App design that works in Arabic and English.

A component system, not a folder of screens. Each component carries what it does when the content is empty, when the text runs long, when the network fails and when the layout mirrors, written as values an engineer can type rather than measured off a picture. The screens come with it, and so do the listing assets at each device class and in each language. If what you are handed is a set of high-fidelity frames and nothing else, the specification work has not been done, it has been deferred to whoever builds it.

Component states, not a translation pass. Both platforms define the mechanism, so the design has to say which properties are directional, what mirrors and what deliberately does not, such as a phone number or a progress direction, and how each component behaves when a label runs half again as long. That is layout work, and it is separate from the typographic half, which is whether the typeface has a real Arabic companion and how the two are paired. That decision is made once, at identity, and this page inherits it.

They can, and we prove it before handover rather than promising it. An engineer builds one screen from the system alone while the engagement is still running, and every question they have to ask is a gap we close. Redrawing happens when a file shows what a screen looks like without saying what it is made of, so the developer rebuilds it in their own terms and the two drift from the first sprint. The test costs a day and it is the difference between a system and a mood board.

Substantially, and most design pages here are written as though it were not. Nobody has to be persuaded to install it and nobody leaves for a competitor, so the work is not delight, it is error recovery, one-handed use in a place with gloves or poor light, and how long a new starter takes before they stop asking. Success is measured in task time and support tickets rather than downloads. Where the tool is really a workflow problem rather than an interface one, automation is the shorter answer and we will say so.

By designing the two moments around it rather than the sign-in itself. The integration switches the person out to the national identity application and returns them, so the screens you own are the one that sets the expectation before they leave and the one that receives them when they come back, including when they come back having cancelled, or much later, or to an app the operating system has since closed. None of that is visible in a screen set, which is why it is a flow decision. The approval timeline sits outside design entirely, and the Dubai page sets out what it waits on.

You do, in your own workspace, from the first week rather than at the end. The component library, the source files and the font licences are held in your name, so a second designer or a new agency opens them without asking us for anything. Ask for that in writing before any design engagement, here or elsewhere: a library you can only reach through your supplier is the quiet version of a lock-in, and it usually surfaces at the least convenient moment.

Yes, and it is usually the right first step. We walk the first session on a handset, name where people stop screen by screen, and rank what is worth fixing against what it costs to fix. Often the answer is two flows rather than a redesign, and saying so is the better answer for you and the shorter one for us. Where the product is a web application rather than an app, the same work sits at web UI and UX design.