#24 What my first data analytics job actually taught me

Hello!

I’m James Janson Young and I convene Ours For The Making. Here I explore the ideas, concepts and innovations that are shaping our future.

👉You can subscribe to the accompanying YouTube channel Ours For The Making  here . 

What my 1st data analytics job taught me

What did my first data analytics job actually teach me? In particular, about how to have a successful career as an analyst?

Well, it revealed that there are three levels to being a successful analyst beyond being technically proficient. 

The trouble is, in my experience I’ve observed that most analysts only ever get to level two. Overlook these three levels and your work either doesn’t get off the ground, or worse, you ability and hard work get ignored. Over time, you get handed fewer and fewer interesting projects and start falling behind. So, let’s avoid that.

To explain these mysterious levels, I want to tell you the story starting at Level One. 

👀 Level One: keep your eyes open

I want to take you back in time. 

It’s 2008. And I have a problem: I’m not really sure what I’m getting myself into.

When I landed my first data analytical role, the assignment didn’t have “data analytics” written on it. It was 2008. Data analytics wasn’t really a thing. 

I started the project as a kind of general analytical/research capability for a small project team. (I.e. a bit of a general dogsbody)

And, if I’m completely honest, I didn’t fully appreciate the ‘data’ potential of the assignment beyond preparing some general, descriptive statistics.

But then something arrived in my inbox from the client and my view of what was analytically possible was about to be transformed…

Eventually – once I could open the file which initially exceeded the size of my inbox.

As I scanned through the raw data and digested it a little, I spotted an opportunity to have a play with the data and see what was possible. And maybe, this was a chance to build my analytical skills…

I decided to build… a model.

There weren’t really any models lying around for me to look at and apply. So, I returned to old maths text books and chatted to likeminded colleagues (more on that in a moment).

I opened up an excel spreadsheet. And got started. 

And this is Level One in a nutshell: Keep your eye’s open. You don’t need to wait for the perfect data modelling opportunity, have a job title “Data analyst” or access to the whizziest software. 

Spot the problems lurking in plain sight that are crying out for a data analytical fix and try stuff out, using the tools you have to hand.

But, although showing curiosity and initiative is all good stuff, it’s not enough. As I was about to discover. 

In order to be successful, I found that I needed to move up to Level Two… 

This was because, having taken a chance with my analytical approach, I quickly encountered another problem. But, not with the data. Or even Excel.

You see, using data to build a model that would churn out numbers to be featured in a published report was something a bit new and untested. The stakes were pretty high.

Colleagues were understandably cautious. 

Data modelling suffers from a “black-box” syndrome where the numbers that magically pop out of models give few (i.e. no) clues to the recipient about how those numbers were generated, and whether they’re valid or robust. 

Don’t assume people presented with your work will ‘get-it’. 

In fact, assume, and prepare for, the opposite.

I needed a way to answer legitimate questions and concerns. 

So I pursued an approach I now call “Mutually Assured Construction”. This brings us up to Level Two.

🏗️ Level Two: “Mutually Assured Construction”

“Mutually Assured Construction” has two governing principles. But, both need to be applied in order for it to be successful.

• Principle 1: “Test internally”

I connected with a few likeminded colleagues interested in the data modelling space. 

We formed a loose analytical group that reviewed and critiqued the model at various stages of its development. Its assumptions, logic and architecture. 

• Principle 2: “Built in public with the client”

The model’s assumptions and architecture were shared with the client so it got to see the inner workings and provide feedback.

This bi-pronged approach was critical to persuading colleagues and the client of the validity and value of this kind of work. 

This might sound obvious, but too frequently problems that arise with analysis can be traced back to simply not talking with the right people, early enough and often enough, inside and outside your organisation. 

And, most analysts stop here. And they do good solid work on a project by project basis.

Good job. Well done. 

But, by stopping here, they’re missing out on a way to really strap rocket boosters to their careers.

You see, up until this point, Levels One and Two, we’ve essentially been talking about “Project Thinking” – principles that you can apply to your experience in individual projects.

Yes, a project is arguably the fastest way to learn and sharpen analytical skills.

But for those of you who’d like to progress their careers further and faster, there’s a whole other level to explore…

🔧 Level Three: build your profile

Level Three represents the most important lesson I learnt from my first data analytics job. In order to open up new, wider opportunities, you need to maximise your project experience.

How?

By seeking out and teaming up with likeminded data nerds, you can contribute to something much bigger. By building a platform to support and showcase your organisation’s capability and capacity to do more and better data analytical work.

And being a part of building that platform can do wonders for your profile. Which, in turn, can open up all sorts of future opportunities…

For example, the more your colleagues hear about what you can do for them, the kinds of problems you can solve, the kind of value you can add, you won’t have to go looking for analytical projects. They’ll come to you.

Level Three Profile Thinking is how you leverage your project experience to build forwards towards your next analytical opportunity by building outwards across your organisation, and beyond.

Of course, the role of the analyst continues to be in flux, but there is a clutch of skills that are essential for any (data) analyst to continue to be successful.

The trouble is, that too often, they are overlooked by analysts. I’ll be talking more about them in the next newsletter. 

Articles of curiosity

🎹  The death of the key change  – this analysis of the 1143 songs that topped the Billboard Hot 100 between 1958 and 2022 reveal something curious about the structure of music and its decline in 1991 to almost zero %. It chimes with what I discuss in  What we get wrong about the future of music  and how streaming platforms are serving up more of the same to us. For instance, a whopping 7 in 10 songs streamed on Spotify in 2017 were from playlists – either editorially curated or machine-generated. A ‘great musical flattening’ could be taking place before our very, err, ears.

ICYMI 

🤔 Need help to decide what your next strategic step should be? Well, here’s a quick way to spruce up an old familiar concept that’s fallen out of fashion in a really quick way to make it way more invigorating – and useful. Find out how here in the latest edition of the Quick Concept Series  here . 

🗞️You can read all previous issues of this newsletter (The Makers) that explore how we enage with the future  here  (scroll to bottom of the page).

Thanks 🙏

Thanks for reading, watching, subscribing and being a Maker. I really appreciate it.

This is the first in the new direction, so, please to hit reply and let me know what you think. You know what they say: “those who show up; shape it.” 

Or, something a bit like that.

If you’ve enjoyed this edition of The Makers, you’d be doing me a kind and generous favour by sharing it with someone who might enjoy it also. 

And if you have questions or comments, do hit reply. 

Until next time…

Best wishes,

James