Direct answer
Last Look in Forex is usually described as a mechanism where a liquidity provider can accept or reject an order shortly after it is received. The main limitation is that this creates uncertainty: the execution you expect from an order request is not guaranteed to become a fill, and the size and timing of outcomes can vary with changing market conditions and provider rules.
Because Last Look is not driven only by your order, its “effect” cannot be treated as a stable, universally measurable feature. In particular, outcomes can depend on market liquidity, volatility, and spreads at the moment the provider evaluates the order. Also, historical relationships between request and fill quality are not reliable predictors when costs and trading conditions change.
Mechanism or definition
“Last Look” is best understood as a post-receipt decision process. In simple terms, your order is sent to a provider, and then the provider may perform checks before finalizing the trade. If the provider rejects, the order may not execute as requested.
It is important to separate stable mechanics from variable conditions:
- Stable mechanics: the idea that the provider gets an opportunity to accept or reject after receiving the order.
- Variable conditions: what checks are applied, how quickly they happen, and how market pricing evolves during that short window.
Since there is no assumption of real-time market data here, any discussion of impact must be framed in terms of general failure modes (how things can go wrong) rather than exact, numeric outcomes.
Evidence or example (failure modes)
A common failure mode is order rejection when market prices move rapidly between order arrival and the provider’s evaluation. Even if you submitted at a reasonable time, a fast jump in the implied tradable price can lead to a rejected request, leaving you without the intended execution.
Another failure mode is asymmetric execution quality. Two traders may submit similar orders under similar quotes, but if their timing relative to micro-movements differs, the set of orders that pass evaluation can differ. This can make observed fill behavior look inconsistent across time.
A third limitation is that observed “good” performance can be conditional. For example, if you measure fill rates and price improvements during a calm period, you may infer a pattern that does not hold during higher volatility. Historical results reflect the specific market regime and cost structure at the time, not a promise for future conditions.
Limitations and risks
Key limitations and risks include:
- Uncertain execution outcomes: Because orders can be rejected, you may not achieve the price or fill that your order request implied.
- Dependence on market state: Rapid changes in volatility and liquidity can increase rejection likelihood or alter effective execution quality.
- Provider-specific behavior: The practical effect depends on the provider’s evaluation rules and timing, which may not be transparent at the same level for every counterparty or venue.
- Cost and estimation mismatch: Even if an order is accepted, costs such as spread and any evaluation-related timing can make realized outcomes differ from simplified assumptions.
- Non-transferability of past observations: Relationships seen in historical data do not guarantee future results, especially when market conditions change.
Verification matters. A self-contained way to test the concept’s effect is to compare order request characteristics (time, price context, and outcome) across different market conditions, while keeping assumptions explicit. If you cannot separate market changes from provider decision behavior, you will likely over-attribute causes.
Verification or next question
To independently verify what Last Look means in your context, you can focus on non-promotional, checkable questions:
- What happens to rejected orders in practice (for example, do they get canceled, re-priced, or handled differently)?
- How do execution outcomes differ between stable and volatile market periods?
- Can you separate changes in the market from changes in execution behavior?
A useful next question is how the provider defines and measures its evaluation window and rejection handling, because those details determine when the limitation is most likely to show up and how much uncertainty it introduces.