Hello there,

​
Ever feel like you're following all the "best practices" but still not getting the results you want?

You're not alone.

Here's the thing: when it comes to data work, "best practices" can be misleading, and sometimes even a trap.

Let's dive into why, and more importantly, what you can do about it.

​

The Problem with Best Practices 🚫

Best practices are like hand-me-down clothes.

Sometimes they fit perfectly, but often they need some serious tailoring.
​
In the data world, here's where they tend to fall short:

​

1. They ignore context

Your company isn't Microsoft, Google, Amazon or

​

2. They become outdated

Technologies evolve, regulations shift, and circumstances change. Yesterday's "standard" could be today's obsolete approach.

​
3. They create a false sense of security

Following best practices to the letter doesn't guarantee success. It might just mean you're really good at following instructions, not solving problems.

​

The Real-World Approach: Your Data Strategy Playbook 📘

​

Instead of blindly following best practices, let's evaluate the real world and be realistic.

Here's your actionable game plan:

1. Understand your unique situation

• Assess your data: Is it messy, incomplete, massive, or all of the above?
• Evaluate your team: What skills do you have at your disposal?
• Clarify organizational needs: What problems are you actually trying to solve?

Pro tip:

Create a "Status Sheet" that outlines these factors.

Update it quarterly to stay on top of changes.

​

2. Experiment and iterate

• Start small: Choose one process to experiment with.
• Test and learn: Try different approaches and document the results.
• Be ready to pivot: If something's not working, don't be afraid to change course.
  ​

Here's something to try:

Create two or three versions of a key dashboard - one strictly following best practices, and the other two not so much (but designed based on how you think your stakeholders will use the data).

For example one could be radically simplified, and another one highly interactive (with text tables - gasp!).

Share these with different stakeholder groups and track which one gets the most engagement, use and positive feedback over a month.

Pro Tip:

Create an "Experiment Log" template to consistently track your tests. Include fields for assumptions, hypothesis, methodology, results, and lessons learned.

​

3. Focus on outcomes, not processes

Evaluate what works for your organization from the results and impact:

• Track usage: Are people actually using your dashboards and reports?
• Measure impact: Are decisions being made based on your analysis?
• Gather feedback: Regularly check in with stakeholders about the value they're getting.
  ​

Pro tip:

Set up a simple feedback system. It could be as basic as a monthly "data impact" survey or as sophisticated as analytics on your BI tools.

​
Real-World Best Practice Showdowns 🥊

​
Here are some common advice you will hear (or read) that need further evaluation and investigation.

That is, do not follow them blindly.

​
1. Advice: "Always use extracts for better performance in Tableau"

Reality Check:

✅ Great for: Smaller, infrequently changing datasets

❌ Nightmare for: Real-time data needing constant updates

🤔 Questionable for: Already optimized, fast data sources

​

2. Advice: "Use Common Table Expressions (CTEs) for complex queries"

Reality Check:

✅ Excellent for: Improving code readability and handling recursive data structures

❌ Problematic for: Queries where performance is critical

🤔 Consider alternatives: Temp tables or restructured queries for speed-critical operations

​

The Takeaway: Your Data, Your Rules

Best practices aren't useless – they're valuable starting points.

But remember:

• They're guidelines, not commandments
• Always consider your specific context
• Be ready to adapt or even discard them when necessary

​

The true best practice?

It's the one that works for you and your unique situation.

​

That's it for this week.

Stay curious, keep experimenting, and trust your judgment.

You've got this!

​

Until next time,

Donabel

​

P.S. What's your experience with best practices?

Hit reply and share your data adventure stories.

Your insights could be the game-changer someone else needs! 📊💡

​
​