My two-year-old cousin, Grayson, has a wooden train set. It has the typical variety of track segments: long and short straight sections, curves, bridges, y’s and x’s. You can build awesome tracks, like a figure-8 with bridges. But Grayson doesn’t. He assembles the track piece by piece, in whatever order the pieces end up in his hands. His train yards are chaotic, like LA after Sharknado.

Too often we approach problems like two year olds. We take what we have (the track in my hand), and try to make something of it. Without a vision for the finished product, you’ll end up with shit (and it’s only cute if you’re a two year old). You’ll also waste a lot of time. Competent designers say, “I want to build that” – and then gather the right set of pieces.

Ever look at a boarding pass? What is all that data? It’s nuts. Maybe there’s a good reason for it (check google). The evolution of boarding passes has been to retain and not reinvent.

UX in cars. What UX? If cars were a SaaS product, no one would buy them.

Think of anything antiquated – it’s because the engineer started with what he or she had – and it was typically yesterday’s tools, yesterday’s ingredients. So he built yesterday’s product.

In my role managing marketing programs for VMTurbo, I faced a battle with the operations team who maintained the website and other marketing technology tools.

Me: “I need to understand how people get to our website”

Ops: “No problem, Marketo cookies people when they first hit the site”

Me: “But what if they don’t convert when they originally hit the site? How do I see the source of each visit, including the original and the one they converted on?”

Ops: “Well Marketo doesn’t do that. So we can’t answer that.”

Wrong answer. You’re starting with the ingredients: Marketo. So you’re myopic. You can’t see beyond what you have. You need to see the goal – which is multi-touch attribution – and then find the right tools to make that happen.

On the other extreme, you don’t want to reinvent the wheel every time. The Ops team was right – we did have a tool (and an expensive one) – so could we make that work? No sense reinventing Marketo functionality.

I’m not advocating that Greyson grow his own trees to make his own train tracks. The single biggest driver of human innovation is reusable components, which has led to specialization within supply chains, and that specialization has led to declining costs and higher quality. In software, it’s API services (or, more antiquated, libraries). In apparel, it’s separation between cotton farms, weavers, designers, manufacturers, and retailers (and there are obvious counter examples). But these supply chains mean you’re contributing only a piece of the value creation. You start with X and turn it into Y. This system works because the final product is agreed upon.

New solutions require new ingredients. Don’t hold yourself back. Think outside the box; start with the vision and not the ingredients.