in the most simple form of it showing 10 years worth of data of people smiling and seeing how many smiled back. This looked interesting for analysis based on if there were any trends over the 10 years.

2: Grouping the data to see the numbers of smiles returned based on comparing different groups together. For example gender, or attractiveness.

I worked with the data in these two areas and came up with some interesting insights answering these questions. The insights were:

1: There was indeed a trend with there a decline in received smiles over the ten years. This was shown both in the first annual total and the final annual total, and across a majority of the years’ results.

2: The groups were interesting to compare and there were often clear winners. For example, a black, straight, above average attractiveness woman would likely receive the most smiles back.

Presenting the Data

For the data storytelling, I decided to focus on two methods: a single page interactive visualisation, and a written article (probably linking to the first.)

The Visual

Something I personally think is really important with visualisations of data for sharing widely, is clean UI and UX. In other words, a casual user being able to navigate the information quickly and easily with no external knowledge required.

This means that the data needs to be visualised in simple ways that do not require more than basic graph reading skills to read, and that there needs to be assets included for explaining and to make it engaging. For this I used a combination of the visualisation tools on Tableau, and external design tools for the assets.

Here are sceenshots and links to the two visualisations. The first focussing on the total annual number of smiles returned, and the second on the social filtering of the data.

The Written

As I am enjoying using Medium, I decided that I also wanted to do some written narratives for these two data stories too. With the written form I could focus a little more on the context to sell the story of it being data from a decade long game with friends, and give a little more context to the findings with qualifiers and provisos. Also, while the social commentary visualisation grouped the information of which gender recieved the most average smiles back etc, I didn’t offer my own guiding interpretation of this. In the data narrative of it, I had more scope to sell a particular angle if wanted.

• A Decade of Declining Smiles?
• A Fun Social Commentary Based On A Decade Of Smiles.

Final Thoughts

The Smile Game was never intended to be anything other than a bit of healthy competition between friends to help make the commute and other outings like chores more entertaining. Nonetheless, it was easy to take the results of the game and turn them first into usable data, and then into data stories. I am not for a second saying the insights are scientific and something that should ever be actioned upon (though others playing the game and the world having more random smiles in it can’t be bad?), but the insights were interesting! At least, they were to me.

It was also quite a fun little task in seeing what could be done with the data in a public facing way with accessible and engaging data stories.

Thanks for reading!

How I Turned Smiles into Data and Insights was originally published in Dataology on Medium, where people are continuing the conversation by highlighting and responding to this story.

The YouGov Body Image Big Survey of 2021 aimed to learn what Brits think about body image and how important it is to them. This post looks at the data to see what it says. All quoted statistics are from this survey’s published results.

The Data

The survey study took place in February 2021 with a sample size of 2,271 UK adults aged 16 or older. The full results can be found here:

Their Results

The YouGov website usefully gives us some insights fully formed for us to look at and their headline finding is that:

The vast majority of people in the UK — nine in ten (89%) — say that in today’s society physical appearance matters, either a great deal (46%) or somewhat (43%).
https://yougov.co.uk/topics/health/articles-reports/2021/08/05/yougov-body-image-study

However, for me personally this did not come as news. I am a 90s baby, and as such I’ve spent my whole life being bombarded with marketing that states that beauty matters. Indeed, the survey findings support this. Saying that:

Six in ten Britons think the media are promoting unattainable body image expectations, and eight in ten say the fashion industry has harmed appearance perceptions.
https://yougov.co.uk/topics/health/articles-reports/2021/08/05/media-fashion-companies-and-pressure-having-perfec

My Thoughts

As I already knew this, I was instead really interested in looking at the other end of the spectrum to try and find refreshing and uplifting statistics to prove that we are moving towards a culture where this is removed — or at least lessoned. To do this, I wanted to compare the age groups to see if the youngest group surveyed (16–24) cared less about image than the older groups. (I am in the next group up of 25–39s.)

Sadly, I did not find an overwhelming change towards a less image centric future.

For instance, while under half of those aged 25 and up thought ‘physical appearance matters a great deal in today’s society’ (42% of over 60s, 47% of 40–59s, 47% of 25–39s), the 16–24 year olds were the only group to be over half with 51%.

This difference between the age groups was strengthened with a perhaps more pertinent question: ‘To what extent, if at all, do you feel under pressure to have a certain body type?’ While 40–59 year olds and the 60+ groupings had under 10% answer ‘a great deal’, the younger two groups had above 20% respond that way. Furthermore, the younger of the pairing responded the highest: 27% of 16–24 year olds and 21% of 25–39 year olds.

