Analytica - visionary modeling

10/15/2026

AI can give you an answer in seconds.

The awkward part comes when someone asks, “Why?”

Generative AI is very good at producing answers that sound convincing.

But important decisions need more than a convincing answer.

What assumptions were made?

What evidence supports them?

What happens if those assumptions change?

How uncertain is the outcome?

Can another person inspect the reasoning?

That's why we're interested in AI-assisted decision making, not AI making the decision for you.

Analytica combines decision modeling with Assista, an AI collaborator designed to help you work with your model.

The model provides something a chatbot response can't:

A visible structure you can inspect, challenge and change.

AI can help with the thinking.

The decision model shows its work.

Don't just ask AI for an answer.

Build a decision you can explain.

Meet Analytica + Assista:
https://vist.ly/5nhic

10/14/2026

Sometimes the hardest part of a complex decision isn’t building the model.

It’s figuring out what you should be modeling in the first place.

Which uncertainties matter?

Which assumptions need to be challenged?

What data is actually useful?

And how do you turn all of that into a decision your organization can act on?

That’s where decision analysis consulting can make the difference.

At Lumina, our work goes beyond Analytica software.

We help organizations approach complex decisions with greater structure and clarity—bringing together decision analysis, uncertainty analysis, risk analysis and decision modeling to tackle problems where the answer isn’t obvious.

Because sometimes you need the right tool.

And sometimes you need experienced people helping you ask the right questions.

Facing a complex decision? Let’s talk about how Lumina can help.

We can develop a custom decision model tailored to your organization's needs or train your consultants to create it themselves.

Book a time with James - https://vist.ly/5ncgb
Book a time with Max - https://vist.ly/5ncgd
Book a time with Karen - https://vist.ly/5ncgf

10/12/2026

What if you could understand decision modeling without needing a PhD, a statistics textbook, or 14 open spreadsheets?

That’s the idea behind a new series from exploring modeling, probability, uncertainty, and AI in a way that’s made for the rest of us.

The first four episodes tackle some surprisingly big questions:

→ What is a model?
→ Do models still matter in the age of AI?
→ What exactly is a probability?
→ The Hidden Accomplice

No heavy sales pitch. No Analytica deep dive.

Just interesting ideas about how we understand uncertainty and make better decisions.

🗓️The first episode drops November 3, with a new episode every week.

🎬 The trailer is live now: https://vist.ly/5mzvh

10/08/2026

Can you trust AI advice if you can’t see how it reached the answer?

In this webinar, and explore a different approach to AI-assisted decision making: building a transparent decision model that humans can inspect, challenge, and improve.

Using Assista with Analytica, they demonstrate how AI can help structure models, explore uncertainty, run sensitivity analysis, and uncover insights—while keeping the assumptions and reasoning visible.

They also put ChatGPT and Assista side-by-side on the same business decision to see what happens.

🎥 Watch the full webinar on YouTube ⬇️

How can you improve decision making in your organization by collabo...

10/07/2026

The worst time to discover a bad assumption is after you've made the decision.

Most complex decisions aren't difficult because you don't have enough data.

They're difficult because you don't know exactly what will happen next.

Costs change.
Demand changes.
Timelines move.
Forecasts miss.

And every model contains assumptions.

The question is whether you can see which assumptions actually matter before you commit to a decision.

That's where decision analysis and uncertainty analysis change the conversation.

Instead of asking:

“What does our model predict?”

You can ask:

“What happens to our decision when our assumptions are wrong?”

Analytica is decision modeling software designed to help analysts model uncertainty, explore risk and compare possible outcomes visually.

Because confidence shouldn't come from pretending uncertainty doesn't exist.

It should come from understanding it.

Ready to make uncertainty part of the analysis instead of something buried in a footnote?

Explore decision modeling with Analytica:
https://vist.ly/5md82

10/05/2026

🧑‍🏫The best teachers don’t just teach us what to think.
They teach us how to think.

How to question assumptions.
How to look at a problem from another angle.
How to work with uncertainty instead of ignoring it.
And how to make better decisions with the information we have.

Those skills extend far beyond the classroom.

Analytica is used in universities to help students explore decision analysis, decision modeling, risk and uncertainty, Monte Carlo simulation, and complex problem-solving.

But behind every model, every analysis, and every student learning to ask “What if? is a teacher who helped them think a little differently.

This World Teachers’ Day, we're celebrating the educators preparing the next generation of analysts, researchers, scientists, engineers, and decision-makers.

Thank you for teaching people not just to find the answer—

but to ask better questions.

Happy World Teachers’ Day.

Are you using Analytica in the classroom? We’d love to hear what you and your students are modeling.

Tell us in the comments or tag your university.

https://vist.ly/5m363

Discover how leading universities worldwide use Analytica for quantitative modeling and decision analysis. Explore our discounted rates.

09/30/2026

Have you actually tried Analytica yet? 👀

We've spent this month talking about complex modeling, Monte Carlo simulation, influence diagrams, risk analysis, and some of the newest features in Analytica 7.1.

So, we're curious where you are in your Analytica journey.

And if you're in the -planning to try it- camp, here's your nudge:

You can get started for free. https://vist.ly/5kfc2

09/29/2026

You shouldn't have to introduce your AI to your model every Monday morning.

Analytica 7.1 now saves your Assista conversations locally and associates them with your model.

Come back later and you can reopen an earlier conversation and continue working.

That's particularly useful when modeling happens the way it usually does in real life:

Over days.

After meetings.

Between revisions.

And following the inevitable "one more scenario" request.

Your conversation can evolve alongside the model.

❓Would persistent AI conversations change how you use an assistant while modeling?

09/24/2026

Your risk model is only as useful as the assumptions inside it.

A beautifully built model can still give you misleading results if the assumptions underneath it aren't realistic.

That's why good risk modeling involves more than entering data and calculating an answer.

You need to be able to see your assumptions, question them and understand what happens when they're wrong.

Analytica lets you represent uncertainty directly in a risk model and explore how different assumptions affect possible outcomes.

You can run scenarios.

Perform Monte Carlo simulation.

Explore sensitivity.

Identify which uncertainties have the greatest influence on the result.

And because the model is visual, you can trace those relationships back through the decision.

The goal isn't to eliminate uncertainty.

It's to understand enough about it to make a better decision.

Which assumptions in your risk models deserve a second look?

09/23/2026

Your next decision model might already exist.

It could be sitting in a PDF, spreadsheet, Word document or report.

With Analytica 7.1, you can bring those files into Assista and start working with the information directly.

👉Extract a table from a PDF.

👉Bring spreadsheet data into your model.

👉Use information from a report.

👉Even work from a model described in a document.

For analysts who spend a surprising amount of time moving information from one place to another, this can remove a lot of the boring part before the interesting work begins.

What file would you hand Assista first?