_edited.png)
In 2016 I landed my first official analytics job. Before that, I had dabbled in working with data - I was officially the go-to person for anything related to spreadsheets in my previous jobs - but this was the first time I had a matching title to go with what I did: Insights Analyst.
The role involved answering business questions, partnering with a specific business team and helping them gather and analyse their data while running legacy reports and developing new ones. When I first joined, the team primarily used Excel and the company was not known for its innovative approach.
I had this woeful report I had to update monthly, feeding from multiple spreadsheets I’d receive by e-mail and collate together in a truly dreadful spreadsheet, which was passed down to me from a long line of analysts prior. I remember my manager telling me he knew there were inconsistencies and errors in the spreadsheet, but no one really took the time to look into it, so we’d just run with the report as it was. That didn’t sit right with me, so I set myself to rebuild it in Excel at the time.
But then along came Tableau. I still remember the first time I downloaded the free trial license. After several months of advocacy from myself and a colleague, we were able to convince our manager and team to give it a go, following the lead of the Business Intelligence team. There were rumours that the company was planning a significant data transformation and Tableau would be a part of the new technology stack. After a lot of back and forth, the two of us finally got our Tableau licenses, and the rest is history.
Or, so we thought. In the end, things did not turn out as expected. Tableau really revolutionised the way I worked. It forever changed my career path. No other BI tool has ever come close to giving me that same spark of joy. But it wasn’t all rosy and perfect. In a slow and painful burn, one by one, our projects faded and failed. But why? What happened? The tool was solid. The strategy was there, and the intentions were good.
The answer lies in two concepts intimately connected to any data initiative’s success, regardless of tools, but which are so often overlooked and misunderstood: Data Literacy and Data Culture.
The first bumpy steps on the journey to analytical maturity
In 2018, I was more deeply entrenched in using Tableau, assisting my team and stakeholders in transitioning from Excel and gathering requirements for new reports. The analytics landscape of my employer was changing rapidly, and a new challenge arose: adoption.
We were facing quite a bit of pushback a couple of years in. There was no clear strategy for managing change and adoption from the business side. The company had no standardised methodologies in place, and each team member developed and delivered projects independently, without a consistent approach.
If we’re creating something better, cleaner, leaner, and quicker - why was it that no one wanted to use it? We had a strange phenomenon of teams only accessing our reports as if they were some ERP system transaction, dumping a table into a spreadsheet and taking it from there (SAP Business Objects, anyone?). What was stopping them from using the carefully curated visuals right there, in front of them, created based on their supposed requirements? Eventually, I asked the question out loud, and the burden of fixing it fell on me.
I’m not one to shy away from hard work, so I took it to heart as a challenge. I rolled up my sleeves and set to research. Someone out there must have gone through this before me, right? Could I single-handedly figure out how to increase stakeholder engagement?
As with anything analytics, the first question is often inadequate, and we should always keep inquiring. That was when I first came across Jordan Morrow’s work on Data Literacy at Qliq. If there is one thing I’ll always give Qliq kudos for, it is that they were the first tool vendor to truly understand that Data Literacy was the real driver of change and improvement in the data landscape. Back then, they already had a treasure trove of free material and resources to be used for free by anyone interested in upskilling their own Data Literacy or in promoting it with their teams. That initiative evolved and became the outstanding Data Literacy Project.
The idea of Data Literacy stroke a chord in my heart. This was the missing piece. I realised we invested a lot of time and energy in all the shiny new tools but none of it in ensuring we, our stakeholders and the technical teams were ready to take on so much more data, analytical complexity and speed in our day-to-day processes. The skills just weren’t there.
What is Data Literacy and Why does it Matter
Be Data Literate - The Data Literacy Skills Everyone Needs to Succeed by Jordan Morrow is a book released in 2021 aiming to demystify the data literacy discussion. It is down to earth, full of examples, and easy to read and follow. It has a conversational tone. It feels as if the author himself is sitting on the porch with you while also talking passionately about his favourite subject.
The first couple of chapters are dedicated to getting the reader on board with the data revolution. Morrow explains a bit about the skills gap facing us when it comes to collecting, reading and interpreting data and how they are essential skills for the present and future. He touches a bit on the importance of changing the way we behave towards data consumption and the impact of democratising access to information.
The author then introduces the first framework of the book. An analytical maturity model is explained as a puzzle consisting of four levels we need to piece together: Descriptive Analytics, Diagnostic Analytics, Predictive Analytics, and Prescriptive Analytics.