These results were mirrored with the questions about sex too. While only 4% of over 60s said they thought about the way their body looked during sex a great deal, the numbers increased down the age groups with 18% of the younger two groups saying this. Likewise it was the two younger groups who answered in the highest numbers to the question ‘a great deal’ to this question too: ‘How much do you worry about what your partner thinks about the way you look? (If you don’t currently have a partner, think about when you did.)’

Furthermore, despite this awareness of the pressures of body image in society, the youngest age group were also the ones who personally thought that image mattered the most in others too. For example, there was one question which listed a number of physical characteristics and asked poeple to say which they found more attractive. The included factors around weight and height. While 34% of the over 60s responded that they ‘Don’t mind any’ and 20% of the middle two ages said the same, just 18% of the 16–24s responded this way.

These results suggest that while the youngest group surveyed said they felt the most pressures around beauty and image from society and peers, they also held some of the aggravating views themselves too.

Final Thoughts

This does not of course mean that the future is definitely going to continue getting worse in regards to these concerns. It could instead simply be a case that younger people care more and you care less as you get older. However, we know the power of change that can come from the younger generation in today’s age — just look at climate change activism. As such, I think this is still really interesting to see and I do wonder how things will progress by the time of the next similar survey.

Thanks for reading! :)

Only 1% Of Brits Do Not Care What You Look Like & This Is Unlikely To Change. was originally published in Data Narrator on Medium, where people are continuing the conversation by highlighting and responding to this story.

I recently wrote a post exploring the stats from ‘The Smile Game’ — a decade long game between myself and some friends to see who could get the most strangers to return our smiles. I broke the stats down by social group averages like by gender and attractiveness to see which grouping got the most people smiling back. You can view that here. Rather than stopping there, I thought I’d have another look at this unplanned decade of data to see if there was a trend regarding the number of smiles we received back each year.

Sadly, there was! This post has a quick look at this unscientific and unplanned decade of smiles.

The Data

7 people smiling at strangers for 10 years. 12,038 returned smiles.

As said, this is in no way a scientific study. Instead, it is a game played between overly competitive and bored friends during commutes. Furthermore, the stats could in truth be complete lies as we all logged our own returned smiles of the last ten years in our phone notes and other places. As such, we have to trust the honesty of the individual players… something that sometimes may have been flexible. For instance, one particular friend who will go unnamed always has a sudden spike in their stats whenever they noticed they were falling a little behind on the group chat…

Nonetheless, while the stats may not be the world’s most accurate and clinical of studies into smiling in the UK, it is still a little bonus pool of information.

‘Interesting’ not ‘Fact’.

Even if we take the data as fully accurate to the game, then the numbers themselves are still hugely impacted by our individual levels of effort and thus cannot fully support a conclusion about British society at large.

For example, while contestant G got 790 returned smiles in the first three years of playing, they only got only 439 smiles returned in the following 7 years. This could be because society suddenly got grumpier and no longer appreciated their beautiful smile. It could be because they grew a beard and that put people off. Or, it could be because they got a new partner and they suddenly found themselves with more interesting things to do than smile at strangers on the bus. We simply cannot know! (We do know. It’s the last one.)

As such, any insights and trends gained from this data must be taken with a huge grain of salt and seen as fun and interesting, not’ facts’.

The Trend

Nonetheless, the data we have sadly suggests a gradual decline in the number of people returning smiles in the UK over the ten year period. While 1,243 people returned our smiles in 2011, just 648 did in 2019. Over the 10 recorded years, 6 of them showed this decline:

1,243 — 1,201–1,132, — 1,007–839–684.

However, there were also a few fabulous years showing a spike in returned smiles. For instance, after the second year saw us lose smiles, the third year saw a huge hike jumping to an all time high return of 1,726. Speaking for myself, I did put in more effort in the third year for various reasons and thus saw more smiles returned. We also had all 7 of us playing in the same year, while most years saw at least one of us not fully invested in trying to win the chocolate bar given to the winner of the year.

Even so, the decline happens over a majority of the years and is supported by the starting and end figures.

Summing Up

So there we have it! According to our self-recorded stats based on a fun game where we all put in varying degrees of effort at different times over the ten years, we can see a clear trend in declining smiles in Britain!

As such, it is clear what we must do: Smile more!!

After all, even if nobody smiled back at us once over this time and we were just a bunch of weirdos on public transport smiling at people… our game still brought over 12,000 smiles to the world.

The stats only record the smiles we got in response to our own successful smiles too. As such, that 12,038 is actually 24,076 recorded smiles added to the UK streets thanks to our game with us smiling first to get the smiles back.

Furthermore, these are just the successful ones. I know personally that I probably only got around 1/25 of my smiles returned over the decade. If we take that across the board, that’s almost half a million smiles added to Britain thanks to this game alone! Imagine if we all started playing?? The decline would be over, and there would be peace and joy across the UK!

