Are you doing it all Wrong?!

mediumThis post was originally published by Bruno Aziza at Medium [AI]

Data Governance, External/Internal Data & The State of AI

1 — A Show-Don’t-Tell Approach to Data Governance

“For many companies, data governance is the business equivalent of flossing. They know it’s good for them, but they’d rather be doing something — maybe anything — else”. Ha!

  1. Start with the Basics: The CDO & Data Council.
  2. Deliver Quick Wins to Demonstrate Value. You Have Selected the Right Use-Cases because they are (1) Visible to senior management, (2) High impact for the business (3) Don’t require overly complex remediation and (4) Your effort isn’t limited to a specific business area or function.
  3. Gradually Scale As You Improve.
Image for post

Did you know this?! “60% of respondents rate their data governance capabilities at various levels of underdevelopment”

  1. Make Sure the Business Side Drives the Transformation
  2. Focus on — and Facilitate — Change Management
  3. Define Clear Accountabilities (“CDO is accountable for KPIs associated with the maturity of capabilities while Data owners and stewards are accountable for key quality indicators for their respective data domains”)
  4. Create Communities

More here

2 — Why External Data Needs To Be Part of Your Data Strategy

“In most cases, you can’t build high-quality predictive models with just internal data.— Asif Mahammad Syed, the vice president of data strategy at the Hartford Steam Boiler Inspection and Insurance Co.

MIT Highlights 5 best practices that everyone can use.

  1. Know the external data landscape & know your options
  2. Use external data to know customers better
  3. Use external data to add real-world context to internal decision-making
  4. Use external data with care
  5. Remember that external data is only successful as part of a centralized digital strategy
Image for post

In a 2019 Deloitte survey, 92% of data analytics professionals said their companies needed to increase use of external data sources.

The piece is rich but there are 2 great nuggets in the article:

1- Best practice to observe: Create a “data-hunting team” that looks for data sources outside the company for a given use case, and then helps acquire, clean, and incorporate the data within the company. Success lies not just in the vision, it lies in the execution!

2- Greatest of all quotes: Data by itself doesn’t produce that intelligence. Data has to be put in the context of the business outcome to give you a valuable insight.”

3 — Why Data Engineering is so Important

“When a traditional company considers exploiting their data, the most efficient and first-step action should be improving the data engineering process.” — Lissie Mei, Data Scientist at Visa (via Trifacta’s post here)

Image for post

With Some Humor!

Image for post
Image for post

4 — The life-changing magic of making with ML

I’ve featured her work before but this blog contains all of Dale Markowitz’s hits! Follow her as she uses: the Perspective API to analyze text, taps into the Video Intelligence API to create a smart family video archive, showcases the power of Speech-to-Text, Translation, and Text-to-Speech to automagically translate and dub videos…all the way to using the Text-to-Speech API and AutoML tables create a PDF-to-audiobook converter.

There are all great but I’m a dad so this one is my favorite!

5 — Are Organizations Succeeding at AI & Machine Learning?

Interesting Report by RackSpace. The top causes for failure include lack of data quality (34%), lack of expertise within the organization (34%), lack of production ready data (31%), and poorly conceived strategy (31%).

  1. On average, companies are spending $1.06M on AI and machine learning initiatives.
  2. 31% identify data quality as a primary challenge to getting actionable insights out of AI and machine learning projects
  3. The most common ways that businesses report using AI and machine learning functionality are: as a component of data analytics (40%), a driver of innovation (38%) and through its application to embedded systems (35%).
Image for post

EXTRAS:

Python turned 30 this week!

Thanks Jeff Sternberg for the reminder and thanks Dustin Ingram for the work on the dataset!

Image for post

What is BRUNO?!

Image for post

Thanks Vincent for the share! 🙂 More here

Image for post

I hope you got value from this: please leave a comment & a like. Want to connect? I’m on LinkedIn @ linkedin.com/in/brunoaziza

Spread the word

This post was originally published by Bruno Aziza at Medium [AI]

Related posts