Companies need a technology turnaround right now and that's a huge opportunity for mid to senior Data Scientists and leaders.
Playing a key role in a turnaround is a career maker. I've been part of 4 and here are some lessons learned.
1/12 #DataScience#MachineLearning#Strategy
Companies only change when they've been through enough pain. That low point and the months immediately after it are where Data Scientists can put forward ideas that will gain traction. Don't push the elephant. Let it lead.
2/12 #DataScience#MachineLearning#Strategy
Companies usually plan without enough information to build a good plan. Data Scientists can help the business understand the possibilities created by our work. That knowledge is critical from the earliest stages.
3/12 #DataScience#MachineLearning#Strategy
Data is a critical success factor for planning. The business must learn about itself, its customers, and the marketplace. Data will provide insights into all 3 and give Data Scientists a seat at the table.
4/12 #DataScience#MachineLearning#Strategy
Quick wins are essential. Businesses need to show investors and customers that their turnaround plan is working from the earliest stages. Deliver value quarterly but setup for longer term projects.
5/12 #DataScience#MachineLearning#Strategy
Run 2 cost savings and 1 revenue generating Data Science project per quarter if you have the resources. Deliver those incrementally over the course of a year if you don't, but deliver results every quarter.
6/12 #DataScience#MachineLearning#Strategy
That means breaking projects into parts where each deliverable fits into a quarter and produces revenue or cost savings while working towards the larger, end of year goal.
7/12 #DataScience#MachineLearning#Strategy
The Data and Analytics org does not run the business. Senior leadership decides the direction. The D&A org must move forward with the business not try to force it in a different direction.
8/12 #DataScience#MachineLearning#Strategy
Even if the plan's a mess, the goal is to deliver better results. Whatever the D&A org is given, the job is to make it better. Provide better options supported by data and trust leadership to continuously improve the plan.
9/12 #DataScience#MachineLearning#Strategy
This process builds trust in the D&A org. Senior leadership should see the team as a lever for free cash flow and margin preservation. Once they do, relentlessly deliver the next 2 quarters. Then present a long term vision.
10/12 #DataScience#MachineLearning#Strategy
Senior leaders trust results. Once the D&A org establishes a track record senior leaders are ready to put more of the business's growth into the D&A org's hands. We can't talk long term without trust.
11/12 #DataScience#MachineLearning#Strategy
That's what being a difference maker during a turnaround means for a Data Scientist or Data Science leader. This formula sets us up to succeed and drive results. That's a career maker.
12/12 #DataScience#MachineLearning#Strategy
• • •
Missing some Tweet in this thread? You can try to
force a refresh
The Data Science learning path today is different than it was 3 years ago and looks nothing like it did 7 years ago. This thread has the main layers and example resources covering the basics, assuming you've got basic math covered.
1/18 #DataScience#MachineLearning
1. Research Methods. We do a lot of research and experimentation now. Data Scientists used to be model-centric but that's changed because our work must meet higher reliability requirements. I wrote an intro post: vinvashishta.substack.com/p/a-basic-intr…
2/18 #DataScience#MachineLearning
2. Causal Inference. Data Science has taken a hard turn towards causal inference, again to meet increasing model reliability requirements. An education on CI always starts with Pearl. ftp.cs.ucla.edu/pub/stat_ser/r…
3/18 #DataScience#MachineLearning
The only way Data Science continues to move forward is with an equal focus on research and applications. To generate value, models must perform reliably in production over several years.
For even mid-level Data Scientists, total compensation starts at $250K+. It goes up to $400K. ML Engineers can cost even more depending on their breadth of platform/architecture knowledge and ability to deploy models at scale.
Machine Learning is working to improve computational methods to discover causal relationships in massive datasets. The line of research is emerging, and we are in the earliest stages of exploration.
Our biggest challenge is, as always, data. Massive datasets are dirty and difficult to completely clean. The biases and assumptions baked into their gathering methodology add another challenge, also complicated by the dataset's size.
List out your accomplishments over the last year. Talk in terms of team value and business value. List specific accomplishments.
Put a checkmark next to the ones your manager is aware of. Put a star next to the ones upper-level management is aware of.
2/13
Schedule a meeting with your manager. Don't just drop by because your manager will feel ambushed. Give them time to prepare and think through how they will respond.
3/13
Platforms like Substack will replace the book publication route for sharing expert knowledge. Let me explain:
I built my Substack after being offered several publishing deals because it allows me to deliver value more efficiently.
Your favorite authors have a few niche areas of expertise, so a newsletter format gives you access to a broader range of topics for the same price.
Newsletters are better for rapidly moving fields like technology. My posts are living documents, and I control the update frequency. Books update annually at best.