Finder replacements for a Mac, and when to switch

Looking for a Finder replacement on a Mac usually starts with a specific moment of friction, not with a considered evaluation. A copy between two folders took four steps. A server connection dropped again. A path had to be typed into a terminal that was already open in the wrong directory. The search that follows tends to produce ranked lists of apps, which is the wrong shape of answer, because these three complaints have three different solutions and only one of them is a general file manager.

What follows sorts the options by the problem they solve, lists prices as the vendors state them, and marks the parts of macOS that stay with Finder no matter what gets installed.

Name the complaint before opening a download page

The useful first step costs nothing. Write one sentence describing the last time the current setup was annoying, using the words that were true at the time.

If the sentence is about moving files between two places, the answer is a dual pane manager, and almost any of them will feel like an improvement.

If the sentence is about a server, an SFTP connection or cloud storage, the answer is probably a transfer client rather than a file manager. Transfer clients handle protocol edge cases, resume behaviour and server quirks better, because that is all they do.

If the sentence is about typing paths, the answer is proximity between the file list and a shell, which only some managers offer.

If the sentence is vague, and it often is, the honest version is usually some form of "too much switching between windows". That is a layout problem, and replacing one two pane app with another two pane app does not touch it.

Getting this right saves a week. Each of the categories below is genuinely good at one of these jobs and mediocre at the others, so a mismatch reads as "the app is fine but nothing improved".

What stays with Finder regardless

No third party manager becomes the system file browser on macOS. This is a platform constraint, not a limitation of any particular app, and it sets a ceiling on how complete a switch can be.

The desktop is drawn by Finder. Open and Save dialogs in every application are Finder based, so the sidebar, tags and recents that appear there come from Finder's configuration. When another app offers "Show in Finder" or "Reveal in Finder", it opens Finder. AirDrop, Quick Look defaults and the standard Get Info window belong to the same system.

Some managers work around parts of this, and a few can hide the desktop icons or register themselves for certain actions, but the practical result is the same for everyone: a replacement is a second, better window for file work, not the removal of the first one. Anyone expecting Finder to disappear from the workflow entirely will be disappointed regardless of which app gets installed.

This matters for the decision because it changes what a replacement has to be worth. The app is not saving the cost of Finder. It is earning its place next to Finder.

The direct replacements, with current prices

These occupy the same slot: a two pane window, keyboard driven navigation, remote volumes, and more file operations than the standard window exposes.

App	Price as listed	Built in terminal	Notes
ForkLift 4	$19.95 single user with one year of updates, $34.95 with two years	No	Strong on remote volumes and sync
Path Finder	$32.95, 1 Year License	Yes	Deep local browsing and organising, 30 day trial
Commander One	$29.99 PRO Pack, one time, or $99.99 for five Macs	Yes, terminal emulator	Free version on the Mac App Store
Marta	Free, with a Patreon	Yes	Native Swift, Lua plugin API, task queue
Double Commander	Free, open source	No	Cross platform, familiar to Total Commander users
muCommander	Free, open source	No	Java based, broad protocol support

Two observations are worth pulling out of that table.

The paid options are inexpensive relative to the time they claim to save, and the gap between them is small enough that price should not drive the choice. A perpetual licence with a limited update window, which is how several of these are sold, means the app keeps working after the window closes and stops receiving new versions.

The free options are not stripped down demonstrations. Marta in particular is a complete native manager with an archive handler, a task queue and a plugin API, which makes it the most direct answer for anyone whose complaint was purely about paying.

How to read the licence terms in that table

Two pricing models appear above and they behave differently over time.

A perpetual licence with a limited update window, which is how ForkLift and Path Finder are sold, means the version installed keeps running indefinitely and stops receiving new releases when the window ends. Renewing buys further updates, not continued use. The practical risk is a future macOS release changing something the last supported version cannot handle, which is a real but occasional event.

A one time payment, as with the Commander One PRO Pack, sits in the same family with a different boundary between paid and free upgrades. Free and open source options avoid the question entirely and replace it with a different one, which is whether the project is still actively maintained.

None of these differences justify picking on price alone, since the spread across the whole table is smaller than an hour of most people's time. The reason to read the terms is to know what happens in year three, not to save ten dollars in year one.

The transfer clients, which are a different category

Confusing these with file managers wastes evaluation time. A transfer client connects to servers and cloud storage and moves files between them. It does not replace the local browsing window.

