Key takeaways
- Liquidity aggregation should be controlled by clear routing rules, not only by the tightest visible bid and offer.
- LP groups, symbol rules, order size limits and failover behaviour should match the broker's risk model.
- Dealing and risk teams need routine reviews of rejects, slippage, stale prices and LP availability.
Liquidity aggregation rules are routing controls
Liquidity aggregation is not just combining several LP feeds into one price stream. For a broker, the main question is how orders move through that pool once clients trade.
A bridge or price engine may see several LPs at the same time. If the setup only follows top of book or a simple priority list, the result can be too blunt. A small order, a large order, a volatile symbol and a high risk client group should not always use the same path.
Routing rules let the dealing desk decide which LPs are used, when they are skipped, and what happens when prices, depth or execution quality change. They do not replace risk judgement. They make that judgement consistent.
Build LP groups by purpose
Many brokers start by placing all connected LPs in one pool. That is simple, but it hides useful differences. LPs may perform differently by symbol, session, order size, depth, reject rate or account group.
Common LP group structures include:
- Primary execution group for normal client flow
- Backup group used only when the primary group is unavailable
- Symbol specific group for instruments where one LP performs better
- Size based group for larger orders or thin books
- Account group specific routing for different client segments
- Test group for new LPs before they receive full production flow
The goal is not to create unnecessary complexity. The goal is to avoid sending every order through the same path when the risk profile is different.
Metals may need different depth and spread rules than major FX pairs. Crypto CFDs may need tighter stale price controls than slower instruments. A prop firm server may need stricter feed stability checks and consistent execution rules across challenge servers. The right setup depends on the broker's instruments, LP relationships and risk appetite.
Symbol mapping also matters. If symbols are inconsistent across platforms, LPs and internal books, routing rules can fail and reports can become unclear. A maintained symbol library reduces this operational risk. Brokers that need to clean up instrument naming and mappings should review their symbol mapping workflow before adding more routing complexity.
Control price construction, markups and failover
A routing rule is only as good as the price it uses. The broker needs to define how the aggregated price is built before deciding where orders go.
Key price construction settings include:
- Which LPs are included in each symbol group
- Whether the system uses best bid and offer or weighted depth
- Minimum depth required before a price is shown
- Maximum spread allowed before a quote is blocked or widened
- Stale price thresholds by symbol or asset class
- Whether an LP is removed automatically after feed issues
These settings affect client experience and broker risk. If the book shows a tight spread on very small depth, larger orders may slip or reject. If failover is too slow, clients may see frozen prices. If failover is too aggressive, usable liquidity may be removed too often and spreads may widen.
Markups also need control. A fixed markup may suit some symbols. Others may need spread multipliers, session based settings or account group differences. Markups should not be used to hide poor pricing. If a symbol is often outside acceptable spread limits, that is a pricing or liquidity issue, not only a markup issue.
Teams should document which controls belong in the aggregation layer and which belong in the pricing policy. For setup options, review the price aggregation and stale price controls and the markup and spread limit tools.
Use execution feedback
Routing should not be based only on quoted price. The best visible price is not useful if the LP often rejects, partially fills or returns poor execution during active markets.
Useful dealing desk checks include:
- Rejects by LP, symbol and order size
- Positive and negative slippage by LP
- Fill ratio during normal and volatile sessions
- Average response behaviour by symbol group
- Frequency of disabled or stale LP streams
- Difference between quoted depth and actual fill outcome
These checks help the broker decide whether an LP should stay in the primary group, move to backup, or be limited to certain symbols. They also help risk managers see when flow is becoming harder to hedge.
This matters for brokers running a mixed A-book and house book model. If the routing layer fails, exposure may stay internal when the team expected coverage, or it may be sent out when the house book was meant to retain it. Coverage logic, routing rules and exposure reports should be reviewed together.
Prop firm operators face a related issue. They may not hedge external flow in the same way as a retail broker, but they still need consistent pricing, execution rules and monitoring across servers. If one server behaves differently from another, disputes and support load can increase.
Set an operating routine
Liquidity aggregation rules should not be set once and forgotten. LP behaviour changes, client flow changes, symbol lists change and platform load changes. A routine keeps the setup usable.
A practical weekly or periodic review can include:
- Check LP uptime, disabled sessions and stale price alerts
- Review top symbols by volume and reject rate
- Compare execution quality across LP groups
- Check whether backup LPs were used and why
- Review large order handling and partial fills
- Confirm that markups and spread limits still match policy
- Check whether any symbols need separate routing rules
- Review alerts that were ignored, repeated or misconfigured
Incident reviews are also important. If a symbol was disabled, spreads widened sharply, or clients received unexpected rejects, the team should identify whether the cause was pricing, routing, connectivity, LP status or internal configuration.
Monitoring should reduce guesswork. Alerts should tell the desk what changed, where it changed and whether action is needed. Too many low value alerts create alert fatigue. Too few alerts leave the desk blind. Brokers can align dealing and technical teams by using shared dashboards for LP status, feed health and routing events. See LP and feed health monitoring for controls that support this routine.
The final step is ownership. Someone should own routing policy, someone should own technical configuration, and someone should review execution outcomes. In a small brokerage this may be the same person, but the responsibilities should still be clear.
XiroLink can support this model with LP aggregation, markups, Centroid compatible FIX, market data controls and monitoring in one bridge environment. The tool matters, but the main value comes from clear rules, clean symbols and a review process the desk actually follows.
Frequently asked questions
What is the difference between liquidity aggregation and LP routing?
Liquidity aggregation builds the available price and depth from several LPs. LP routing decides where an order should be sent, using rules such as symbol, size, account group, LP status and execution quality.
Should a broker always route to the tightest price?
Not always. The tightest price may not be the most reliable if it has frequent rejects, poor depth or unstable execution. Brokers should balance price, fill quality, latency, reject behaviour and risk controls.
How often should LP routing rules be reviewed?
They should be checked whenever LP behaviour changes, after major symbol changes, after platform incidents and as part of a regular dealing desk review. The exact schedule depends on the broker's flow, LP mix and operating model.