Wavebox: the setup order that holds up

Installing Wavebox takes about five minutes, and the setup wizard walks you through it without asking anything difficult. That is precisely the problem. The wizard asks you to name three spaces at a point where you have no basis for naming them, and the answer you type becomes the structure everything else hangs from. A week later the arrangement does not match how the work actually splits, and rebuilding it means signing into every account a second time. The install is trivial. The order of decisions is not.

Four decisions that belong before the installer

Settle these four before double clicking anything. All four are technically reversible, and all four are expensive to reverse.

The login address you create the account with. The number of spaces and the axis you cut them on. Whether anything deserves a separate profile. What you are bringing across from the browser you use now.

Writing those down first converts the setup wizard from a series of open questions into data entry. Skipping it means accepting the defaults, and the defaults in the wizard are examples rather than recommendations.

The login address is the decision you cannot cheaply undo

Account creation offers three routes: sign in with a Google or Microsoft address, create an account with any address and confirm it by code, or enter the address a team invitation was sent to. The third route skips the rest of the wizard entirely and opens a profile preloaded with whatever template the administrator configured.

For the first two, the consequence to understand is that the address you use becomes the primary sign-in for the default space. Apps added to that space stay authenticated against it without further work. Start with an address you rarely use, or a throwaway created to test the thing, and the account you use most is permanently a second class citizen inside the setup.

Changing the default space later is possible. It is done from the spaces list in settings, through the three dot menu next to a space name. What the documentation notes alongside that instruction is the part that matters: after reassigning the default, you log back into the apps in the newly assigned space. One wrong address at the first screen produces a re-authentication queue as long as your app list.

The same address governs a second machine. The knowledge base is direct about it:

When installing on a second computer, you must login using the same email address as your first computer in order to pick up your subscription and any cloud/sync profiles or team templates. Source: kb.wavebox.io

If a laptop and a desktop are both in the picture, pick the address that will still be correct on both of them in a year.

The install itself, in order

Once the four decisions are made, the sequence on macOS is short.

Open the downloaded disk image in Finder and drag the icon into the Applications folder. Create the account by one of the three routes above, checking the junk folder if a confirmation code does not arrive. Import bookmarks and passwords from Chrome or Edge, which is optional, and which the documentation states happens without passwords leaving the machine. Create spaces, up to three in this screen. Assign an email address to each space. Drag the apps you use into each space. Add an ad blocking extension and choose the search engine for the address bar.

Two of those steps get skipped and cause trouble later. Assigning an address per space is the first: that address becomes the default identity for sign-ins inside the space, so setting it during the wizard removes a decision from every future login. The extension step is the second, because the free tier allows exactly one extension. If the plan involves both an ad blocker and a password manager, that plan requires the paid tier, and it is better to know that on day one than on day eight.

Space or profile: a test that takes ten seconds

The documentation frames this as a room full of filing cabinets. The machine is the room. A profile is a cabinet. A space is a drawer inside a cabinet.

A space separates login state and nothing else. Because the drawers share a cabinet, search runs across all of them at once, unread badges from every space roll up into one view, and an app from one space can sit beside an app from another in split screen. A client inbox and your own inbox, signed in as different people, side by side.

A profile separates everything: settings, extensions, passwords, the entire space structure. Switching profiles is switching worlds, and nothing follows you across.

Situation	Use
Accounts must not mix, but the day should stay unified	Space
Company finance, or a client needing a sealed environment	Profile
Another person uses the same Mac	Profile, theirs
One large project whose unread count drowns everything else	Profile

The last row is the one people miss. Unified badges and unified search are the point of spaces, and for a big enough project that unification becomes noise rather than signal. Putting that project in its own profile gives it a clean, self contained view with nothing else leaking in.

Cut groups by task, not by application

With spaces settled, groups come next. A group is a rounded square in the sidebar, carrying its own colour, icon and unread badge, and holding a strip of apps and tabs. A group containing a single app can hide that strip entirely.

The organising principle is the one used for folders on a phone home screen: group by what you are doing, not by what kind of software it is. Published examples run along two axes. One is by task, such as travel, finance or streaming. The other is by whose accounts they are, such as client one, client two, side project, school. Pick one axis and hold it. Mixing both is what produces a sidebar nobody can read.

Each group has a space attached to it, which means apps and tabs opened while working inside that group land in that space automatically. That is a convenience rather than a constraint, and a group can hold apps drawn from more than one space when the work genuinely requires it.

The sidebar offers two layouts: one showing only the groups belonging to the current space, and one showing every group across every space. Past roughly five spaces, the second layout stops being readable, so plan on switching to the first as the structure grows.

How many spaces, and how to count them

The second decision goes wrong in a predictable way: people count services and end up with far more containers than the work requires.

The number that matters is not how many accounts you have. It is the largest number of accounts you hold on any single service. Three Gmail addresses and two Slack workspaces need three spaces, not five. The three mail accounts set the floor because they cannot share a cookie jar, and the two Slack workspaces then distribute into two of those three. Counting service by service and adding the totals produces a structure with containers nobody uses.

This matters because every additional space is another full round of signing in. Six spaces means six sets of authentication, including whatever two factor steps each service imposes. Spreading accounts that never collide across separate containers adds all of that work and changes nothing about the separation you actually get.

A faster way to arrive at the right number is to count invitations. Accounts you were invited into, by clients or by other organisations, are the ones that genuinely belong to someone else and have to stay apart. Accounts you created yourself usually collapse into one container without any conflict at all.

Frequency is the second filter. Something opened many times a day earns permanent space in the sidebar. Something opened once a month occupies that space permanently while returning almost nothing, and is better created at the moment it is needed. Starting deliberately small and adding containers as collisions actually appear produces a structure that matches the work, rather than one that matches an idea of the work formed during the first hour.

The seven days are a test plan, not a demo

The trial runs for seven days with full access to the paid features and no card required. Treating it as a period to be impressed by is a waste of it. Four specific things should be verified while the paid features are still on, because each one determines whether the arrangement survives contact with real work.

Whether unread counts are correct for the services you actually use. Per-service handling differs, and a service that counts wrong turns consolidated badges into something you learn to ignore.

Whether the link engine routes links the way your day requires. Paste a link from a chat and confirm it opens as the right identity, not merely in the right application.

Whether every extension you depend on installs and behaves. Dropping to the free tier leaves one, so the count matters more than the compatibility.

Whether a second machine picks up the configuration, which is where the address decision from earlier gets its exam.

If seven days does not produce answers, the documented route is to reply to the trial ending email and ask for an extension. Deciding on incomplete evidence because a timer expired is the worse option. What the paid tier actually unlocks, and the count where the free tier stops, are laid out on Pricing, and the practical questions that come up during a trial are collected in the FAQ.

Rebuilding costs scale with how far out you go

The reason setup order matters is that undo cost is layered.

Moving an app between groups takes a drag. Recutting groups is a few minutes of rearranging. Adding a space, or reassigning which space is the default, means signing into every app in that space again. Splitting something into its own profile means reinstalling extensions and re-entering passwords on top of that.

So decide outward to inward: profiles, then spaces, then groups, then where individual apps live. Done in that order, every correction stays contained in a layer you can fix quickly. Done in the opposite order, starting by arranging apps because that part is visible and satisfying, each outer change collapses everything inside it.

What to change first

Write down the four decisions before opening the installer: the address, the number of spaces, whether anything needs its own profile, and what comes across from the current browser. That single page is what separates a setup that survives a month from one rebuilt twice. Deciding the surfaces before filling them is the same principle behind Workspaces, and SpaceDeck documents which services can sit on those surfaces.