“80% of failures can be avoided in the design phase”

This statistic, and various versions of it, is peppered all over design literature. In an earlier blog post, I pulled at the loose threads of this statistic and in doing so can hopefully discourage a few designers from using it uncritically in their arguments.

But if it is true (though unvalidated), what is it actually saying?

It’s saying that design is too important to be left to designers. We know much of that 80% isn’t under the control of the designers on a project and that instead will be the result of demands from product or marketing, of technical constraints, of security or of legal requirements.

It’s okay to say this. Product design should be done by multidisciplinary teams, not by lone designers.

Stating that 80% of failures can be avoided in the design phase is also implicitly suggesting that 80% of failures could be predicted (and so designed out) in the design phase.

But we don’t start our projects with the benefit of hindsight. Design requirements are often too uncertain upfront, some even unknowable, for design and redesign to be bounded to a design phase.

Though the traditional linear design-then-build model might still be right for the construction of nuclear infrastructure and major civil engineering projects, it’s not generally true in the world of product design any more and not true at all in the world of digital product design. Agile, iterative design methods mean multidisciplinary teams get to design out problems as they arise, wherever they are in the product development lifecycle.

Even with a multidisciplinary team working to an iterative design process, we’ll find problems that could have been (and were) predicted, failures that could have been avoided.

That’s okay too.

As much as we try to bake in good practice from the start , and “shift left” in testing, deployment and design so we can see and address problems much earlier in the process, we can’t avoid taking on design- and technical debt altogether.

Yes, the earlier you deal with your debt, the less it will cost you in the long run.

But ask any product owner and they’ll tell you, there’s always more to do than can be done and an opportunity cost to everything in the team’s backlog. So you have to prioritise what the team works on next — based on what matters most to your customers and your business.