Electronic exchanges and online stockbroking changed different parts of the same transaction. Exchanges automated the market where orders meet. Online brokers gave investors a way to submit those orders without speaking to a representative. The two developments eventually connected, but they did not arrive together.
That distinction matters. A price on a computer screen does not necessarily mean a computer executes the trade. Equally, clicking “Buy” in a brokerage account does not usually send an order straight to the exchange where the company is listed.
This chapter in the history of trading technology follows the shift from electronic quotations and private dealing networks to automated order books and online investor accounts. The central change was not simply faster communication. Software began performing tasks previously handled through conversations, handwritten instructions and repeated data entry.
Electronic Trading Began Before Online Stockbroking
Nasdaq put quotations on screens
Nasdaq launched on February 8, 1971, as the National Association of Securities Dealers Automated Quotations system. It allowed market makers in over-the-counter stocks to update their buying and selling prices electronically. Brokers could compare those quotations without collecting every price through separate telephone calls.
However, the original system automated quotations, not the entire transaction. Dealers still negotiated trades by telephone. Calling Nasdaq an early electronic stock market is reasonable; treating its 1971 launch as the arrival of a modern automated order book is not. Nasdaq’s history of market automation makes that distinction clear.
During the 1980s, systems including the Small Order Execution System, or SOES, extended automation into execution. The October 1987 crash exposed weaknesses in telephone dealing when market makers became difficult to reach. Automatic execution addressed a different problem from electronic price display: a visible quotation was of little use if nobody answered the phone.
Private networks and home computers followed separate paths
Instinet began operating a computer network for professional investors to arrange block trades in 1969. These private connections show why electronic trading should not be confused with internet trading. Institutions could communicate trading interest through computers long before household internet access became common.
Retail access developed separately. During the 1980s, some brokers offered proprietary software and direct dial-up connections for customers using personal computers. By the mid-1990s, brokerage websites were combining account access, research, quotations and order entry. The SEC’s 1997 report on technology in securities markets documents both developments and distinguishes customer order-routing services from systems that actually matched trades.
The difference was architectural, not cosmetic. One system connected professional buyers and sellers; another connected a customer to a broker. Both used computers, but they occupied different positions in the trading process.
From Electronic Prices to Electronic Markets
The next step was to turn trading instructions into executable orders inside a shared system. Instead of displaying a dealer’s price and leaving someone to negotiate, an electronic order book could accept instructions, rank them and execute compatible orders.
Globex extended automation into futures
CME launched Globex on June 25, 1992. Although it traded derivatives rather than ordinary company shares, its arrival belongs in the same history: exchange trading was becoming a networked service rather than an activity tied entirely to a physical floor. The launch date is recorded in CME’s product and platform anniversary record.
Electronic derivatives trading and online stockbroking should still be kept separate. A futures trading platform serves a different product and market structure from a retail share account. Their shared feature was the replacement of parts of a human communication chain with electronic instructions.
London introduced the SETS order book
The London Stock Exchange launched SETS in October 1997. It provided an electronic order book for FTSE 100 shares and other liquid securities, supporting both price formation and trade execution. Market participants could submit orders or execute against orders already displayed.
London did not move every security into one identical trading model. Quotation services and hybrid arrangements continued alongside the order book. The London Stock Exchange’s 2001 listing particulars describe SETS alongside those other services.
This illustrates a recurring feature of market automation: installing computers did not settle the question of how trading should work. An exchange still had to decide which securities suited an order book and where dealer quotations remained useful.
Xetra supported trading without a local floor presence
Deutsche Börse launched Xetra on November 28, 1997. Its electronic order book supported growth in trading and participation without requiring the old physical presence on the Frankfurt exchange floor. Transparency was part of the design, not simply an incidental benefit of replacing paper.
Xetra also introduced volatility interruptions at launch. Those mechanisms slowed trading during sharp price movements, giving participants time to reassess their orders. Deutsche Börse’s account of Xetra’s development shows that automation involved controls as well as speed.
Together, these developments demonstrate why there is no single date when trading “went electronic.” Quotation display, order submission, matching and risk controls moved into software at different times.
What an Electronic Order Book Changed
An electronic order book records buying and selling instructions. A matching engine applies the venue’s rules to decide which compatible orders trade. That sounds straightforward, but the ordering rules determine who receives an execution when several participants want the same price.
Price and arrival time are common priorities, with further distinctions for displayed and hidden interest. Under Nasdaq’s published order-priority model, displayed limit orders at the same price execute in the order received, ahead of non-displayed interest at that price.