…Maybe not. But nonetheless, interesting! To me anyway — I hope you’ve enjoyed too. Thanks for reading :)

A Decade of Declining Smiles? was originally published in Data Narrator on Medium, where people are continuing the conversation by highlighting and responding to this story.

This post looks at the fabulous viz by Mariona Banyeres who has visualized the data of 939 New York Times bestsellers. While there is a lot of really interesting data on offer, this post looks at what makes up a bestselling family. I was particularly fascinated by this topic as the vizz shows that words making up the family unit are present in 649 titles. This is almost 70% of the books looked at! Meanwhile, the actual word ‘family’ was in over half of these. As such, I wanted to look a little closer at the data so I can see what I need to include to one day write myself that chart-topping novel!

The data in this post is from Tableau Public and is available to view here. Meanwhile, this is is what the author Mariona Banyeres says about her data:

First published in 1931, the New York Times’ list of best selling books needs no introduction. Known to be only partly based on book sales, the method by which it is compiled is shrouded in mystery. In the course of this project, the plots* of 939 titles which reached the #1 spot of the list between 1931 and 2020 were scraped and processed, and the resulting 140k noun phrases parsed, in search of recurring characters and themes. Delve into the results all around for a glimpse of the colourful, bustling, crime-ridden world of NYT bestsellers.
*from Goodreads and Wikipedia
https://public.tableau.com/en-gb/gallery/new-york-times-1-adult-fiction-best-sellers?tab=viz-of-the-day&type=viz-of-the-day

Who Is In The Family?

Out of the top nine most mentioned family roles, it is perhaps unsurprising that mother and father are right up there with 256 father mentions and 233 mother mentions. The difference between the numbers is not particularly drastic, however, it is enough to make mothers only third on the list with daughters coming in above with 244. Sons bring up the top four with 227 mentions.

Based on this, it seems that a family comprising a father, mother, son, and daughter would do quite well with audiences as these are far and away from the most frequently included. There is then a huge drop down to the next members on the list with cousins getting a collective 71 mentions, grandmothers 44, uncles 38, grandfathers 37, and aunts 22.

And, of course, the pets! It seems there is a clear answer to the age-old debate of cats or dogs with these results. While cats get just 16 recorded entries in this data visualization, dogs get a whopping 56. 53 of these are simply ‘dog’, while there are also 3 puppies, 2 hounds, and a single chihuahua.

Who makes the best husband and wife?

Banyeres includes analysis of the most common adjectives accompanying the words woman/wife, and man/husband. According to these, the best wife for a New York Times Best Seller is:

A young first wife.

While the best husband for a New York Times Best Seller is:

A young abusive/late husband.

It is very interesting and sad to see that an abusive husband was the joint highest type according to the common adjectives, but at least it offers a potential plotline with him passing away! (Or perhaps regularly being late for things, but the first is more dramatic.)

Who Would Be A Good Supporting Character?

The data suggests that friendship is incredibly important with 368 bestsellers including this. One potential suggestion could be for a housekeeper to be friends with the family. While only 16 of the stories included one, 158 of the stories had key themes surrounding money and wealth and a housekeeper would be a natural inclusion in this.

What Should Their Jobs Be?

113 of the stories prominently included lawyers, with 189 mentioning courtrooms and associated words too. As such, it seems good that one of the pair should be a lawyer. Meanwhile, even more of the stories included ideas of law enforcement and intelligence services.

From this, it seems we have a spy and a lawyer as our central couple. It’s a good job they have a housekeeper as they’ll no doubt be very busy!

Anything Else?

The final insight I’m going to consider for my perfect bestselling family is that health issues are common with 246 of the books including them. As such, one of the children is going to suffer a wound or injury (the top two examples), while the other unlucky child shall sadly have cancer.

Summing Up

So there we have it! From the insights granted by Mariona Banyeres’ viz, I have learned what makes a bestselling family in Adult Fiction! We have:

A young first wife who is a spy married to an abusive husband who dies potentially during his work as a lawyer. Fortunately, the family is incredibly close with their friendly housekeeper who looks after the children while the parents are busy with work. This is important, as the daughter suffers from serious injuries while the son has cancer. And, of course, there’s at least one dog to be looked after too!

Perhaps the wife will go on a courtroom crusade for justice, resulting in a large payout adding to their already sizeable wealth enabling them to live happily ever after.

Now just to get the thing written! Thanks for reading.

The Perfect Fictional Family According To The Data was originally published in SYNERGY on Medium, where people are continuing the conversation by highlighting and responding to this story.