After introducing us to these concepts, Morrow defines Data Literacy as a broad term. According to the author, Data Literacy is “the ability to read, work with, analyse and communicate with data”. The first definition I saw back in 2018 mentioned “argue with data” in place of “communicate with data”, and I do remember that to be the first bit of feedback I received - “argue with” could be easily misinterpreted as questioning for the sake of disagreeing, often in an unproductive way. It is nice to see that the author starts his chapter by making the same distinction. While I personally believe “communicate with” takes a tiny bit away from the inquisitive aspect of being data literate, I agree it is a better fitting term for a broader, initial definition.
The book then explores in more detail what each of the four parts of this definition means, giving examples and making the point that everyone, regardless of level or profession, needs these skills. This is how we bridge the gap and learn how to do more with the data we have.
I’ve met people who proposed that Data Literacy was a too strong and charged term. Literacy is often associated with knowledge, and its lack is often incorrectly attributed to the opposite: ignorance. I’ve heard long discussions on how the term had been co-opted by Data & Analytics teams to express their frustrations and assign blame to this invisible force that seems to haunt their stakeholders: their lack of Data Literacy.
This book is a great starting point to debunk this myth. Data Literacy is a complex mix of technical and behavioural skills. It’s not about learning statistics or the ins and outs of some software. It is, first and foremost, about learning how to read and consume data. I’ve seen Data Fluency proposed as a possible substitute, but in my view, Data Fluency is nothing more than a later stage in the Data Literacy journey.
I like to compare it with learning a new language, and so does Jordan Morrow in Be Data Literate. If I had to learn a language that represents its alphabet differently from what I know - say, Japanese, for example - I’d consider myself completely illiterate in that language. That doesn’t take away from the other languages I already know. But in this particular new one, I am illiterate - I can’t read or communicate with it. I’d have to learn what its symbols mean, start putting them together to derive some meaning (words), and then put those together and form sentences. I’ll only get to fluency in Japanese once I can read complete sentences and start using them to communicate my thoughts to others. There’s much more to being fluent than just learning how to read, though. The same goes for learning the language of data.
Another essential framework discussed in Be Data Literate is what the author calls the three C’s of Data Literacy. These non-technical skills permeate the entire journey towards becoming more fluent in the language of data: Curiosity, Creativity and Critical Thinking.
Be Data Literate’s core as a book is in bringing together the four levels of analytics (descriptive, diagnostic, predictive, prescriptive) alongside the four skills in the definition of Data Literacy (to read, to work with, to analyse and communicate with data), peppered with the three C’s (curiosity, creativity and critical thinking) as a recipe to follow and improve Data Literacy.
Our world is increasingly surrounded by data, and we often see glaringly bad examples of its uses. One does not need to learn a specific software or be well-versed in technical jargon to be data literate: it is about understanding the inner workings of what makes an analysis good and solid to deserve our trust. Data literacy matters because understanding our world and being able to make a well-informed assessment of it matters.
Data-Informed Decision-Making
Jordan Morrow presents us with one last framework, a step-by-step process for making data-informed decisions in a business context. This framework brings together all the other frameworks mentioned in the book. He walks us through six steps:
First, we need to ask questions. Your questions will relate to which of the four analytics levels we are in our journey.
Second, we need to acquire the data that will answer these questions. This is related to the concept of working with data, and the complexity of this step will be directly linked to the complexity of our questions. The further we are in the four analytics levels, the more complex acquiring data will be.
Third, we analyse the data. This is where the Critical Thinking of the three C’s framework shines.
Fourth, we integrate the analysis with our context, previous knowledge, and intuition and asses our biases and risks. This is when we bring it all together. An appropriate analysis only exists within a context.
Last, we decide. And we not only choose an option, but we take note of it and clearly communicate it across our peers because there is a 6th step to the framework:
We iterate! If we weren’t clear on some aspect of our process before, we should know better now, so we do better next. A truly data-informed culture learns from its mistakes and uses the power of hindsight to move forward.
How to not run a Data Literacy Initiative
I believe Data Literacy is the one big thing that sets apart companies that can truly drive positive change from those that scramble around trying to find a magic bullet that will solve all of their decision-making in one shot.
My first attempt at launching a Data Literacy programme did not go as intended, and culture was the main factor behind it. That experience did not stop me from trying, though.
I have learned a lot since then, and Be Data Literate by Jordan Morrow was part of that learning and self-reflection as well. Although I am now more aware of what the shortcomings were, I still stand by what I tried to do and if given the chance, I’d probably try it all over again - just better.
