Wi-Fi & eSIM app

Problem

People open a connectivity app when the connection is the problem. At launch there is no session and nothing to fetch, so whatever is on screen is all they get.

Hypothesis

If one check runs before anything is drawn, the app can show a state of its own instead of a half-loaded one.

Solution

One check at launch, two exits. Connected: normal sign-in and routing. Offline: a dedicated screen with guidance and Try again, and no cached account or plan pretending to be current.

Problem

One screen had to serve three people: someone who has bought nothing, someone with Wi-Fi, and someone with Wi-Fi and an eSIM. A catalogue sells customers what they already pay for, and a dashboard hands a newcomer an empty room.

Hypothesis

If Home is composed from what the account holds, one page can be a shop for some and a control panel for others.

Solution

Home is built from state. No account: two doors, plans or sign in. One service: its card with speed, devices and next payment, and the eSIM below as the one thing left to add. Both: nothing to buy, because nobody can hold two of either. Buying the eSIM is activating it, since there is only one universal plan.

Problem

The web portal had seven items in its nav bar and plans were asking for an eighth, although plans are a journey, not a place.

Hypothesis

If the shell keeps three tabs, the navigation matches how the product works.

Solution

Seven pages became three tabs: Home, Support, Account. There is no Plans tab, because Home already shows what this account can still buy.

Problem

Partner venues and universal plans sat in one list, and nobody could tell what was actually theirs.

Hypothesis

If the app carries only universal plans, the catalogue stops confusing people.

Solution

Two universal catalogues, base Wi-Fi and mobile eSIM. Venue registration stays in the captive portal on purpose.

Problem

Buying on the web took seven steps, with the plan chosen inside the checkout, so comparing and paying blurred into one long form. Every step was a place to give up.

Hypothesis

If browsing leaves the stepper and only what must be sequential stays, buying reads like buying.

Solution

Three steps: plan, account, review and pay. Comparison moved into the app shell, because it is shopping, not checkout, and with the integrations in place a purchase takes seconds instead of minutes.

Problem

A plan was something to read: tier, price, renewal. Pausing, upgrading or adding a laptop happened elsewhere, and since nobody buys a second copy of a service, the account is the only place those changes can live.

Hypothesis

If each service screen carries its state and its actions, the account stops being a receipt and becomes where the product is run.

Solution

Account opens on the services and where each stands. Each service gets its own screen: Active, or Paused with the date it resumes, and under it devices, upgrade, cancel. Devices have their own screen with the network name, password and a QR code that connects a phone without typing.

Problem

People usually need help at the moment the connection is failing, often before they have an account. Support that lives inside an account arrives too late.

Hypothesis

If help is its own tab and also a link on the screens before sign-in, nobody has to be signed in to reach a person.

Solution

Support is one of the three tabs and a Need help link on every screen before an account exists. It hands over a phone number and an email instead of a form, so it still works when the connection is the question.