Product specification has a point of diminishing returns.
Like most things in life, a nice chart makes it easier to understand.
Time flows from left to right. The green line is how complete a specification is and the red line is how hard it is to make the green line move.
Important: This chart assumes that product management has provided details of the problem space and requirements. How much of the specification they provide is variable and won’t be discussed in this post. Also the way PMs and Designers split up the work will also be omitted from this post.
Phase 1. Creative Burst
This is the beginning. Effort is low to yield alot of specification details. This is where basic information architecture and the structure of the system is designed. You can get about 30% of all the details in this short burst of time. Engineers should be aware of these designs on a loose basis. About 20 minutes a week is plenty of time to understand where its going.
The difference between a great product and a terrible one is often created in this phase.
Phase 2. Specification
This is where effort really starts. There are so many details on a single screen. Look at any UI of a product. Count the number of independent things on the screen. It’s not uncommon for dozens or even hundreds of variables/affordances to be present on any given screen. On a single blank screen of PowerPoint, I counted well over 150 individual things that need specification. Each of these needs programming and detail. How does it work? What does it do? Where does it come from? Can it be customized? What does it look like? The “job” of product design is to detail out as much of these questions as possible.
During this phase, we hit a point of diminishing returns. In other words, the effort to detail the design is higher than the value of that detail. Most organizations keep pushing the designers for more and more detail, despite the ROI being terrible. Enlightened teams switch to a different mode of design at this point.
Heresy: Designers should sit with engineers and pair on the work.
I know this is controversial, but the best designs I have ever done were accomplished sitting side-by-side with an engineer. Of course there are moments when you say, “Let me mock that up”, but often it’s just simply tweaking the UI with the engineer at the desk.
The devil is in the details and this is (in my experience) the best way to achieve excellence.
Phase 3. Development
The whole point here is that you shouldn’t wait for 100% specification to begin development. In fact, 100% usually is a pipe dream, an impossible end. This is because there is always room for improvement and the time it takes to get that improvement isn’t worth it. The ROI is terrible. It’s much better to get the thing to market and learn from the audience.
If you don’t have designers sitting with engineers, you will lose the benefit of design thinking at this stage. Get the best of both worlds. Don’t have designers throw something over the wall and then resign themselves to whatever outcome arrives.
There are many other nuances and facets to this work. There is no way I am able to be comprehensive in a single post. However, I hope the chart comes in handy and maybe helps someone along the way. This post was just a phase 1 exercise.