Holi, Soy Anita. 

Here’s how most of us pick a new AI tool for the classroom : we see it do something impressive, generate a quiz in 3 seconds, turn a paragraph into a slick slide deck, translate a worksheet into 5 languages instantly, and then we go looking for a reason to use it. What first, Why, if it comes up at all, come last, as a justification rather than a starting point. 

It’s such a natural order that most people never notice it’s backwards. 

Quick answer, if you’re skimming: the order most people use to choose a tool is What (what can it do) - How ( how do I use it) - Why ( why did I even want this). Flip it. Start with Why, the actual learning outcome you’re chasing then How you’ll get students there, and only then What tool fits. Tools chosen in that order get used past week one. Tools chosen in that order get used past week one. Tools chosen the usual way get bookmarked and forgotten. 

The order almost nobody uses

There's a well-known communication framework ( 3 concentric layers, usually drawn as rings) that separates why something exists, from how it operates, from what it actually produces. The reason it gets taught so often isn’t the diagram. It’s the observation underneath it : most people describe their work starting from the outside ring and working in. What we make. How we make it. Why do we bother? 

Flip that order, and something interesting happens to your decision-making. 

Applying it to your next AI tool

Why - before you open a single tool, name the actual learning outcome. Not “I want to save time” ( It’s also real…. But it’s not a Why. It's a side effect.) What should a student be able to do differently after this lesson than before it? This is the same idea from an earlier Book Notes post : design comes before tools, not after. Skipping this step is exactly how a genuinely well-made tool ends up solving a problem you don’t actually have. 

How - once you know the outcome, decide the approach. Will students get there through practice and repetition? Through feedback and revision? Through something more exploratory, where they’re allowed to get it wrong a few times first? This is still not about software. It’s more about the shape of the learning experience you’re building toward. 

What - only now do you go looking for a tool. And notice what happens: the search gets dramatically narrower. You’re not scrolling through “50 best AI tools for teachers” hoping something jumps out. You’re looking for one specific thing that serves an outcome and an approach you already decide on, before you ever knew its name. 

Reverse the order - What first, Why maybe never 

You would get a very different outcome. You get a tool you’re excited about for a week, a login you forget by month two, and a “why did I sign up for this” moment when the renewal email arrives. 

Where this framework actually came from, for me

I didn’t pick this discipline up for the sake of a tiny blog metaphor. It’s the actual thread running under my own “why” for being in this field at all. I want to use technology and learning design to make education more meaningful, accessible, and future-ready. That connects a classroom I taught in across four countries to a partnerships and product marketing seat in tech..because underneath both, I keep circling back to the same question : how do you design an experience that actually changes what someone can do afterward? 

Start with Why isn’t something I borrowed from one world and awkwardly bolted onto the other. It’s the same question, asked in 2 different rooms. 

If you’ve ever felt a little sheepish about a subscription you can’t quite remember why you got it, that's probably because you make choices with the wrong order. 

I’m building out a fuller decision framework around this, including how I evaluate vendors and tools for my own work for both the classrooms and the corporations. Over in the newsletter with the Shine Lab. If flipping this order changed how you’re about to choose your next tool, that’s where the rest of the thinking lives. 

Subscribe to keep thinking this through with me

— Anita