Looking back, with the power hindsight gives me, I was able to put together my own ten-step framework of how not to run a Data Literacy Programme within your company, which I share below. I hope it helps you ease the bumps in your journey:
Start small: Imagine trying to teach someone how to swim by throwing them into the deep end of the pool. Not the best idea. The same goes for a data literacy initiative. Start small, with one department or specific area where data can make a significant impact, and then expand from there.
Don’t ignore the “gut-feelers.” These are the folks who prefer to make decisions based on their intuition rather than data. Some are downright averse to using data altogether. Some of these will even be in leadership roles! Instead of trying to convert them, try to understand where they’re coming from and work with them to show that this doesn’t have to be a mutually exclusive relationship. Data and intuition can coexist!
Don’t assume that leadership buy-in is a given. Just because the leadership team is on board with the idea of a data literacy initiative, it doesn’t mean they’ll be fully supportive of it. Keep them informed of the program’s progress and benefits, and be prepared to address any concerns they may have.
Don’t forget about data governance. Make sure that everyone is aware of the rules and regulations surrounding the use of data and that there are processes in place to ensure that the data is accurate and reliable. A data literacy program can’t be successful if everyone is confused about what the correct sales numbers are because each team uses slightly different definitions. Document business logic, transformations done to the data, where the accurate data can be found and what defines it. Make sure there are guidelines that help build trust. Circulate those guidelines with the relevant teams.
Don’t neglect the impact of data siloes. Even if your data is accurate and reliable, it may be siloed in different departments, making it difficult to get a holistic view of the company. Be sure to address this issue and figure out a way to make the data accessible to everyone who needs it.
Don’t assume that democratising data will solve everything. Making data available to everyone in the company is great, but it’s not a magic bullet. Make sure that the people using the data understand how to use it properly and can interpret it correctly. See item 4 again. Democratisation without governance is a recipe for disaster.
Data Visualisation is much more than just pretty charts. Just because you have data doesn’t mean people will automatically understand it. And just because you put several charts on a screen doesn’t mean people will know how to read them. Data Viz is powerful in helping people adopt data in their daily processes, but it has to be thoughtful, carefully crafted, and support the people who consume the information so they can understand it.
Don’t ignore the importance of training. Onboarding new tools and technology matters. Ensure that everyone who needs to use the data is adequately trained on how to do so. This goes into tech training as well. Rolling out a fantastic tool like Tableau or Power BI seems easy, but your users may struggle with concepts such as interactive filters and buttons.
Don’t forget accountability. Ensure there are people responsible for keeping the data accurate and reliable. Name leaders that are accountable for KPI definitions and their results. Don’t confuse accountability with blame. A culture that instigates fear will avoid accountability like the plague. This erodes trust in the process and in the data.
Don’t forget the importance of management. A data literacy program is not just about the data. It’s about the people and the culture. Make sure that management is fully committed to the program and that they are leading by example. Don’t expect your peers to follow your initiative if their managers are not on board.
Implementing a data literacy program is a challenging task. It will be disruptive and demand effort from everyone involved. But it is very rewarding to see the results of a properly data-informed decision coming to fruition.
Data Literacy doesn’t happen in theory. It happens in practice. Be Data Literate by Jordan Morrow focuses on the practical advice I wish I had before I jumped into such an initiative head first.
Should you read it:
Be Data Literate - The Data Literacy Skills Everyone Needs to Succeed is a book of frameworks. It briefly goes over the concepts that make Data Literacy and then presents a series of steps anyone can take to help them get there. If you summarise the main ideas in Be Data Literate, you’ll end up with a concise framework to help you on your journey towards understanding and using data more effectively. The author knows his audience. He’s clearly talking to those in business who use data daily but are unsure how to be better with the challenges it throws at them.
The book doesn’t go into profound detail, nor it should - it serves as a primer to get you started with the subject of Data Literacy and then go your own way. One nice feature is that the author provides the sources and resources at the end of each chapter, making it easier to identify and explore them if a particular subject sparks more of your interest.
If you want an introduction to the concept of Data Literacy and how it ties into your journey to getting better at reading, working with, analysing and communicating with data, this book is an excellent recommendation. It is concise, quick and easy to follow while being packed with valuable information.
Always check your local library first to see if any of the books I recommend are available. If they’re not, consider donating a copy!
Get a copy at your local library | Amazon
If you subscribe to my monthly Newsletter, you’ll get a summary of all recommendations, plus more of my data viz musings.
You can also follow Data Rocks on LinkedIn
Comments