The last ten percent

At my first design job, we used to talk about the last 10%. This was the time that we’d set aside at the end of a project, to sit down and review the website together.

We’d compare the (almost) finished website against the mock-ups, the navigation diagrams and the content plan. We’d stop making the site and try to experience the site. We solved content issues, white space issues, typography issues, interaction issues. We were designing in the browser, although we didn’t call it that.

We were designing in the browser, although we didn’t call it that.

Sometimes we’d even be shifting pixels about. Yeah I know pixel perfection is an unrealistic goal, but if every element is 5 pixels out, then the overall effect can be a design that feels disjointed and unbalanced.

It would always take longer than 10% of the project budget to get this done. But the effect was usually huge. It turned a site from an average one into a great one.

Part of the downfall of that business was that later on, we tried to cut out this process. We tried to compartmentalise our individual skills. We tried to force a waterfall approach and to become more streamlined. We stopped allowing for the final 10%. This meant that we shipped websites that weren’t finished, even though we had ticked every box in “our process”.

For a long time I held on the the crazy idea that better planning, better skills and more experience would lead to projects that didn’t need this review process.

Whatever you call it — the last 10%, testing, proofing, quality assurance — this is when the real work starts. It’s much more agile and much more realistic. It’s how really good products and services are designed.

Now, the last 10% usually takes up about 90% of the budget.