The thing that took me a while to internalize is that when you hit buy on a retail brokerage app, your order almost never touches the exchange you are picturing. There is no floor, no matching engine on the NYSE quietly pairing you with a seller. Your order gets sold, in bulk, to a firm you have probably never heard of, and that firm decides what price you get. This is not a scandal exactly, and it is not free money for you either. It is just the plumbing, and once you see how the plumbing works, the zero-commission headline stops being the number you care about.
Let me trace one order from the moment you tap the screen, because the path is more interesting than the marketing.
Where your order actually goes
Say you buy 100 shares of some liquid large-cap stock. Your broker does not run to the exchange. Instead it hands your order to a wholesaler, also called a market maker or an internalizer. A handful of firms dominate this business, and they pay your broker for the privilege of receiving that flow. That payment is the PFOF, payment for order flow. Fractions of a cent per share, sometimes a bit more, sometimes rebated in other forms, but it adds up to real money across millions of orders.
The wholesaler now has your order and a decision to make. It can fill you from its own inventory, it can match you against another retail order flowing in at the same moment, or it can route you out to an exchange. In practice it fills you itself most of the time, because retail flow is the flow it wants. Retail orders are what people in this business call uninformed, which sounds insulting but just means you are not a hedge fund with a reason to think the stock is about to move. You are noise, statistically speaking, and noise is safe to trade against.
Here is the part that confuses people. The wholesaler is required to fill you at or better than the National Best Bid and Offer, the NBBO, which is the tightest publicly quoted spread across exchanges at that instant. So the wholesaler often gives you a price slightly inside the spread. If the market is 50.00 bid, 50.02 offer, you might get filled at 50.009 instead of the full 50.02. That fraction of a cent per share is called price improvement, and brokers love to quote it because it is technically a benefit to you.
Why price improvement is real and also a little misleading
Price improvement is a genuine thing. On a typical retail order you probably do get filled a hair better than the quoted offer, and across a year that saves you something. The wholesaler can afford to hand you that fraction because it is capturing the rest of the spread and it is doing so with almost no risk, since your order is small and uninformed and it is netting thousands of orders like yours against each other.
Where it gets slippery is the benchmark. Price improvement is measured against the NBBO, but the NBBO is a quote for a standard round lot and it does not reflect the price you might have gotten if your order had gone somewhere it could actually interact with resting liquidity or midpoint dark pool matches. The wholesaler is grading its own homework against a benchmark that is easy to beat by a rounding error. So a broker can truthfully say it delivered price improvement on the large majority of orders while the size of that improvement is tiny, and the more meaningful question, whether you would have done better in a different execution model, is one the statistics quietly skip.
The other thing worth naming is that the spread you pay is not visible to you. There is no line item. Commission went to zero and the cost moved into the fill price, where you cannot see it and cannot easily compare it. That is the whole trick. The money did not disappear, it changed its clothes.
Why PFOF persists anyway
I used to assume this arrangement survived purely because it is profitable and the disclosure is weak. That is part of it. But the honest version is more mixed. Retail traders genuinely do get fast fills, usually at or slightly better than the visible quote, at zero commission, on stocks that are cheap to trade. For someone buying 20 shares of an index-heavy name and holding for years, the difference between this and a theoretically perfect execution is a couple of pennies they will never notice. The model is efficient for the boring, high-volume, low-stakes trading that most retail activity actually is.
It gets worse for you in the corners. Wider-spread names, options, anything less liquid, and larger sizes are where the invisible cost stops being a rounding error. Options PFOF in particular tends to be far richer than equities, which tells you something about where the spreads and the margins live. If your trading concentrates in those areas, the execution model you are on is doing more to your returns than you think, and the zero-commission label is telling you nothing about it.
How to actually read a broker's execution quality
The useful move is to stop judging by the commission headline and start reading the disclosures that describe execution. Regulators require brokers and market centers to publish order-routing and execution-quality reports, and while they are dry, they are where the real information lives. Here is roughly how I look at them.
- Find where the broker routes its orders. If nearly all non-directed flow goes to one or two wholesalers, ask what the broker is being paid for it. The routing report usually discloses payment received, often per hundred shares or per contract.
- Look at the execution-quality stats by order size and type, not just the blended average. The average is dominated by tiny liquid trades and hides how the broker does on the orders where execution actually matters.
- Compare effective spread to quoted spread. Effective spread is what you actually paid relative to the midpoint at the time of your order. A broker delivering good execution shows an effective spread meaningfully tighter than the quoted spread. That single comparison tells you more than any price-improvement percentage.
- Check price-improvement in dollars per share or per order, not just the percentage of orders that got any. Ninety-nine percent of orders improved by a hundredth of a cent is a worse deal than eighty percent improved by half a cent.
- If you trade options or less liquid names, weight all of the above heavily, because that is where the spread you cannot see is widest.
None of this requires you to abandon a commission-free broker. Plenty of them route reasonably and the cost to a long-term, large-cap buyer is negligible. The point is to know what you are actually paying and where, so the decision is yours instead of the wholesaler's.
A practical rule I have landed on. If your trading is small, liquid, and infrequent, the execution model barely matters and you should optimize for whatever else you value, interface, features, tax tools. The moment your size grows, or you move into options or thinner names, or you start trading often enough that fractions of a cent compound, pull the execution-quality report and read the effective-spread numbers before you read anything the marketing put on the front page. The commission being zero was never the interesting number. What the fill actually cost you always was.