The Asset Outperformer toolbar carries an exchange selector, sitting between the score range boxes and the country selector. At capture it read All Exchanges, which is its unfiltered state, and the board below it ranked 171 crypto assets without reference to where any of them can be traded.
That selector is the only venue handle the interface offers, and it is worth building process around. It is also, in its default state, a control that fails open, and the difference between those two sentences is most of the work.
A permissive default is a control that fails open
Start with what the selector does not do. It does not remember your desk's policy. It does not announce that it is unset. It does not distinguish between a user who deliberately chose all venues and a user who never touched it. Every session starts permissive, and a screen that starts permissive produces a shortlist containing names your desk cannot trade, presented in exactly the same visual language as the names it can.
Compare that with how the phase selector behaves on the same toolbar, which at capture read Phase B (Uptrend) and was visibly non-default. A filter that is on and says so is a filter you will notice. A filter that reads All is invisible in a screenshot, in a meeting, and in a compliance review six months later, which is why the venue state has to be captured outside the tool rather than inferred from it.
The distance from a ranked row to an order is also short here. Each row ends in an Action cell with a trade button, so the interface is not purely analytical. Permissive default plus a short path to execution is the standard shape of an operational incident, and it is worth naming as such in the risk register rather than treating the screen as read-only.
What the selector has to acquire before it is a control
Three things turn a dropdown into something an auditor will accept, and none of them live inside the interface.
The first is a canonical list held elsewhere. Your approved venue list is a governed document with an owner, a review date and a change history. The dropdown is a way of applying it, not a place to store it. If the only record of which venues are approved is what somebody selected in a screen, the policy does not exist.
The second is state capture at the moment of decision. Whatever the selector was set to when the shortlist was produced has to be recorded alongside the shortlist, because the shortlist alone is not self-describing. Before you build a workflow on this, check two behaviours directly: whether the selector accepts more than one venue at a time, and whether its setting survives a page reload and a switch between the crypto and stocks tabs. The toolbar also carries Templates and Presets controls, so test whether a saved template captures the venue state or only the score and phase settings. Those answers determine whether your process is one click or one manual step per run, and the manual step is the one that gets skipped.
The third is an exception log. Venue policies are broken occasionally and legitimately, usually under time pressure, and a policy with no exception path is a policy people route around silently. Give the exception a form, a named approver and a review, and you will hear about the breaches instead of discovering them.
Running the screen twice to price the policy
The most useful thing you can do with the selector is not to set it and forget it. It is to run the board twice at each rebalance, once unfiltered and once restricted to approved venues, and record the difference.
That delta is the cost of the venue policy expressed in names. Count how many of the top decile survive the restriction and which specific names were removed. Do it for a few quarters and the delta becomes measurable in return terms as well, because you can track what the excluded names did after you declined to hold them.
This matters because venue policy is normally justified on risk grounds alone and reviewed on nothing. Once you can put a number on what the restriction costs in foregone selection, the annual conversation changes from an assertion that the policy is prudent to a comparison between a known cost and a stated risk appetite. That is a conversation an investment committee can actually have.
The counting is also the early warning. If the surviving share of the top decile is drifting down quarter over quarter, either the venue landscape is moving away from you or the ranking is drifting toward assets you structurally cannot reach, and both are worth catching before they show up as a performance question.
When the top-ranked name lists only on an excluded venue
This is the case that decides whether the policy is real, and it always arrives attached to the most attractive name on the board.
Filtering it out does not make it disappear. It makes it invisible on your screen while remaining visible on everybody else's, which means the question arrives later, from someone who ran the unfiltered board, phrased as why you did not own the best performer. If the only answer available is that it was not on the list, the answer sounds like an oversight rather than a decision.
So handle it explicitly. Record excluded names at each rebalance with the reason for exclusion, rather than discarding them. When one of them is the top-ranked asset, escalate it as a policy question rather than an investment question: does the desk want to onboard this venue, or does it accept foregoing this class of name. Both answers are defensible and the escalation is what makes either one defensible later.
What you should not do is take the position through an indirect route because the direct one is blocked. Reaching an excluded venue's asset through a wrapper, a derivative or a third party does not satisfy the policy, it launders it, and the operational and counterparty risk you have taken on is usually larger than the venue risk the policy was written to avoid.
Where venue restriction bites hardest, and what it still will not tell you
The restriction is far more binding on the crypto tab than on the equity tabs, and for a structural reason. A token's liquidity is frequently concentrated on a single venue rather than distributed across fungible ones, so excluding that venue does not raise your cost of trading the asset, it removes the asset. On the crypto board at capture, 100 of 171 ranked assets were shown as outperforming the basket, and the share of those reachable under a conservative venue list is a much smaller number that only you can compute.
Finally, be clear in the documentation about what the selector is not. It is not a borrow or short availability check, it is not a custody eligibility check, it is not a settlement or fiat on-ramp check, and it is not a sanctions screen. Each of those is a separate control against a separate list, and a screen that has been filtered to approved venues has passed exactly one of them. Writing the filter into a procedure as though it covers the rest is the failure that turns a useful convenience into a false assurance.