Consider a simplified order book with two offers:
| Shares offered | Price per share |
|---|---|
| 100 | $25.00 |
| 200 | $25.02 |
Assume an incoming order buys 150 shares, permits both prices and encounters no other trading interest. It takes 100 shares at $25.00 and another 50 at $25.02. The total is $3,751 before fees, rather than the $3,750 implied by buying everything at the first displayed price.
Now change that instruction to a buy limit order capped at $25.00. Only 100 shares can execute against the offers shown. Whether the remainder stays available or is canceled depends on the order’s instructions.
The practical lesson is that automation does not manufacture shares at the price a buyer wants. It processes the available instructions. Faster matching and adequate liquidity are different things.
This also explains why execution quality cannot be judged by the speed of a confirmation alone. In the example, a fast fill at two prices and a partial fill at one price represent different outcomes, even if both appear almost immediately.
Online Stockbroking Brought Order Entry to the Investor
The retail change was less visible inside the exchange but more visible at home. Investors could enter a security, quantity and order type themselves rather than relay those details in conversation. The trade ticket became a screen.
That convenience also transferred responsibility for entering instructions correctly. A customer who meant to buy 10 shares but entered 100 could not expect software to recognize the intention behind the number. A clean interface was not a substitute for checking the order.
Early online brokerage also exposed the gap between submitting an instruction and receiving an execution. The SEC staff’s January 25, 2001 examination report on online brokers documented complaints about account access, delayed processing and order errors. It also highlighted confusion over pending orders and cancellation requests.
Those were operational problems, not just inconvenient loading screens. If an investor could not tell whether an order had executed, entering it again risked creating a duplicate. Requesting cancellation did not mean the marketplace had confirmed cancellation.
Consider an investor who submits a purchase and then loses the connection. There are at least two possible situations: the broker never received the instruction, or the trade proceeded while the customer lost visibility. The same blank screen can therefore conceal very different positions.
The appropriate design question is not simply “How quickly does the page load?” It is “Can the customer establish what happened to the order?” Clear status messages, confirmations and a usable support route belong to the trading service, not its decoration.
The Broker Did Not Disappear Behind the Buy Button
Online stockbroking can look like direct access because the customer controls the screen. In an ordinary retail account, however, the broker still receives and reviews the order before arranging execution.
The broker may execute it itself or send it to an exchange, an alternative trading system or a wholesale broker-dealer. Listing a company on one exchange does not mean every transaction in its shares occurs there. FINRA’s explanation of the online trade lifecycle sets out these separate stages.
The distinction can be summarized without treating every service as an exchange:
- The brokerage interface accepts the customer’s instruction.
- The broker’s systems review the order and arrange its execution.
- The execution venue or counterparty completes the trade when its terms can be met.
- Clearing and settlement arrangements handle the resulting obligations and transfers.
A trade confirmation and completed settlement are therefore not the same event. The longer transition from paper records to account-based holdings belongs to the history of share certificates and electronic ownership. Electronic execution changed how a bargain was struck; ownership records and settlement required their own systems.
For readers comparing old and new trading methods, this is the most useful correction to the “middlemen disappeared” story. Some manual interactions disappeared. The functions behind those interactions did not simply vanish.
Lower Commissions Changed the Economics, Not the Investment
Online access also made the brokerage service easier to separate into its parts: execution, research, advice, account administration and support. A customer wanting only order execution could assess a different service from someone wanting regular investment guidance.
Discount and full-service brokerage remain different propositions, even when both offer an app. Some firms offer zero commissions on certain products, while other charges and service differences still matter. FINRA’s stock-investing guidance distinguishes these brokerage models and warns against trading excessively simply because entering an order is easy.
A hypothetical example shows why commission reductions mattered. A $20 purchase commission on a $500 share investment consumes 4% of the amount invested before the share price moves. Cutting that commission changes the economics of a small purchase substantially.
But removing the commission does not remove the difference between buying and selling prices, the possibility of a worse execution price or the risk of choosing a poor investment. Cheap access and sound judgment are separate purchases. Only one comes with an order button.
Electronic Access Was the Foundation, Not the Finish
The clearest way to read this history is as a sequence of separate changes: displaying prices electronically, transmitting orders, automating matching and giving retail customers control over order entry. No single launch accomplished all four.
Electronic execution must also be distinguished from automated decision-making. A person can choose every trade manually and still use an electronic exchange. Software that decides when, where or how much to trade belongs to the related history of algorithmic and high-frequency trading.
The lasting achievement of electronic exchanges and online stockbroking was a shorter route from an investor’s instruction to a market transaction. The lasting caution is just as clear: easier access does not guarantee a fill, a favorable price or a profitable decision.