Transmit 5 from Panic lists at $45 and connects to a long list of protocols and cloud providers. Cyberduck is free and open source, with a donation model and a paid version on the Mac App Store, and covers a similar protocol range.

If the original complaint was a dropped SFTP session or a sync that behaved unexpectedly, one of these will fix it and a general file manager probably will not. It is also perfectly normal to run both: a manager for local work and a client for anything that leaves the machine.

Terminal file managers

For readers whose complaint was about typing paths, there is a category that lives entirely inside the shell. Yazi, ranger, nnn, lf and Midnight Commander are all free and open source, installable through Homebrew, and they navigate with the keyboard alone.

The advantage is exact: the file browser and the command line are the same window by definition, so there is never a path to copy between them. The cost is equally exact. Previewing images and PDF files depends on terminal support and configuration, drag and drop to other applications does not exist, and the setup is a config file rather than a preferences window.

These tools suit people who already live in a terminal. For anyone who works visually and occasionally runs commands, the direction of travel is the reverse: a graphical manager that has a shell inside it.

What macOS already does, before anything gets installed

A share of the reasons people look for a replacement are covered by features that are already present and rarely enabled.

The path bar and the status bar are both off by default, and both are one menu item away in the View menu. Turning them on removes a large part of the disorientation that pushes people toward other apps, because the full path and the item count are visible at all times.

Column view navigates a deep hierarchy with the arrow keys alone and shows several levels at once, which is closer to a two pane manager than most people expect. Tabs work in the standard window, and two tabs dragged into a split arrangement covers a fair number of copy tasks.

Bulk renaming is built in. Selecting several files and choosing Rename Items provides find and replace, adding text, and sequential numbering by name, date or counter, which is the main reason dedicated renaming apps get bought.

Smart Folders turn any search into a permanent view in the sidebar without moving a file. Tags add a second axis alongside the folder tree, and a file can carry several at once. Quick Look previews almost anything with the space bar, and holding the option key while dragging copies instead of moving.

None of this makes a replacement pointless. It does mean the honest comparison is against a configured standard window rather than an unconfigured one, and a few of these settings will resolve the complaint entirely for some readers. Spending ten minutes here before spending money is a reasonable trade.

When the real problem is the number of windows

The complaints that no app in the table above resolves sound like this. Files are listed in one window. Commands run in a second. An assistant that renames, summarises or sorts things runs in a third. Every task crosses at least two of them, and each crossing costs a few seconds plus the attention needed to re-find the place.

Counting the crossings for a single ordinary task is a fast diagnostic. Sorting a folder of screenshots into dated subfolders typically means selecting in the file window, switching to the terminal, confirming the path, running the command, switching back to verify. Five context changes for one small job, repeated many times a week.

A manager with an embedded terminal removes one of those. Removing the third window as well means the assistant reads the same directory that is on screen, which is what a file manager built around folders, a terminal and an assistant in one window is for. How that arrangement differs from the two pane managers above is set out in the comparison, and the pricing page covers what it costs.

Testing a candidate in two weeks

Evaluations in this category go wrong in a predictable way. The app gets installed, driven around a folder of sample files for an afternoon, declared impressive, and then quietly abandoned because the old habit never changed.

A test that produces a real answer has three properties.

It uses real work. Feature checklists confirm that something is possible. They say nothing about whether the tool gets reached for under pressure, which is the actual question.

It runs for about two weeks. The first few days measure novelty and the friction of unfamiliar keybindings, both of which fade. What remains after two weeks is the answer.

It changes one thing. Trialling three managers at once produces no usable comparison, because none of them accumulates enough hours to form a habit.

The trials support this. Path Finder offers 30 days, ForkLift and Commander One provide trials as well, and the free options can run in parallel at no cost. Setting a calendar reminder for day fourteen with one question written on it, whether the original annoying sentence still happens, turns an open ended evaluation into a decision with a date.

If both candidates feel fine and nothing measurably improved, the complaint was almost certainly not about features. That result is worth having, because it redirects the next hour of searching away from another two pane manager.

What to change first

Start with the one sentence describing the last annoyance, then pick the category it points at rather than the app with the best review score. Try one option for two weeks with real work, not with a folder of test files. If the sentence was about window switching rather than features, look at Atriens before buying another two pane manager.