Design at Anested has one governing sentence: interfaces should explain themselves. Everything else we practice — every review comment, every rejected mockup — is a consequence of that single test. If a user needs a manual, the design is not finished.
First, hierarchy before decoration. A screen is a ranked argument about what matters most. If everything is bold, nothing is; if every element demands attention, the user receives none of it. We set type scales, spacing, and alignment before anyone is allowed to pick a color, because hierarchy communicates even when styling fails.
Second, states are the design. Empty states, loading states, error states, success states — that is where users actually live, especially in the first week of using a product. We design them first, not last. A beautiful happy path attached to a cryptic error state is not a beautiful product; it is a broken one with good marketing.
Third, motion must earn its render cost. We removed nearly all animation from our own corporate website, and what survived is hover feedback and nothing else. Every animation is a claim on the user's attention and the device's battery. Restraint, we have found, reads as confidence.
Fourth, accessibility is a floor, not a feature. Contrast ratios, focus states, keyboard navigation, and semantic markup are enforced in code review exactly like any other correctness issue. An interface that only works for some users is an interface with a bug.
And fifth: if we catch ourselves writing a tooltip to explain a control, we first try to redesign the control. Sometimes the tooltip wins — but it has to win the argument. The manual is the bug, and every principle above is just a different way of not writing it.
Need help with something we wrote about?
Our engineers answer support queries directly — hosting, databases, code, or anything from this post.