Do it now, while the need is real

Most of us know the feeling. The list is long: emails to send, forms to fill in, requests to raise, some research into what happened last time. None of it is hard. It's administrative, and it isn't the part of the job where the spark happens.

People tend to split in two when they hit this. Some think, that's just how it's done. That isn't a fault. It's familiar, it's proven and it's safe. Others think, there has to be a better way, and they tinker. They fumble something together that helps a little.

In both cases I see the same gap. People either accept the work as it is, or they can almost touch a better way and don't know how to get there. So they carry on.

Here's one I lived. At Unity we needed assets from customers for our channels, and they came with conditions: permission, approval, the right formats. That meant outreach to a studio, a few meetings and emails about the use case, the dimensions and the formats. Then legal needed a waiver signed. The customer's legal team reviewed it and sent back redlines, ours reviewed those, and it went back and forth. Then the assets arrived in the wrong format, or we needed a different type, and the cycle started again. Even then, nobody was sure where the assets lived or who to ask when something went wrong. It had run that way for years because that's how it was done.

Meanwhile we had good tools: Airtable and Brandfolder, which is built for assets. I was looking at 140 requests in a short window, and I knew I didn't want that headache. So I broke it down. What do we need from a studio? A meeting or an email could get it, but the answer was an intake form.

The form told customers the file types, dimensions and rules, and gave them control over how their content was displayed and how long we could use it before asking again. I worked with legal, and we cut the waiver out as a separate document and built it into the form as a checkbox. Assets went into Brandfolder with clear instructions, so any team could see what was allowed. Legal had a log of who signed, who requested and who to contact. Three to four weeks became under an hour.

It wasn't hard to build. What was hard was seeing it, because everyone is under pressure to just get the work done.

I've heard the answer many times: "Great idea, but we don't have time right now." I think that's backwards. The real need is the only moment you get a real prototype, real users and real feedback. If you wait until you have time, the need has gone and you've missed the chance to learn.

To a lot of people that sounds crazy. Why run a little experiment when all this work is piling up? It might be a gamble, but it can pay off. You don't need a perfect solution. It can be scrappy, and with AI it's easier than ever. I've spent a lot of time playing with tools and code, enough to understand what's possible, and it's a lot. So when I look at labour, I often ask how I can automate some or all of it.

That's how I approach a lot of problems. I try to understand the problem, find the friction and imagine it from several people's perspectives. Then I ask myself: is this painful to do, and why? What would make it awesome?

I see so much opportunity to improve how work gets done, and it doesn't always mean buying expensive tools. Sometimes it means turning a regular document into a shared template, instead of a master copy that sits on your manager's personal drive, shared with everyone, with strict instructions to make a copy.