Hi Reader,

The conversation around AI has taken us backwards.
Let me explain.

There are 3 ways to think about your work:

1. Inputs: the things you use to do the work – tools, platforms, resources, AI models
2. Outputs: the things you build – pipelines, dashboards, data models, reports
3. Outcomes: the things that change in the business because of the work – lower churn, higher revenue, faster decisions, better customer experience

For years, I’ve argued that most data teams spend too much time thinking about outputs, and not enough time thinking about outcomes. Now AI has pushed many companies even further back.

Instead of asking:

They ask:

That’s an input question.
When companies start with the input, they often miss the real problem.

In my latest Helicopter Data Podcast episode, I spoke with Juan Sequeda, Principal Data Strategist at ServiceNow, about this exact issue.

Juan is one of the leading voices on knowledge graphs, semantic layers, and ontologies.
He explains why AI doesn’t just need more data:

• It needs context
• It needs meaning
• It needs to understand how business concepts relate to each other

That’s where knowledge graphs become really interesting.
This conversation helped me think differently about what companies actually need to make AI useful – not just impressive in a demo.

Watch the full talk here 👇

Why AI Needs Knowledge Graphs [36:31] Data without knowledge can be misleading and sometimes even useless. Join me and Juan to learn how to make AI useful for the business

Have a great weekend!

Regards,
Shachar

P.S. Are you serious about growing your data career? Check out my website.

♻️ Please share these links with 2 colleagues or friends to help them unblock their career:

• ​Career Unblocking Mini-Course and Community →
• ​My Website (how I work) →
• ​YouTube Channel →