For Data Scientists & AI/ML Engineers — free blueprint + 10-minute self-check

The meeting ends with "very insightful." No decision is made. Nothing changes.

Most strong analyses stop one level below the one that gets acted on — and a 10-minute check shows if yours do.

By Richard Dobis — ex-senior director, led 50 data scientists, sat in the calibration rooms where promotions are decided.

The same analysis, delivered three ways:

  • Level 1 — Numbers: "The coefficient for in-person visits is 0.62. Accuracy 84%." → the room nods politely. Nothing happens.
  • Level 2 — Meaning: "A visit is worth six calls — but above eleven a year, the effect saturates." → "very insightful." Still no decision.
  • Level 3 — Recommendation: "Redirect visit capacity to the 340 high-value clients at five to eight visits." → a decision is made, with your name on it.

Same data. Same model. Same skill. Three different careers.

If you deliver well, hear "great work," and still watch bigger scope go elsewhere — the fix is not to become louder. It is to check which level your work actually reaches.

Check where your last three deliverables landed

The Output Value Ladder blueprint: the three levels in full, real before/after examples, and the 10-minute self-check.

Free. 15-minute read, 10-minute self-check. Use it in your next meeting.

By submitting, you agree to our Privacy Policy.

Who is telling you this

I heard "great work" and "not quite ready" in the same review — for years, while holding ML patents and an international AI award. So I know this problem from the inside.

I also know it from the other side of the table. I later became a senior director, led 50 data science professionals, ran projects directly for SVPs in a 50,000-person organization — and sat in the calibration meetings where promotions are actually decided. I watched strong technical people lose that room, and I learned exactly what the people in it need to hear.

That is what I mentor Data Scientists and AI/ML Engineers on now.

From people I've mentored

  • "Richard was really instrumental in shaping my career."
    — Pavel, Director of AI

The Ladder is yours to keep — and it works on its own.

Many readers apply it in their very next meeting and watch the room react differently. The model transfers completely; nothing is held back.

What no blueprint can carry is your situation. Your career runs on a whole chain — the work that reaches you, the level of your output, how its impact becomes visible, how trust turns into scope. The Ladder checks the most common break. If your weakest link is elsewhere, the self-check will show you that too — and finding it is a different job than reading about it, because it depends on your role, your manager, and the decision actually on your table.

The Ladder is one model from Impactful Analytics™ — the method I mentor with. Start with it. It checks the most common break first.

Free. 15-minute read, 10-minute self-check. Your next meeting is the test.

Richard Dobis, mentor for data scientists and AI/ML engineers on career growth

This site is not a part of the Facebook™ website or Facebook™ Inc. Additionally, this site is NOT endorsed by Facebook™ in any way. FACEBOOK™ is a trademark of FACEBOOK™, Inc.

 Privacy Policy 

richard.dobis@dobismentoring.com