This post brings together a few quotes about data storytelling before offering my thoughts at the end. It's designed as a simple introduction to the topic.

What is Data Storytelling?

Data storytelling is the process of creating a story out of data analysis findings, allowing regular business people (laymen) to understand the possibly complex insights and apply them to a business decision or action.

Crafting a narrative is the most effective way to communicate, and visuals are the easiest medium for the human brain to understand. Therefore, data storytelling with visualizations is the most effective way to present data.
https://www.import.io/post/8-fantastic-examples-of-data-storytelling/

The phrase “data storytelling” has been associated with many things—data visualizations, infographics, dashboards, data presentations, and so on. Too often data storytelling is interpreted as just visualizing data effectively, however, it is much more than just creating visually-appealing data charts. Data storytelling is a structured approach for communicating data insights, and it involves a combination of three key elements: data, visuals, and narrative.
https://www.forbes.com/sites/brentdykes/2016/03/31/data-storytelling-the-essential-data-science-skill-everyone-needs/?sh=403d161f52ad

Why Is It Useful?

If you want people to make the right decisions with data, you have to get in their head in a way they understand. Throughout human history, the way to do that has been with stories.
https://mitsloan.mit.edu/ideas-made-to-matter/next-chapter-analytics-data-storytelling

Data storytelling turns data from neutral fields in a database into opinions, arguments and insights.
https://www.tableau.com/about/blog/2020/8/importance-data-storytelling-next-decade-data

My Thoughts

Data can be truly fascinating and hugely important. However, not everybody cares about that raw data. In fact, some people will run a mile from it. Instead, these people want to be shown why this data is interesting and why it is useful. Data storytelling is the way to do this as, like is said in one of the quotes, storytelling has been used throughout human history to help share information.

My Data Storytelling Publication

The new publication ‘Data Narrator’ looks to share some of the amazing data out in the world through engaging storytelling in the form of posts here on medium. Data can be on literally any topic under the sun, and the same sheet of numbers can have many different stories to tell.

I am excited to see what data stories we can start sharing together. To learn more and get involved, go check it out here!

An Introduction to Data Storytelling was originally published in Big Data Today on Medium, where people are continuing the conversation by highlighting and responding to this story.

This post looks at the data collected from a decade of my friends and I smiling at strangers to see who got the most smiles back. While it’s a game and was never intended as a social experiment, I thought I’d plot the results into some analytics anyway! Why not?

The Context

The game began in 2011 when my housemate and I started to get bored of our London Underground commute to Uni. Like many, we ended up turning it into a game and we competed to see who could get the most strangers to return our smiles.

If you have ever been on the tube, then you know this is not necessarily that easy. In fact, it is against the unwritten rules that you should keep yourself to yourself and avoid eye contact — let alone smile.

While it started off as just the pair of us, it soon spiralled outwards to our partners and other friends. Others have come and gone over the years, but there has been a core group of us who have played for a decade! (Even if some of us do occasionally stop for a while or only remember to play in the few days we see each other.)

The Data

The interactive Viz can be found on Tableau Public here.

The number of smiles we received were logged by ourselves in either our phone notes, or somewhere similar. As such, it is honestly anybody’s guess as to how accurate they truly are. But, of course, my personal results were completely accurate!

I have not included people’s names, nor the ability to filter the pie chart topics to show the relevant smilers. I initially did, but then I realised it was possible for my friends to see which of them was deemed unattractive by the crude app and so disabled it!

The Analysis

As suggested by my visualisation and the chosen data to focus on, what jumped out at me was the opportunity to do a little bit of social commentary with the data. After all, this was a decade of statistics on who while out and about in the United Kingdom was smiled at the most.

Of course, for it to be a better social commentary the results should have been done more formally than our own self-monitoring. We should also have made sure our friend group was truly diverse and representative with equal numbers of people making up the groups.

Nonetheless, I could plot the results based on average annual smiles returned by a few groups. These being: By Gender, By Race, By Sexuality, and By Attractiveness. The first three are self-ID, while the final category was determined crudely by a free phone app rating our photos as either above, or below average attractiveness. Once more, I apologise to my friends and to science.

The Results!

So according to the stats, if you and your friends want to start playing the smile game too then you better start saving the pennies as your black, straight, above average attractiveness female friend is certain to win!

…Or, perhaps, our champion of the last decade is just particularly skilled and would’ve won whatever the circumstances. Who knows? Clearly we need to do another decade and find some more people to find out!

Stay tuned and I shall see you again in 10 years. Thanks for reading.

A Fun Social Commentary Based On A Decade Of Smiles. was originally published in Data Narrator on Medium, where people are continuing the conversation by highlighting and responding to this story.