Skip to content

Analytics: The Mixed Bag

Mr Ben
Mr Ben
5 min read
Analytics: The Mixed Bag

Analytics has a branding problem.

The industry is stretched thin. The careers blur org lines. The craft has a cluttered workbench. The results are muddled.

Yet, there is no place I would rather be.

Analytics: the industry

You will not find “Analytics” in a SIC code search. It does not fit cleanly into classic math, computer science, business, or finance degrees, but it still does not stand out enough to warrant its own major.

Look closer though - every organization requires analytics. It likely goes by a different name.

  • Knowledge Discovery
  • Data Mining
  • Big Data
  • Big Data 2.0
  • Business Intelligence
  • Machine Learning
  • Data Science
  • Data Engineering
  • Growth

The names are buzz-worthy for a time, then fade after failing to meet inflated expectations. But faster than Uncle Sam can mint a roll of dimes, a new term is coined.

The same concept’s recurrence, with different names, hints that the core issue remains unresolved. The need is still there.

At the root of analytics - no matter what we call it - is the need to create an accurate, shared understanding of our world. Developing that understanding is a fun and impactful challenge to meet.

Analytics: the career

No one starts in data. No one leaves data. Analytics is essentially the Hotel California of careers.

So far, my journey has been:

IT College Grad → Founder → Programmer Analyst → Business Analyst → Data Scientist → Analytics Engineer

Being a founder was too ambitious out of the gate. I did not have the knowledge or systems in place to make the business a success. The failure was a sharp reality check. I still needed to understand how the business pieces snapped together. Running a company gave me the outline, but analytics has given me the color.

Once you find your way into a data role, you get visibility into a wide range of topics. You need exposure to all these dots because your job is to help connect them for others. You may be the lowest person on the totem pole in the room, but a career in analytics gets you in the room where it happens.

What you do next with all the organizational knowledge and leader relationships is up to you. There is no clear career path out of analytics. But when you decide what your next step is, your time as an analyst will leave you uniquely positioned to succeed.

Analytics: the craft

I like the idea that tools define a craftsman. When I say “carpenter,” images of hammers, saws, and planers come to mind. Tools are so recognizable in a craft that our emojis brought their tools along for digital career day.

<span class=“text-5xl”>🕵️‍♀️ 👨🏻‍🌾 👨🏻‍🏫 👩🏽‍🔬 👨🏾‍🚒 👩🏻‍⚖️ 🧙🏼‍♂️</span>

You can see the detective has their magnifying glass, the teacher with their chalk, and the wizard with his staff.

So when I say “analytics professional,” what tools come to mind?

<span class=“text-5xl”>👨🏻‍💻</span

Okay - a computer is a good start. But let’s turn that screen around.

WHOAH! That’s a lot of tools. If you cannot borrow the magnifying glass from our emoji detective friend, there is a simpler way to think about it. What are the analytic jobs to be done?

At a minimum, you must collect data, format it for reporting, find the insight, and communicate the findings. It is possible to do all this in Excel, but please … don’t.

When we focus our tools on delivering those four steps reliably, the tools begin to click into place. My stack looks like this:

Fivetran for collecting data. dbt for formatting data. Mode for finding and communicating results.

Your analytics stack will never be perfect. As trouble areas improve, new desires emerge. As Tristan Handy put it - we are pushing the bubble under the screen protector to the edge.

Analytics: the results

If you are still with me, you have persevered and have results you are excited to share with others. All the work comes to a head. Then it falls flat with your audience, or they ask for a “quick update,” or they accept your results but move on without a thank you.

Communicating and visualizing results is a skill. We all get better at this with reps and patience with those we help. The underlying work is going to be invisible to stakeholders. But if your results inform someone’s decision without exposing the details to them, that is a sign of expertise.

I think a slight reimagining of the “Handyman Homily” gets this point across well.

A startup founder hit on product-market fit. Their company was in hyper-growth! Scaling at breakneck speed. The old processes couldn’t keep up, and the leadership team was flying blind. On top of it all, the founder was courting investors for their series A. The team needed analytics.

The founder put out a call for help on Twitter, and luckily an analytics engineer responded in a DM that they could help. The next day the founder asked the engineer to pull a report with customer count and MRR over time. The engineer wrote:

SELECT reporting_month,
	COUNT(DISTINCT customer_id) AS customer_count,
	SUM(mrr) AS mrr
FROM crm.transactions

Then out popped a track record of the company’s meteoric rise. The leadership team rejoiced, and the investor’s eyes turned to dollar signs 🤑. The founder profusely thanked the analytics engineer and asked how much they owed the engineer. Our analytic protagonist calmly responded half-a-million dollars. The founder threw their hands back and said, “no way! That only took you 10 minutes if that!”. Again, calmly and cooly, the analytics engineer responded, “It did not take 10 minutes. It took my whole life.”

Analytics: the mixed bag

So you get it. Analytics is a messy place. What we call ourselves will change in 6-months. We don’t have a career plan for you. Our tools do not always work together. Then all the work you put in is hard to see.

For all these drawbacks of analytics, it has more upside to offer. Companies will always need you because you are adding to their senses. You get to be challenged intellectually. You sit in an exciting role where you work with computers, and humans, taking advantage of the best form of leverage - permissionless leverage. Your code - SQL or Python - codifies an understanding of the world you developed. Your communication - PowerPoints or memos - shares that new understanding with others.

When you are aware of the downsides of analytics, you can adjust your approach to turn it into an upside. That’s taking analytics: the mixed bag and turning it into analytics: the goodie bag.


Mr Ben Twitter

Data 👨🏻‍💻 • Fam 👫🏻• Tacos 🌮


Related Posts

Members Public

Writing DAG

Sometimes writing is hard and other times it's downright impossible. Stack the odds in your favor with a writing system. Writing is not one skill. It's more than stringing words together. You need to research, outline, draft, edit, edit, edit, and share. A common failure mode is attempting to cram

Writing DAG
Members Public

One-Way or Two-Way Doors

Benn Stancil wrote about measuring analytics success [] by speed of decisions made - not the outcome of the decision. Tristan Handy followed up in the Analytics Engineering Roundup [] like a true analyst. He added a caveat. Tristan cited

One-Way or Two-Way Doors
Members Public

DPIM Framework: Bundling Better Infra, People, Models, and Data

I cannot stop thinking about this tweet.

It has been tumbling around

DPIM Framework: Bundling Better Infra, People, Models, and Data