Most idle games can be played on instinct: buy the most expensive thing you can afford, repeat. That mostly works here too, but it will cost you hours, because Idle Broker has one mechanic that punishes instinct — automation. This guide explains what is actually happening under the numbers.
This is the rule the whole game is built on. In a typical clicker, owning a generator adds a number to your income and that number ticks in continuously. Here, a generator sits idle until a trade is started. The trade then runs for a fixed duration and pays a single lump sum when it settles.
Two consequences follow, and both catch people out. First, an un-started generator earns exactly nothing, no matter how many you own. Second, the Return/s number in the header is an average — your balance will jump in steps, not climb smoothly. If you sit and watch it, it will look broken. It is not.
The TRADE button has no trade of its own. One press starts a trade on every asset you own that isn't automated yet, and the badge on it counts how many that will be. Mashing it does nothing — an asset already mid-trade ignores you — which is deliberate: this is not a game about wearing out a mouse button. What it does reward is being there: trades you start by hand carry the Midas Touch bonus, and automated ones do not.
One consequence worth stating plainly, because it defines the opening minute: you cannot earn a cent without owning something. A run starts you with just enough seed capital for one Intern, and hiring them is the first thing you do.
Automation costs 15× a generator's base price — adjusted, like every price in the city, by the market phase you are in — and is bought once, per generator type. From then on that generator restarts its own trades forever, with no input from you.
The part that matters more than the convenience: only automated generators earn while you are away. Offline earnings are calculated from your automated income alone, capped at eight hours. If you close the tab with a hundred un-automated Interns, you come back to nothing.
So the common new-player mistake is buying a wide, shallow spread — a few of everything, nothing automated — and wondering why the game feels dead after a break. The stronger line is almost always: automate the Intern first, even though 225 coins feels enormous at that stage, then automate each new tier as soon as it is affordable rather than buying more units of it.
Rule of thumb
When you can afford either another unit of a generator or automation for it, buy the automation. One automated unit out-earns several manual ones over any real span of time, because you are not there to start the manual ones.
Advertisement
Every generator has a base trade duration. Owning more units of it shortens that duration at four thresholds — a bigger desk is a tighter desk:
| Units owned | Trade speed |
|---|---|
| 1 – 9 | Base duration |
| 10+ | 7% faster |
| 25+ | 15% faster |
| 50+ | 28% faster |
| 100+ | 40% faster |
These thresholds are worth aiming at deliberately. Sitting at 23 units of something is leaving a speed-up on the table for the price of two more; the jump from 9 to 10 is the cheapest one in the game.
Base trade durations and prices for every generator. Longer trades pay proportionally more — a Crypto Exchange trade takes three times as long as an Intern's, and pays far more than three times as much.
| Generator | Cost | Automation | Trade |
|---|---|---|---|
| Intern | 🪙 15 | 🪙 225 | 5s |
| Coffee Cart | 🪙 150 | 🪙 2.25K | 6s |
| Stock Portfolio | 🪙 1.1K | 🪙 16.5K | 8s |
| Investment Fund | 🪙 12K | 🪙 180K | 10s |
| Hedge Fund | 🪙 130K | 🪙 1.95M | 12s |
| Crypto Exchange | 🪙 1.4M | 🪙 21M | 15s |
Each additional unit of a generator costs 15% more than the last, so prices climb steeply within a tier. That is usually the signal to move up a tier rather than keep buying the one you are on.
Advertisement
Prestiging retires your firm: coins and generators go, permanent upgrades stay. The instinct is to delay it as long as possible. That instinct is wrong.
The honest signal is time-to-double. Early in a run your balance doubles in minutes. Late in a run it takes hours. The moment you notice you are waiting rather than deciding — when whole tiers are out of reach and nothing you can buy meaningfully changes the rate — you have hit the flat part of the curve, and everything past that point is worth less than a reset would be.
A good first prestige typically feels early. Your second run will pass your first run's peak in a fraction of the time, and the run after that faster still. The reset is not a wall to be postponed; it is the actual progression mechanic.
Prices in a city are not constant. They move through three phases as your star total closes on the cost of the next city, and the Investments panel names the one you are in:
| Phase | Prices | What to do |
|---|---|---|
| Opening bell | −25% | Build hard. This is the cheapest the city ever gets. |
| Tight market | +40% | Automate what you have and leave. Eight hours offline beats grinding through it. |
| Closing bell | −15% | Push for the stars that buy the next city. |
The squeeze is deliberately a wall. Because it cuts your income it also cuts the stars a run produces, so the middle of a region takes more prestiges than either end — the pacing is real, not cosmetic. It is also the game telling you, in the only language an idle game has, that this is a good moment to close the tab.
Prices are fixed for the whole of a run. The phase is read when you prestige or move city, never live, so claiming a quest that pays stars can never re-price the shop out from under you mid-run.
Gems are the premium currency. You earn 2 from every achievement — thirty of them, so roughly 60 over a full career — and you can buy them in packs. They are tied to your account, which is why the shop asks you to sign in.
They spend on two different things, and the distinction is what makes a free trickle go further than it looks:
If you are spending free gems, buy buffs. A surge is gone in half an hour; Passive Engine is still paying you three cities later. Star Collector is the strongest of the five for anyone planning a long game, because it multiplies every future star payout rather than a single run’s income.
Two things worth knowing before you spend. A time warp pays nothing for un-automated generators, because they earn nothing while unattended — the game refuses the purchase outright if you have automated nothing, but it will happily sell you a warp that pays less than it could if you buy one before automating. And the city skips on the World Map (40 gems for London, rising to 700 for the Moon) buy the move without spending stars — the stars you already hold stay put and keep counting toward the city after.
Automation costs 15× base price and is the only income that runs while you are away. A payback table for every tier at 1, 10 and 25 units, the rule of the twentieth unit, and the two situations where buying it wastes your money.
Stars come from a square root, which makes long runs worse value than they look. What each star costs at your stage, the plateau test for timing a reset, and why moving region is the one decision you cannot get wrong.
Replacing the coins-per-second engine with timed trades kept the average income identical and changed everything else — including one ownership speed-up that turned out to be a 1.67× income multiplier by accident.
A tester reached the halfway point of a weeks-long ladder in half a day. The cause was one constant that never scaled, and the fix was a single exponent — plus why the balance simulation behind it is only trustworthy about shape, not schedule.